artificial intelligence and machine learning - canadian … · 2020. 11. 2. · hospital, in...

INSIDE:

Enterprise messagingClinicians at Michael GarronHospital, in Toronto, have beenusing a real-time messagingsystem called Hypercare tocontact each other quickly.Page 4

Quebec leads in PET/CTOver half of the PET/CT examsconducted in Canada are done injust one province -- Quebec. Toits credit, the province hassupported the installation of thedevices in rural settings alongwith urban centres.4

Machine learningSt. Michael’s Hospital, in Toronto,is building an early warningsystem that can predictdeterioration in patients. Insteadof the usual five to 10 dataelements, this one monitorsabout 100, making use of AI todo it effectively. Page 18

Pub

lication

s Mail A

greem

ent #

40018238

ARTIFICIAL INTELLIGENCE AND MACHINE LEARNING – SEE PAGE 18

CANADA’S MAGAZINE FOR MANAGERS AND USERS OF INFORMATION SYSTEMS IN HEALTHCARE VOL. 25, NO. 8 NOVEMBER/DECEMBER 2020

T W E N T Y- F I V E Y E A R S

DIAGNOSTICIMAGING

PAGE 14

NYGH deploys clinical decision support for DIBY JERRY ZEIDENBERG

TORONTO – To improve the orderingof diagnostic imaging exams, NorthYork General Hospital has imple-

mented one of Canada’s first “clinical deci-sion support” systems that quickly shows re-ferring physicians what the best options arefor imaging their patients.

“The CDS tool is a continuation of whatwe’ve already been doing with ChoosingWisely. We’re leaders there too,” said Dr.Ryan Margau, chief and medical director,Medical Imaging, at North York General.

Choosing Wisely is an international pro-gram that encourages physicians to check onthe appropriateness of the tests and proce-dures they order. The new system, however,gives doctors a tool that’s integrated into theordering module of their electronic recordsystem, so they’ve got a quick guide at thetouch of a few buttons.

“It’s embedded into the EMR, and it helpsfacilitate decisions using evidence-basedguidelines,” said Dr. Margau.

The Clinical Decision Support system in-stalled by NYGH, called iRefer CDS, is from

Toronto-based MedCurrent, a companylaunched by radiologist Dr. Stephen Herman.

Dr. Herman, a specialist in thoracic imag-ing at the University Health Network, cre-ated MedCurrent in 2013. As it happens, un-til now all of MedCurrent’s sales have beenmade outside of Canada – the installation atNorth York General is the company’s first inthis country.

“We look forward to bringing the bene-fits we have seen in other regions to

NYGH, including reducing patient expo-sure to radiation, optimizing clinical work-flow and achieving operational cost sav-ings,” said Dr. Herman.

The MedCurrent CDS platform embedsthe latest version of iRefer, a set of interna-tionally recognized referral guidelines fromthe Royal College of Radiologists in the UK,into a cloud-based offering that leveragesMicrosoft Azure. The platform also provides

a business intelligence module and contentauthoring tool that allows organizations toembed various best practice guidelines, suchas Choosing Wisely or iRefer, into their ex-isting electronic workflows.

So far, the MedCurrent system has beenapplied to the use of MRI exams at thehospital.

Dr. Margau explained that MRI is a so-phisticated modality with many nuances. It’snot just a matter of whether or not a patientshould receive an MR exam, but what kind.

“There are issues of whether the examshould be contrast-enhanced, of the se-quences that are used, and other factors,”he noted.

Because the system has been integratedinto the Cerner electronic health record so-lution used at the hospital, it can automati-cally pull data from the EMR to determinewhether the patient has conditions thatshould be noted. It will then provide a cau-tion to the referring physician.

“It will customize the exam recommen-dation by accounting for the patient’s age,gender, kidney function, allergies, and other

Dr. Ryan Margau, chief and medical director, Medical Imaging at North York General Hospital, has been one of the team leaders drivingthe implementation of a clinical decision support system for DI at the hospital. It’s among the first hospitals in Canada to make use ofCDS in radiology, and the solution is already helping referring physicians make the best choice of exams for their patients. SEE STORY BELOW.

It’s embedded in the EMR and it helps facilitate decisions usingevidence-based guidelines.

PHO

TO:

CO

URT

ESY

OF

NO

RTH

YO

RK G

ENER

AL

HO

SPIT

AL

CONTINUED ON PAGE 2

variables,” said Dr. Jeremy Theal, NYGH’schief medical information officer and agastroenterologist. “It takes all of this fromthe EHR.”

It can also show him or her what otherDI exams the patient has had, and in thisway, the doctor can avoid duplication.

Already, after just one month of usage,“providers are changing the type of MRIsthey’re ordering at least 10 percent of thetime,” said Dr. Margau.

As well, the MedCurrent system hasprompted them to cancel their MRIs in 2percent to 3 percent of the cases. “They’redeciding instead to do other kinds of ex-ams, such as ultrasound,” he said.

Another advantage of the system is that ithas turned MRI ordering from a paper-based process at the hospital to an electronicone. Before now, ordering for all modalitieswas done digitally, except for MRI.

“The message that we’re getting from thedoctors is that this has been great, they nolonger have to go looking for a paper form,”said Dr. Theal. “It’s a huge convenience.”

Dr. Theal said he’s heard no complaints

from the ordering physicians about theswitch to the decision support system forMRI. That in itself is an endorsement. “Ifphysicians don’t like something, you hearabout it,” said Dr. Theal. “Especially whenyou’re introducing new steps in the order-ing process.”

So far at the hospital, since the iReferCDS solution has been implemented, themajority of MRI exams have been orderedby general internists, neurologists, GI spe-cialists and ER physicians.

Dr. Theal noted that iRefer CDS pro-vides analytics, and can show trends andpatterns for modalities, patients and physi-cians. If a physician’s ordering patternseems to be an outlier, “we can have a con-versation to see how to optimize their or-dering practices,” he said.

It’s not meant to be a punitive issue,however, but an educational one.

In the future, the hospital plans to ex-tend the use of the MedCurrent system toother modalities, such as CT and ultra-sound. In this way, patients will receive theoptimal test, with customized procedures.

NYGH also intends to extend the CDSsystem to outside physicians – such as fam-

ily doctors – who direct their patients to theimaging department at North York General.

At the moment, the roadblock is thatthese physicians are using EMRs otherthan Cerner, such as OSCAR, Accuro orPractice Solutions. But Dr. Margau said it’s

just a matter of building the integrationfrom one system into the other.

Since clinical decision support in DI hasbenefits for the hospital and patients, onemay wonder why more hospitals inCanada haven’t adopted the systems, espe-cially since they’re widely used in theUnited States and the United Kingdom.

Dr. Theal noted that CDS requires ahigher level of computerized infrastructurethan is found in most Canadian hospitals.

“In Canada, we’re still below averagewhen it comes to HIMSS EMRAM,” hesaid, referring to the scale that’s used tomeasure the level of sophistication that ahospital has, in terms of computerizedequipment and software applications. Thescale runs from 0 to 7, with 7 indicating avery high-performing facility.

“Unless you have electronic ordering,you can’t take advantage of this,” he said,noting that hospitals with computerizedprovider order entry (CPOE) are at theEMRAM 4 level or higher. Canada has arelatively small proportion of hospitals atthis stage of development.

“And even those who have e-orderingare still at the early stages of their imple-mentation,” he said. “They may be workingon other priorities.”

For its part, North York General is ahighly computerized hospital, and hasachieved Level 6 on the EMRAM scale. Itis a Canadian leader when it comes toadopting new solutions to enhance pa-tient safety and medical outcomes, andscores highly in international surveys oftop performing hospitals.

A recent Newsweek study lists NYGH asthe second highest-performing hospital inCanada, just behind the University HealthNetwork. Adopting CDS is another step onthe ladder of continuous improvement.

Art DirectorWalter [email protected]

Art AssistantJoanne [email protected]

Publisher & EditorJerry [email protected]

Office ManagerNeil [email protected]

Address all correspondence to Canadian Healthcare Technology, 1118 Centre Street, Suite 204, Thornhill ON L4J 7R9 Canada. Telephone: (905) 709-2330.Fax: (905) 709-2258. Internet: www.canhealth.com. E-mail: [email protected]. Canadian Healthcare Technology will publish eight issues in 2020. Featureschedule and advertising kits available upon request. Canadian Healthcare Technology is sent free of charge to physicians and managers in hospitals, clinicsand nursing homes. All others: $67.80 per year ($60 + $7.80 HST). Registration number 899059430 RT. ©2020 by Canadian Healthcare Technology. Thecontent of Canadian Healthcare Technology is subject to copyright. Reproduction in whole or in part without prior written permission is strictly prohibited.

Send all requests for permission to Jerry Zeidenberg, Publisher. Publications Mail Agreement No. 40018238. Returnundeliverable Canadian addresses to Canadian Healthcare Technology, 1118 Centre Street, Suite 204, Thornhill ON L4J 7R9.E-mail: [email protected]. ISSN 1486-7133.

CANADA’S MAGAZINE FOR MANAGERS AND USERS OF INFORMATION TECHNOLOGY IN HEALTHCARE

Volume 25, Number 8 Nov/Dec 2020

Contributing EditorsDr. Sunny MalhotraTwitter: @drsunnymalhotra

Dianne [email protected]

Dianne [email protected]

Dave [email protected]

CONTINUED FROM PAGE 1

NYGH among the first in Canada to roll-out decision support for DI

On Nov. 17 during Digital Health Week, join us for a two-hour program that willaddress important health care topics including virtual care and e-mental health.

Learn more about improving health outcomes for Canadians and creating a more sustainable health system for future generations — register for free.

Don’t miss the final Infoway Partnership: Fall Series virtual event!

infoway-inforoute.ca/partnership-fall-series | @Infoway

Dr. Jeremy Theal: The CDS system can pull information from the EMR and alert the referring physician.

2 C A N A D I A N H E A L T H C A R E T E C H N O L O G Y N O V E M B E R / D E C E M B E R 2 0 2 0 h t t p : / / w w w . c a n h e a l t h . c o m

BY JERRY ZEIDENBERG

TORONTO – Michael Garron Hos-pital – formerly the Toronto EastGeneral – has deployed Hyper-care, an enterprise-wide messag-

ing system that allows clinicians to com-municate with each other through an in-stant messaging app installed on theirphones. Hypercare enables users to se-curely share clinical information and man-age medical on-call schedules.

The Hypercare system runs on a cloud-based network that is independent of thehospital’s existing communication net-works. In this way, it acts as a messagingback-up, in case of outages.

This scenario actually occurred aboutyear ago, in September 2019. The hospitalexperienced a ransomware virus that dis-rupted its email service and paging systems.The Hypercare system, however, which hadbeen deployed in a pilot to a limited num-ber of users, still worked and enabled theenrolled clinicians to communicate.

“Hypercare really showed its valuethrough the code grey time,” said Dr.Patrick Darragh, chief medical informa-tion officer at Michael Garron Hospital.The pilot project of 20 users quickly scaledup to 300 users by February 2020. Andwith the impetus of COVID-19 in March2020, nearly all of the hospital’s clinicianshave joined – about 700 users in total.

“We’ve had such good feedback about[Hypercare] that it’s become an essentialway for clinicians to communicate with eachother in the hospital,” said Dr. Darragh.

It’s not just the quick communicationthat’s appealing to clinicians. The hospitaland Hypercare have together created a direc-tory function that has transformed manag-ing medical on-call schedules from a paper-based task to an automated, electronic one.

Dr. Darragh noted that creating the on-

call schedules used to be done by collatingthe various departmental schedules into apaper-based master list – a painstaking andtime-consuming task. Instead, Hypercareallows departments to create their on-callschedule electronically.

On-call schedules are consolidated au-tomatically and posted immediately to theHypercare messaging app on users’phones. Clinicians can instantly find andcontact the on-call specialists they need.

“You used to have to look up phonenumbers or contact the call centre, find theright person, and wait for a call-back,” saidDr. Darragh. “Now, I just open my phoneand it’s all right there.”

“It’s a better use of hospital resources,and it leads to better communications,”he added.

Albert Tai, CEO and co-founder of Hy-percare, said the application easily supportseven last-minute changes – something that

was difficult to do in the previous way, us-ing paper lists and requiring manual prepa-ration of a master list the day before.

Moreover, clinicians can trade shifts on-line, and the changes show up in the sys-tem in real-time.

“The scheduling piece adds a lot ofvalue to the system,” said Tai.

Dr. Darragh noted that of late, Hypercarehas also been used as an effective COVID-19alerting system. When patients were testedfor the coronavirus, adding the final resultof the swab to the electronic record was in-efficient. Dr. Darragh said there could behundreds of these patients with pending re-sults in the system at any time.

Now, using Hypercare, “when the swabgoes positive, [Hypercare] sends an alert tothe most responsible clinicians.” Then, ap-propriate next steps can be taken by the in-fectious disease and infection controlteams.

As well, a Hypercare alert goes out tothe hospital’s infection control team whenan inpatient COVID-19 swab result is neg-ative. By reducing the time that patientsare in isolation, the hospital can conservepersonal protective equipment.

He noted that this alerting system has thepotential to be extended to other areas, suchas positive blood cultures for a number ofconditions, and for deteriorating patients.

Many hospitals have early warning al-gorithms, but it’s sometimes a problem toget the word out to the right personnel.Hypercare can be used to trigger alerts andsend messages to the smartphones of theappropriate caregivers.

Tai noted that Hypercare can be pro-grammed to escalate – if one person does-n’t respond in a given period of time, themessage will be sent to the next person inline, ensuring that nothing falls throughthe cracks.

Messages can go not just to individuals,but to “roles” – such as the attendingnurse, resident or physician. “The individ-ual may change from day to day,” said Dr.Darragh. “Hypercare makes sure the alertgoes to the right person who needs toknow the information.”

Alerts can also be sent to entire groupsor teams.

As well, being adopted throughoutMichael Garron Hospital, the Hypercare so-lution is now being extended to clinicians inthe community. Family physicians andother clinicians who work with the hospitalare using the application – making it one ofthe first cross-organizational communica-tion systems to be used in the province’snew Ontario Health Teams (OHTs). Hyper-care is already being used by East TorontoHealth Partners, an OHT in East Toronto.

“Since April 1, we’ve opened it up to allof our groups,” said Dr. Meera Shah, a

N E W S A N D T R E N D S

BY SHELAGH MALONEY

Canadians and clinicians havemet the unprecedented chal-lenge of the COVID-19 pan-demic head-on by embrac-ing change in the way care is

delivered – from in-person to virtual.What will it take to maintain this mo-mentum and, more importantly, is thiswhat Canadians want for the long term?

The answer to the second part of thatquestion is yes. The appetite for “virtualfirst” when it comes to healthcare visits isbuilding, among those who have tried itand those who would like to experienceit. The answer to the first part of thequestion is more complicated, and it’stied to how Canadians feel about the useof technology in healthcare more broadly.

Late last year, long before we knewabout the pandemic, Canada Health In-foway undertook a broad initiative to en-vision our health system five to 10 yearsinto the future. We would examine the useof emerging technologies for healthcare,

such as artificial intelligence (AI), preci-sion medicine and robotics, for example.And because our health system exists toserve Canadians, we asked them how theythought these technologies would impacttheir healthcare experiences.

We called this part of our work AHealthy Dialogue, and we believe it’s oneof the largest public consultations aboutdigital health ever conducted in Canada.

We engaged with more than 58,000Canadians through: two national surveys(pre-pandemic and a few months afteronset), online focus groups, in-personfocus groups with Indigenous peoples,interviews with people in vulnerable andunderserved communities (e.g., new im-migrants, members of the LGBTQ com-munity), and an online forum.

What became evident through theconsultation was that, while Canadiansbelieve technology can help improvethe health system, they are keen to haveit resolve the issues they are currentlyfacing before addressing other moreabstract or futuristic concepts.

For example, reducing wait times andimproving equitable access to care are apriority over using artificial intelligenceto predict health outcomes. They knowthat technology can improve their healthexperience by providing faster and more

convenient accessto physicians andnurses. In fact, 92 per centof Canadians wanttechnology tomake healthcaremore convenient.They have seenhow technologyhas transformedthe rest of theirlives and are ready

for the same experience when it comesto their healthcare.

Another thing that came throughclearly in the consultation is that Cana-dians are concerned about health equity,so it’s important for us to ensure that allCanadians have access to health services.

Twenty-six percent report that they havepoor internet connectivity, and 40 per-cent said they aren’t comfortable withapps and technology in general.

These findings confirm the need fordigital health literacy in Canada as well asthe need to bridge the digital divide, im-proving access to internet and broadband.

In an Infoway survey conducted a fewmonths into the pandemic, 84 percent ofrespondents told us that, given the op-portunity, they would use more technol-ogy tools to manage their health. Eighty-six percent said the pandemic hadshown them that virtual care tools canbe important alternatives to seeing doc-tors in-person, and 72 percent saidthey’d be willing to use virtual care, evenif they hadn’t tried it yet. That’s solid ev-idence that Canadians are ready to em-brace a much greater use of technologyin healthcare. Bring it on!

Shelagh Maloney is Canada HealthInfoway’s Executive Vice President,Engagement and Marketing.

Canadians are ready to embrace greater use of technology in healthcare

New system eases clinical communication at Michael Garron Hospital

Dr. Patrick Darragh, CMIO: Hypercare has also been used as an effective COVID-19 alerting system.

Shelagh Maloney

CONTINUED ON PAGE 22

h t t p : / / w w w . c a n h e a l t h . c o m4 C A N A D I A N H E A L T H C A R E T E C H N O L O G Y N O V E M B E R / D E C E M B E R 2 0 2 0

Our vision at Orion Health is to revolutionize global healthcare so that every individual receives the perfect care for them.

Find out more at orionhealth.com

Right place

Right patient

Right care

BY NORM TOLLINSKY

The COVID-19 pandemic hasforced public health agencies andgovernments across Canada tomake difficult decisions about so-

cial gatherings, mask use, testing, electivesurgeries and lockdowns. Most if not all ofthem have some kind of analytics strategyto guide them, says Greg Horne, SAS Insti-tute’s Global Principal for Healthcare. “It’sjust a question of how sophisticated thatstrategy is and how it’s being used to drivechange and patient outcomes.”

Alberta Health, one of the most adeptpractitioners of health analytics, hit theground running when the news of thevirus began circulating around the world.

“Once we heard about the first cases inWuhan, we started thinking about whatwould happen if the virus appeared inCanada,” recalled Larry Svenson, provin-cial health analytics officer and executivedirector for analytics and performance re-porting for Alberta Health. “We startedputting things into play in late January,early February because anytime we seesomething of international concern, weramp up very quickly.”

The essentials, including the infrastruc-ture, the partnerships and the data flows,were all in place. All that was required wasto adapt them to accommodate the specificinformation needs related to COVID-19.

Alberta Health relies on SAS Institute’sViya platform and licenses a variety ofother SAS visualization and statistical soft-ware, including SAS Grid Computing. Italso uses R Studio and a variety of othersoftware products.

Modeling based on the early experienceof Wuhan and Italy, said Svenson “painted apicture that was quite frightening if we hadnot acted at all,” prompting the province toimpose the necessary public health mea-sures to flatten the curve and prevent ad-verse impacts on the health-care system.

“As we started having cases in Canada,we began using more local data, but inthe absence of local data, our initial mod-

els were based on what we were seeing in-ternationally.”

The objective was to “understand atwhat point we would see a trigger thatcould cascade into overburdening thehealth-care system and require us to makeadjustments such as slowing down electivesurgeries and ramping up testing.”

Models also tried to forecast whatwould happen when students returned toschool and at what point lab testing wouldexceed the province’s daily testing capacity.

“The models are mathematically quiteintense with sophisticated differential

equations,” said Svenson. “We’re usingtools like AnyLogic and Mathematica tocreate these models and run them on amainframe environment because they’reso resource intensive. We’re quite lucky be-cause we have a person with a PhD in Ap-plied Mathematics whose doctorate wasfocused on modeling infectious diseaseoutbreaks, and we have a strong connec-tion with the math department at the Uni-versity of Alberta both for COVID and forour work on the opioid epidemic.”

Data is collected from diverse sources,including the province’s CommunicableDisease and Outbreak Management sys-tem, Communicable Disease ReportingSystem, Alberta Precision Laboratories, theNational Ambulatory Care Reporting Sys-

tem, the Discharge Abstract Database,physician claims data, and Alberta’s Phar-maceutical Information Network, alongwith feeds from Alberta Education aboutpotential issues in the schools.

“The overriding objective, said Svenson,“is to make sure that we have high-qualityevidence to support the public health re-sponse and understand how the pandemicis affecting Albertans. We report on every-thing from how many new cases we haveon a daily basis, how many people end upin hospital, the number of deaths, howmany outbreaks we’re dealing with and

whether they’re in long term care facilities,schools, workplaces or other locations.

“The chief medical officer of health, Dr.Deena Hinshaw, relies on us to provide herwith everything she needs to know. Wealso provide evidence to the Minister ofHealth and the Premier so they are wellbriefed and can make decisions based onthe best possible evidence.”

Data presented in charts, graphs and in-teractive maps is also uploaded to Alberta’sCOVID-19 surveillance website every week-day, allowing everyone in the province to seethe number of active and total cases by age,gender and geography, the total number ofdeaths, the number of people in hospital, thenumber of patients in intensive care, the av-erage age at death, how many comorbidities

they had, the number of tests conducted,and comparisons with other jurisdictionsacross the country and around the world.

The daily uploading process is almostentirely automated using R scripts.

Svenson credits the “unbelievable tal-ent” of his team and the quality of the re-lationships with organizations supplyingdata for the reputation Alberta Health hasearned in the analytics space.

“It’s the relationships that are far moreimportant than the data,” noted Svenson.“Once you have solid relationships with allof your stakeholders, their willingness toshare the data is a non-issue. It just hap-pens seamlessly.

“We continually work to keep those re-lationships strong, so when we have eventslike COVID-19, we’re not running aroundtrying to establish new relationships ortrying to convince people to do something.We just move straight into implementa-tion and action mode.”

“Alberta has always been a very analyti-cally savvy province,” said SAS Institute’sGreg Horne. “They’ve always had the useof data at the heart of everything they do,so I wasn’t surprised when they were thefirst in this current environment to use an-alytics to inform their public health re-sponse. They’re very forward thinkingabout what they do.

“Globally, we see that hospitals andgovernment departments that are usinganalytics are coping with this crisis in amuch better way than those that are not asproficient at it.”

Of course, the value of analytics inhealth care extends beyond public healthemergencies like COVID-19 and the opi-oid epidemic, notes Horne.

“We see around the world a very consis-tent problem in health care, which is thatthe demand and need for health servicesfar exceeds the ability of any system tofund it. Analytics,” he claims, “can drivevalue into the decisions that need to bemade with the ultimate goal of makinghealth care more accessible, more afford-able and safer.”


BY NEIL ZEIDENBERG

Back in March, soon after theCOVID-19 pandemic wasdeclared, ICES – formerly theInstitute for ClinicalEvaluative Sciences – an

independent non-profit clinical researchorganization – began receiving a datafeed from eHealth Ontario (nowOntario Health – Digital Services)containing information about completedCOVID-19 tests. The information wasstreamed automatically on a daily basiscreating a rich data repository.

However, by using the feed alone, itwas difficult to determine what type offacility the person tested was living in –something essential for decision-makingand managing the pandemic.

“Although we had some ways to

identify individuals testing positive andnegative for COVID in long-term carehomes, we were unable to identifyindividuals in retirement homes,homeless shelters and correctionalfacilities,” said Mahmoud Azimaee,director, Data Quality and InformationManagement, ICES.

ICES acquired master lists ofaddresses for all of these institutions tobe matched with the individual’saddresses in the daily COVID-19 data.Getting perfect matches was nearlyimpossible considering unstandardizedaddresses with countless typos.

ICES made use of a solution calledSAS DataFlux to standardize COVID testdata, taking steps to match addressesreliably by street, city and postal code.This led to an enriched COVID-19 datawith flags for institutions where the

patient lived. Entity Resolution andfuzzy matching features in DataFluxwere used to achieve this goal.

However, there were still someaddresses missing, and better standardswere needed to match the remaining

addresses. “We utilized DataPackCanada Addresses, an add-on to the SASDataFlux to validate the addresses in thedatabase, improving the linkage rate,”Azimaee explained. “The software alsocorrected many misspelled street namesin the database.”

From there they used the positive andnegative cases by institution name. Butwith data streaming into ICES every dayand multiple COVID tests creating mul-tiple records the data grew exponentially.In the end, SAS DataFlux helped ICESmaintain an accurate work list.

To make the most of using SASDataFlux, all team members for thisproject were provided one-week ofsystem training.

More recently, ICES is using SASDataFlux on a new project, even granderin scale. “We’re looking into a databaseof everyone ever issued a health card inOntario – alive or dead,” said Azimaee.“It’s in excess of over 20 millionrecords.” Using the addresses registeredto these individuals they aim todetermine how many of them are family,or live/lived under the same roof.

ICES refines list of COVID-19 cases in Ontario facilities using analytics

Analytical tools and skilled people are key parts of Alberta’s strategy

Greg Horne, Global Principal, Healthcare, at SAS Larry Svenson, Executive Director, Analytics, AH

With analytics, ICES was able to determine the types of facilities in whichpatients were living.


BY DAVE WEBB

Two home-grown BC tech compa-nies are teaming up to providevirtual healthcare solutions toFirst Nations communities. Wel-

Tel Health and Mustimuhw InformationSolutions Inc. (MSI) announced in lateSeptember that they will partner to pro-vide an SMS-based solution to communityhealthcare providers to improve follow-upand communication with patients.

For patients, it’s as easy as replying to atext message.

“Solutions tend to be over-architected,”said Mark Sommerfeld, chief executive offi-cer of MIS, which provides Indigenoushealth, child and family services to 285 FirstNations communities in seven provinces.And that’s not just in the tech field. Som-merfeld recalled a state-of-the-art garbagedisposal system in one remote communitywhich was powered by diesel fuel – prohib-itively expensive to deliver and consume. Anew and better solution was found.

Urban system designers often don’t takeinto account such limiting factors, Som-merfeld said. According to a report by Fu-ture of Good, a “social impact” websitepublished by philanthropic strategy vet-eran Vinod Rajasekaran, only 10 percent ofFirst Nations households have access to re-liable Internet or broadband services. Thishas cut them out of alternatives to com-municate and get telehealth services dur-ing the ongoing COVID-19 pandemic.

“First Nations are on the wrong side ofthe digital divide,” Sommerfeld said.

“Telehealth is not new to First Nationscommunities,” said Gabrielle Serafini,

CEO of WelTel. But often it’s in the form ofoverly sophisticated, over-designed solu-tions that aren’t practical, given geo-graphic and socio-economic considera-tions. “I’d bet you’d find those video-con-ferencing carts sitting in a closet.”

Enter the ubiquitous cell phone. Whilethose in underserviced areas may not havea smartphone or high-speed Internet,most have a mobile that can receive textmessages. Additionally, SMS services areless infrastructure intensive and less ex-pensive – with no data plan required – andthey’re easier to use.

WelTel, co-founded by Dr. RichardLester, an infectious diseases specialistwith Vancouver Coastal Health, began its

work in African nations, includingRwanda and Uganda, where the company’ssolution was used to monitor adherence tomedication for HIV patients.

“Almost all healthcare is voluntary: pa-tients choose when to engage in care, whento take their medicine (if they choose totake it), and whether to return for follow-up visits,” Dr. Lester wrote in a 2013 paper.“In human immunodeficiency virus (HIV)infection and other chronic diseases, thebenefits of medication adherence for thepatient and public health are tremendous.”

The company is currently offeringtracking support and tracing for COVID-19patients in the region. There’s a similarityin the remoteness and marginalization in

Canadian First Nations communities, Ser-afini said.

In Africa, she said, it’s not unusual tohave a cohort of 4,000 people being tendedby two nurses. The SMS-based solution is50 to 100 times faster than following up byphone, she said.

On the patient side, it can be as simpleas a follow-up text message: “How areyou?” There are no app downloads, noportals to sign into. They simply respond.WelTel has found that three to 10 percentof patients require a follow-up each week.

“We (both companies) believe in meet-ing people where they are,” said Serafini.

On the provider side, there’s a lot moresophistication involved. Providers access aWeb portal that serves as a communica-tions hub, said Sommerfeld. It can auto-matically triage messages for follow-up orreadmission; can establish targeted cohortswith customized messaging; and gatherand use rich data, he said. Serafini said thesolution can also be integrated with elec-tronic medical record (EMR) systemsaside from MIS’s. “It’s entirely interopera-ble,” she said, although each EMR systemhas its own integration protocols.

But there is more on the horizon. Withthe rich data being collected, WelTel islooking to analytics and artificial intelli-gence to enhance the solution. The com-pany is working with the University ofBritish Columbia’s Michael Smith Labora-tories, named for B.C.’s first Nobel Laure-ate, on integrating the solution with AItechnologies like machine learning andnatural language processing, together withanalytics technologies, to forecast futureoutbreaks and patterns of epidemiology.


BY DENIS CHAMBERLAND

In their 2019 report, titled Promot-ing an Overdue Digital Transforma-tion in Healthcare, global consul-tants McKinsey maintain that thepromise of automation and its ben-

efits largely remains to be defined inhealthcare. This is partly because the sec-tor has been noticeably late to embracedigital technologies. Still, the opportuni-ties to improve patient care from the ef-fective use of automation appear limitless.

Take artificial intelligence (AI), forexample. In Transforming Healthcarewith AI: The impact on the Workforceand Organizations, released earlier thisyear, McKinsey expect that AI will cometo be applied in three phases: first, toroutine, repetitive and largely adminis-trative tasks (which can take up to 70percent of a practitioner’s time); second,to support the shift from hospital-basedto home-based care (i.e., remote moni-toring); and finally as a clinical deci-sion-support tool, to augment the workof physicians and staff.

All agree that the benefits of AI inhealthcare delivery will be transforma-tive. Some predict that AI will improvethe accuracy and timeliness of diagnoses

and provide better treatment in develop-ing countries and remote regions thatlack radiologists. Others posit that itmay also help to calculate the probabilitythat an individual may contract a condi-tion, therefore significantly enhancingthe chances of healing the individual.

Despite concerns that automationmay threaten human labour, a broadconsensus exists in healthcare thatautomation may actually alleviate thelabour shortage problems which areexpected to plague the sector fordecades. According to McKinsey’sfindings early this year, the general lackof fear of job loss is partly because only35 percent of time spent in healthcare ispotentially automatable.

In that same vein, most analystsbelieve that automation will eliminatesome human effort from the healthcaresystem without eliminating the humansbecause human empathy is central tohealthcare.

In fact, AI that performs its tasks wellcould free radiologists from drudge-work, allowing them more time to inter-act with patients and colleagues. As oneleading team of radiology experts put it,“the only radiologists whose jobs may bethreatened are the ones who refuse to

work with AI.” (T. Davenport and K.Dreyer, AI Will Change Radiology, but ItWon’t Replace Radiologists, HarvardBusiness Review, 2018).

Automation is not only changing thescope of existing jobs, it likely will leadto the creation of new jobs. In healthcarethis is expected to happen at the inter-section of medical and data-science ex-

pertise – for exam-ple, the emergenceof clinical bioin-formaticians. Automation is alsoa vibrant topic ofdiscussion withhealthcare decisionmakers, govern-ments, investorsand innovators, in-dicating that lead-ers are looking at

new ways to adapt their business modelsto the needs of the 21st century.

Because healthcare is under tremen-dous pressure to deliver more and bet-ter care at a lower cost, many expectautomation to form a part of the solu-tion, by helping to reduce operatingcosts, by reducing the need for staffovertime, cutting down on medical

mistakes, and improving efficiency.Without doubt the most important

threat to the success of automation inhealthcare globally is the largely stillnon-existent regulatory framework. Forexample, if a machine misdiagnoses acancer case, who is to blame? The physi-cian, the hospital, the imaging technol-ogy vendor, or the data scientist whocreated the algorithm?

Similarly, cyber threats loom large inhealthcare, a heavily targeted sectorwhich has seen a 63 percent rise in cyberattacks in 2016 alone, according toDerek Marky, in How Automation and AIImprove Healthcare Cybersecurity (2018).Indeed, automation may significantlytransform healthcare in the years tocome, but only if the legal frameworkcatches up and adequately supports thenew technologies.

Denis Chamberland is CEO of MESGroup, which works with hospitals onmanaged services projects, including Man-aged Equipment Services and ManagedICAT Services. MES Group members wereengaged on all of the MES projects con-cluded in Canada. He can be reached [email protected]. Seewww.mesgi.com.

AI will create new kinds of jobs, merging data and clinical work

B.C. companies provide easy-to-use digital solution to First Nations

First Nations communities in British Columbia are benefiting from a host of new health technologies.

Denis Chamberland


Instead of deploying multiple systems to collaborate, clinicians are using a unified solution.

UK doctors combine AGFA HealthCare XEROand Microsoft Teams to share images quickly

SPONSORED CONTENT

Physicians in England have come up witha fast and easy way to share diagnosticimages – and to consult each other onthe images in real-time – by integratingAgfa HealthCare’s XERO viewer withMicrosoft Teams. The combined system

allows a doctor to share the images with other physi-cians at a touch of a button. Clinicians can also chatabout the images, providing advice by voice or text atthe same time.

“Normally, sharing and discussing images wouldtake a significant amount of time – two peoplewould have to log in to a computer simultaneously,share patient details, and debate their findings, andoften these computers take several minutes just toboot up. Image sharing between staff now takes mil-liseconds, and it can be done from anywhere in thehospital,” said James Diss, radiology registrar at ThePrincess Alexandra Hospital NHS Trust, in Harlow,England, a three-hospital network.

The hospital had already been using Agfa Health-Care’s technology, and when the UK governmentrolled out Microsoft Teams nation-wide – in responseto the outbreak of COVID-19 – managers and clini-cians at The Princess Alexandra Hospital decided tocombine them, creating an even more useful solution.

The integrated system is proving to be of immensevalue to clinicians, especially when they have questionsabout diagnostic images and reports. In some cases,the quick feedback from radiologists has reduced theneed for ordering new or additional diagnostic exams.Moreover, different imaging systems can be included,such as vascular imaging – something the group at ThePrincess Alexandra Hospital has already deployed.

Now, other departments are interested in gettinginvolved, too, including cardiology, surgery andophthalmology.

“With this technology, we are providing clinicianswith a much more efficient workflow process, wherethey don’t have to take down details, login to com-puters, request unnecessary tests, or chase people,”said Dr. Diss.

According to Dr. Diss, the COVID-19 crisis hassparked a surge in creativity at hospitals across thecountry. “One of the few positives to come out of thepandemic is the enthusiasm from staff for innovation;we have achieved more in the past couple of monthsthan we have done in the last five years. Staff have seenthe benefits that technology can provide in improvingproductivity and collaboration amongst teams, andmost importantly, clinical care and patient outcomes.”

In collaboration with radiologists and clinicians,Agfa HealthCare has implemented a host of innova-tions to combat the COVID-19 pandemic. Here aresome of them:

COVID-19 Specific Priority Work Lists in Enterprise Imaging: To help the radiology team of a Belgian UniversityHospital to keep track of the COVID-19 cases amongother high importance cases, a dedicated COVID-19worklist has been created. From the Agfa HealthCareRIS, a COVID-19 indication is communicated to En-terprise Imaging, resulting in the creation of aCOVID priority reading task.

To quickly configure the reading priority, withoutthe need for additional manual intervention by users,

an unused order priority was used. In the list area, aspecial color code is given to the COVID-19 cases tohighlight the cases, so they stand out in the task list.Task assignment rules are configured if these specificcases need to be assigned to a specialized group of ra-diologists for reading. Escalation rules can also be em-ployed to ensure quick response on highlighted cases.

Faster Reporting with COVID-19 Drop-down Menus and Text Macros: With suspected COVID-19 cases increasing rapidly, ef-ficiency of radiology reading needs to be maximized.In collaboration with an Enterprise Imaging customer,a series of COVID-19 specific drop-down menus withtext macros have been created, covering ground-glassopacity, vascular thickening and scoring of affected ar-eas in the lungs. Apart from a faster reporting, it willalso result in more consistent reports, which in a longerterm will be beneficial for analytics and statistics.

Balance the load – Image-sharing across hospitalnetworks and regions: In many countries, the COVID-19 outbreak requiresthe collaboration and image sharing across hospi-tals, hospital networks and geographical regions.Some hospitals are centralizing triage of patientsand depending on the case severity, patients are sentfor admission to different hospitals. Others are look-ing to balance the bed capacity and load with hospi-tals inside or outside their network. In the UnitedKingdom we are supporting customers balancingthe COVID-19 load by connecting different sites,using the XERO Exchange technology. They can

view, share DICOM images and reports storedwithin the Enterprise Imaging for Radiology solu-tions. Based on hospital request, patient history canbe shared as well.

With new diseases come new terminology: Since they are not standard in the Radiology lexicon,we received feedback from some radiologists theirspeech recognition software for fast reporting wasstruggling with the recognition of new terms likeCorona or COVID-19. This was slowing down thereporting turnaround, since repeated manual correc-tion of the not-recognized terms was needed.

Customers in Belgium and Luxembourg, usingEnterprise Imaging integrated with the NuanceSpeech Magic recognition software, have quickly re-trained their Nuance Speech Recognition systems andadded those new terms to the speech lexicon. Oncethe new terms added, the speech accuracy for dictatedwords like COVID-19 has improved significantly andsupports fast and accurate reporting in times that pa-tient need fast diagnosis most for timely treatment.

Business intelligence – using data to measure, understand and predict: Customers around the globe have been using Busi-ness intelligence tools to mine their radiology orderand report data. Not only to help monitor the cur-rent pandemic, plan for radiology capacity, but alsoto learn and be better prepared in case any futureoutbreak might occur.

Text search in radiology reports is a powerful toolto find emerging trends. A pandemic usually startssmall, but once gaining traction, can have a steepgrowth pattern. Catching it first signs early can helpstop the spread. Or in this case, finding the root ofwhen and where exactly the disease started toemerge. This is what customers in Italy have done.With text search in radiology reports on term like‘pneumonia’, ‘pneumonitis’, or any other relevantterms, they were able to back-track the start of theoutbreak. At a later stage, once the disease was con-firmed, report text search on terms like ‘COVID-19’can be also be used for further research purposes.

HE

AL

TH

CA

RE

I

NN

OV

AT

IO

N

h t t p : / / w w w . c a n h e a l t h . c o m N O V E M B E R / D E C E M B E R 2 0 2 0 C A N A D I A N H E A L T H C A R E T E C H N O L O G Y 9

BY DIANNE CRAIG

PrecisionOS, developers of a vir-tual reality (VR) training plat-form to enhance surgical readi-ness, are transforming the ap-proach that nurses take to en-

hance their skills with OR training. In a pi-lot project taking place at the BurnabyHospital, near Vancouver, operating roomnurses specialized in orthopaedics are us-ing the VR platform to refresh and advancetheir understanding of surgical procedures.

“We learn primarily by failing,” said Dr.Danny Goel, an orthopaedic surgeon andCEO of Vancouver-based PrecisionOS.“With this understanding, we are challeng-ing the current model by allowing ourusers to make mistakes and learn fromthem in a risk-free environment.”

While Precision’s VR training to datehas primarily been focused on delivering aplatform geared toward surgeons, it re-cently developed a version that caters tothe specific needs of nurses, who ulti-mately work together collaboratively in theoperating room. “This is a novel way totrain all members of the OR team, includ-ing surgeons, nurses, and medical deviceprofessionals,” said Dr. Goel.

During the early months of the COVID-19 pandemic, elective orthopedic surgeryhad been postponed at all B.C. hospitals. Inlate May, PrecisionOS sent two VR headsetsto the Burnaby Hospital that nurses could

use for experiential learning. “PrecisionOScame to us at the right time,” said SandeepRandhawa, clinical nurse educator at Burn-aby Hospital, explaining they had beenlooking for ways to elevate their trainingprocedures for nurses.

Typically, said Dr. Goel, operating roomnurses learn through an apprenticeshipmodel, where they are learning by watch-ing as opposed to learning by doing. “NewOR nurses were monitored by experiencednurses,” said Randhawa, adding that thatwill continue, but VR now further pro-gresses their training.

Randhawa said that in addition to pro-

viding enhanced operating room trainingfor new nurses, the VR experience acts as a“refresher for nurses who haven’t been insurgery for a while, to bring them back upto speed.”

“The feedback has been very positive,”said Randhawa. “They value the ‘safe space’to practice and be comfortable with thatprocedure. It helps them refresh theirmemory.”

Leslie Romero, an operating roomnurse, was among the first of five or-thopaedic nurses there to use the Preci-sionOS VR platform. Romero explained,“it’s giving me a greater understanding and

visualization of the steps for the surgeries.It gives me a chance to look at and to seewhat they are actually doing.”

Founded by Dr. Goel, along with twoexperienced gaming professionals, chiefcreative officer Roberto Oliveira, and chieftechnical officer Colin O’Connor, Preci-sionOS currently has a highly experiencedand well-rounded team of nearly 40 peo-ple, including industry advisors.

“There are experts in surgical educa-tion, human design interface, simulation,haptics, software developers, technologyartists, visual effects and medical deviceexperts,” said Dr. Goel. The company usesFacebook’s Oculus Quest for the headsetand powers its VR platform with the Un-real Engine used by gaming professionals.

The VR learning process, explainsOliveira, starts with selecting a module.“You can select from several procedures.For the shoulder alone, there are nine dif-ferent pathologies. Learners are takenthrough each protocol step by step, as theydrill a guide hole or place a saw. During thiswhole process, there are “audio and visualinstructions guiding you on what to do. Youcan also repeat a step, skip ahead, or skipback,” he said.

“Once you’ve completed a procedure,you are presented with a metrics screenwhich provides specific feedback on howyou did,” added Oliveira. The next timeyou put on the headset and handset, you’ll


BY KIRSTEN LEWIS

One of the most complex chal-lenges in caring for those af-fected by COVID-19 is at-tending to underlying med-

ical conditions that have been shown toinfluence how sick a patient becomes.Having multiple health conditions alsoimpacts timely and accurate diagnosis ofeither COVID-19 or an exacerbation oftheir current conditions due to overlap-ping symptoms.

Heart failure is very common, affectingan estimated 600,000 Canadians. It is bothan underlying risk factor for more severeCOVID-19 illness, and a condition thatmay be missed or overlooked as many ofthe symptoms of COVID-19 are similar.

According to the European Society ofCardiology and Centers for DiseaseControl and Prevention, heart failure pa-tients are at an elevated risk of severe ill-ness from COVID-19, though most pa-tients with the virus experience mild in-fections and fully recover. This meansheart failure patients, if they do get sickwith COVID-19, can become very un-well and require longer recovery times.

This clinical complexity of evaluatingindividuals with difficulty breathing un-derlines the importance of frontlinemedical staff providing accurate pre-screening for COVID-19. Having theright tools allows for COVID-19 to beruled in or out. Appropriate tools also

allow for quick recognition of heart fail-ure symptoms as the care priority.

Great progress has been made sincethe start of the pandemic in aligningknowledge with on-the-ground practicethrough the development of newscreening tools, knowledge networkingand guidelines.

For example, the Canadian Cardio-vascular Society has created a tool to as-sist with remote assessment of patients,providing a process to help differentiatebetween COVID-19 and heart failuresymptoms, ensuring the best diagnosticand treatment approach.

In addition to the risk of diagnosticconfusion between heart failure andCOVID-19, there are other prioritiesthat are supported with screening tools,including keeping patients out of health-care facilities to reduce exposure riskand save resources.

Among the various ways our health-care system responds to the pandemic isan emerging key area: equipping healthprofessionals in hospitals, primary care,or other areas, with the tools to putemerging knowledge around COVID-19into action. These tools help ensure pa-tients get the best available diagnosis,and best possible treatment, from theirfirst point of contact.

Our clinical R&D team has been as-sisting these efforts by authoring usefuldecision-support tools to help provideleading-edge care. Mobilized via point-

of-care technology to be easily accessible,they have been continually updated tomeet the needs of COVID-19 healthcare.These tools help ensure healthcare pro-fessionals can act on the best-availableknowledge when they make decisionsand enable consistent high-quality re-mote screening by phone or through avirtual visit platform.

In a perfect world, clinicians wouldhave ample time outside their key duties

to keep up withthe latest researchdevelopments.Even before thepandemic, thatproved challeng-ing. Now, with ourentire healthcaresystem facing acommon threatand constraintsunlike anythingwe’ve seen before,

it’s become even more difficult. This is compounded by an extremely

high volume of evolving knowledge ontreatment and diagnostics – often withconflicting data. According to CDC data,there are now more than 100,000published COVID-19 research articles.

Decision supports have been provento help clinicians to differentiate be-tween the diagnosis of two similar con-ditions, heart failure and COVID-19,whose overlapping symptoms otherwise

form a diagnostic grey area.To help, we have developed a heart

failure/COVID-19 triage tool that hasillustrated the value of leveragingmedical technology to action the latestclinical research content. The triage toolis part of our extended COVID-19library of order sets and tools that hashad tremendous uptake, with more than7,500 downloads of the triage tool alone.

The order sets and tools support fastdiagnostic testing, early infectioncontrol, and an inventory of bestpractice interventions. This is part ofour ongoing commitment to buildCOVID-19 decision support tools thatmake a difference – adopting the mostup-to-date, evidence-based practices andimproving outcomes for heart failurepatients and others exhibiting possibleCOVID-19 symptoms.

As we enter the second wave ofCOVID-19 infection, these complexmedical challenges are not going away,and ensuring medical professionals areproperly equipped to promptly respondand better diagnose such patients is anongoing priority for us. Putting the best-available decision-support tools in thehands of frontline healthcare staff willhelp the Canadian healthcare systemweather the storm and provide the bestpossible care for everyone.

Kirsten Lewis is Think Research Vice Pres-ident, Research and Development.

Decision support tool helps diagnose COVID, heart failure, or both

Orthopedic surgery nurses accelerate learning using immersive VR

Virtual reality allows surgeons, nurses and other professionals to practice in a risk-free environment.

Kirsten Lewis


h t t p : / / w w w . c a n h e a l t h . c o m1 0 C A N A D I A N H E A L T H C A R E T E C H N O L O G Y N O V E M B E R / D E C E M B E R 2 0 2 0

Replacing a technology that’s been in place for over 50 years provides an

opportunity to enable seamless communications through an intelligent

ecosystem connecting all the people and information needed to deliver

patient care.

With Vocera, determine what to prioritise

next and close the loop faster by

viewing secure messages, phone calls,

and alert and alarm notifications in one

place. Locate people quickly, collaborate

productively and reduce the noise with

an agile workflow platform that delivers

meaningful, actionable information

when, where, and how it’s needed.

For more information,

call 647-490-1637

or visit vocera.com/

pagerreplacementsolutions

2Bleeper.2479.PrintAd.202010.Canada.PRINT.indd 1 10/7/20 10:48 PM

patient car


opportunity to enable seamless communications thr

Replacing a technology that’

e. patient car



Replacing a technology that’



s been in place for over 50 years prReplacing a technology that’


ough an intelligent opportunity to enable seamless communications thr

s been in place for over 50 years pr


ough an intelligent

ovides an s been in place for over 50 years pr

place. Locate people quickly


viewing secur


With V

place. Locate people quickly


e messages, phone calls, viewing secur


ocera, determine what to prioritise With V

, collaboratecklyy,


e messages, phone calls,


ocera, determine what to prioritise

when, wher



oductively and rpr

s needed.e, and how it’when, wher



educe the noise with oductively and r

s needed.



educe the noise with

call 647-490-1637

For mor

call 647-490-1637

e information, For mor

pagerr


eplacementsolutionspagerr


eplacementsolutions

BY NICHOLAS CHEPESIUK

We are at a crossroads in men-tal healthcare. The pandemicis leading us to either turn ablind eye to the growing wave

of mental illness and addiction in Canada oruse the circumstances as an opportunity toadapt and improve mental health services.

In a recent survey done by StatisticsCanada, over half of all participants re-ported that their mental health has wors-ened since the onset of physical distancing.

This survey aligns with research under-taken by CAMH, which shows an increasein the number of Canadians with in-creased alcohol consumption, elevatedanxiety, feelings of loneliness, and who aregrappling with depression.

Soon after the pandemic started, manyhealthcare organizations launched virtualcare for at least some of their services.

Trafalgar Addiction Treatment Centres, aCARF-accredited private provider of addic-tion and mental health treatment that offersresidential and outpatient clinic programs inperson, transitioned all outpatient programsto be fully virtual when the pandemic hit.

“When COVID hit and we had emer-gency orders coming in telling us, ‘you’vegot to lock down,’ we knew that we could-n’t send anybody home without access andsupport,” said Shane Saltzman, CEO ofTrafalgar Addiction Centres.

“We reached out to OnCall Health andsaid, ‘How quickly can you get me acti-vated, signed up, and start onboarding mystaff?’ Within 48 hours we started deliver-ing virtual care.”

Headquartered in Toronto, OnCallHealth provides secure virtual care soft-ware that powers over one million virtualhealthcare appointments each year forhealthcare providers and organizationsthroughout North America.

Although Saltzman admits that his teamhad aspirations to implement virtual care atTrafalgar Addiction Centres beforeCOVID-19, the pandemic acted as a cata-lyst, driving the organization to act fast inthe face of change. Trafalgar launched a vir-tual rehab program to replace its outpatientservices that were typically held in person.

OnCall Health appointed a full-timeaccount manager to ensure Trafalgar’s em-ployees were adequately trained, on-boarded, and able to access knowledgeable

technical support 24/7. Trafalgar was usingthe system in its practice in only two days.

Mental health and addiction serviceslike Trafalgar have seen improved out-comes while using virtual care platformsto connect with patients remotely, stream-line administrative tasks, and even host in-ternal meetings for clinical supervision.

Virtual care allows practices like Trafal-gar the capacity to schedule flexible ap-pointment times that are more convenientthan in-person visits. Providers and health-care organizations can use this versatility toexpand and improve access for their clientswhile increasing revenue for their business.

“This virtual realm makes mentalhealth and addiction treatment muchmore accessible,” said Saltzman. “We cannow provide essential treatment, safelyand securely. We can manage full groupsizes and scale up and down quickly.”

Trafalgar’s four-week Virtual RehabProgram provides patients with 20 hours aweek of individual group and family ther-apy. Online treatment delivery has also en-abled Trafalgar to expand its services na-tionally, with patients from across Canada

participating in the three Virtual RehabProgram groups that it offers each day –with six to 10 people in each group. Beforethe pandemic, Trafalgar was only able tooffer one rehab program.

The virtual care platform allows health-care providers and patients to interact usingone to one and group video conferencing,but it doesn’t stop there. The platform alsoenables secure instant messaging, file shar-ing, screen sharing and the ability to sched-ule and manage phone appointments.

For its part, OnCall Health has seen aspike in the use of subscription text ther-apy services among clients, a feature that

allows providers to use a messaging appli-cation to either complement a video ap-pointment or confidentially engage in pro-fessional counselling. These unique tele-health applications now grant people ac-cess to licensed therapists in the palm oftheir hand.

Virtual care in a mental health contexthas also improved outcomes for providersin unique ways. In one application, a be-havioural therapy provider that assistschildren with autism recognized that videoconferencing allows children to be accom-panied by their parents in the comfort oftheir home. Children get to learn skills inthe same environment in which they needto apply them, making it easier for them toreplicate the behaviours.

Seniors have also reaped the benefits ofvirtual care. Ageing populations are espe-cially vulnerable to the mental health chal-lenges that result from isolation due to so-cial distancing protocols amid COVID-19.Seniors that engaged with virtual care plat-forms for their primary care needs re-ported unexpected mental health benefits.They were also satisfied with the opportu-nity to remain active in the managementof their health.

Healthcare providers are embracing theefficiency of virtual care as it enables themto not only connect with clients virtually butalso efficiently manage operations with fea-tures like automatic scheduling, appoint-ment reminders, consent forms, and securepayment processing. Privacy is also pro-tected with end-to-end encrypted, PHIPAand PIPEDA compliance. Learn how virtualcare is useful to healthcare providers in awide range of industries at https://oncall-health.ca/use-cases/mental-health/

Nicholas Chepesiuk is CEO of OnCallHealth, a provider of virtual healthcare soft-ware.


In July, Montreal-based MEDFARClinical Solutions acquiredPLEXIA Electronic MedicalSystems of Burnaby, BC. Eventhough the two EMR developers

have impressive growth trajectories anda similar vision for healthcare,successfully merging both teams at bothends of the country during the COVID-19 pandemic meant exploringuncharted territories.

Wesley Wong, PLEXIA’s generalmanager, reflects back on the process:“Since mid-March, all of our colleaguesin both British Columbia and Quebecwere working from home. As you canimagine, finalizing a transaction of thissize requires the collaboration of manyindividuals at different levels of bothorganizations. And, with restrictions onair travel, all of this was conducted andcompleted via videoconferencing. I amproud of our teams. Everybody workedreally hard to bring all the pieces togetherand we are pleased with the results.”

COVID-19 forced both organizations

to adapt to an unexplored and unstablebusiness landscape, but it did notdistract PLEXIA or MEDFAR from theirvision to drive excellence and efficiencyin healthcare.

Wong shares his thoughts on thejourney: “I am thoroughly impressed bythe level of respect, understanding andengagement of the entire Montreal teamwhile we worked together to align ourresources, tools, and processes. Fourmonths after the acquisition, both teamsare already reaping the benefits fromworking together.”

The enthusiasm is also shared byMEDFAR’s CEO and co-founder, EliasFarah: “We are confident that our laser-focused EMR strategy improves theexperience of clinicians and theirpatients. With the addition of PLEXIA,we are now the largest and fastestgrowing pure-play EMR provider inCanada. The transaction gives us scale todouble down on our plan.”

The acquisition of PLEXIA acceleratesMEDFAR’s expansion and increases its

footprint in Canada. More than 6,000physicians and 20,000 users working in1,500 clinics delivering 6 million patientvisits per year nationwide have benefitedfrom MEDFAR’s EMR, so far.

Over the past 10 years, MEDFAR hasevolved its MYLE EMR into a CareManagement Solution that follows allaspects of the patient journey, whether

care is delivered virtually or at the clinic.This unique solution fully integrates e-fax, videoconferencing with MYLETelemed, the MYLE Patient Portal,appointment confirmation with textmessages, a statistics and reportingmodule with MYLE Analytics, as well asthe new MYLE Kiosk that allowspatients to self-register at the clinic.

With all these powerful featurestriggered in one click from the MYLEcontrol panel, clinicians and staff areempowered to deliver care withenhanced safety and efficiency.

Interestingly, PLEXIA’s innovationroadmap has been following a similarpath with a comparable vision, but inmirror image for specialists of thefeatures that MEDFAR was developingfor family physicians.

“PLEXIA and MEDFAR are a perfectmatch,” said Farah. “As much as we sharethe same vision and values, our featuresare designed for different groups ofclinicians: MEDFAR with a focus onfamily medicine and PLEXIA with anemphasis on medical specialties. Add tothe equation the fact that PLEXIA grewin Western Canada, while MEDFARexpanded in Eastern Canada and theresult is the combination of two verycomplementary organizations that nowhave the scale to innovate faster andbetter, as well as the critical mass tothrive coast-to-coast.”

EMR vendors join forces to create fast-growing and innovative company

Virtual care eases access to mental health and addiction services


The merging of MEDFAR andPLEXIA creates a company thatoffers a full-featured EMR forboth GPs and specialists.

For critically or chronically ill patients, daily remote monitoring can mean the

pressure, pulse rate, SpO

symptoms change

respond to changes in condition

Book a Cloud DX Connected Health demo today1-888-543-0944 | [email protected] | clouddx.com | @clouddx

Connected Health for Remote Patient Monitoring

W I N N E R

B O L D E P I C I N N O V AT O R A W A R D��

T R U S T E D B Y

“Remote automated

discharge - post surgery - is the way of the future. This technology gives health care providers the ability to detect early signs of complications and optimize medical management,

keep patients out of the hospital and in the process facilitate more elective and urgent surgeries and reduce the spread of COVID-19.”

- Dr. PJ Devereaux, Professor, Director,

Leader of the Anesthesiology, Perioperative Medicine and Surgical Research Group at PHRI, McMaster Health Sciences.

D I A G N O S T I C I M A G I N G

BY NORM TOLLINSKY

Earlier this year, the smallmining town of Val d’Or, 525kilometres north of Mon-treal, welcomed the arrival ofa state-of-the-art PET/CT,bringing advanced imaging

to the 135,000 people in northwesternQuebec’s Abitibi-Temiscamingue region.

The GE Healthcare Discovery MIPET/CT at the 90-bed Hôpital de Val d’Orwill serve patients throughout the sparselypopulated region, including the towns ofRouyn-Noranda and Amos, as well as pa-tients flying in from Indigenous commu-nities in the province’s Far North.

The $7 million project – $3.2 million forthe PET/CT and $3.8 million for an ex-panded suite to accommodate it – wasfunded entirely by Quebec’s Ministère de laSanté et des Service Sociaux, underliningthe province’s commitment to ensuringequal access to advanced medical imagingoutside the province’s major urban centres.

PET/CTs have also recently been de-ployed in two other mid-sized communi-ties across Canada – Kelowna and Sudbury– but Quebec leads all provinces in thenumber of PET/CTs and accounts formore than half of the PET/CT exams per-formed in Canada.

Preliminary data from the CanadianAgency for Drugs and Technologies inHealth’s (CADTH) 2020 Canadian Med-ical Imaging Inventory, scheduled for re-lease in January 2021, reports 67,849PET/CT exams for the 2018-2019 fiscalyear in Quebec and only 23,564 exams inOntario for the fiscal year 2019-2020.

That equates to 1.6 exams per 1,000

population for Ontario’s 20 PET/CTs and8.0 exams per 1,000 population for Que-bec’s 23 devices.

Dr. Francois Lamoureux, a nuclearmedicine specialist at Hôpital de Val d’Orand president of the Canadian Associationof Nuclear Medicine, attributes Quebec’sleadership in PET/CT accessibility to along-running education campaign to edu-cate specialists, general practitioners,healthcare administrators and the generalpublic about the value of PET technology.

Whole body PET scans show abnor-mal metabolic activity at the molecularlevel using an injected radioactive sugarsolution that “lights up” fast-growing le-sions, while CT scans show physicalstructures using X-rays, allowing for pre-cise localization.

PET technology has come a long way inrecent years, said Dr. Lamoureux.

“Ten years ago, when we started talkingabout using more PET technology clini-cally, there were many drawbacks. It wasdifficult to have the tracers and the ones wewere using were very limited. You needed acyclotron that cost $15 million and special-ized staff. Now we have much less expensiveradionuclide generators to produce tracers.”

Using PET/CT imaging, “you canchange the therapeutic approach for pa-tients in 30 to 40 percent of cases,” claimsDr. Lamoureux,

Using a CT or X-ray imaging alone,there’s no way of knowing if a pulmonarylesion or nodule, for example, is benign,requiring in some cases a very invasivebiopsy, he explained.

However, if a PET scan is performedand there is no uptake of the tracer, the pa-tient will be deemed to have a benign le-

sion, and no further investigation is re-quired. “If there is an uptake – because it’sa whole body investigation – you are ableto see not only the primary lesion, but alsothe metastases that could exist elsewhere.”

With the Discovery MI PET/CT at Vald’Or, said Dr. Lamoureux, “we are able todetect lesions as small as four millimetres.To see a lesion using a CT, it has to beabout one centimetre.”

PET/CTs can also be used to assess howa lesion is responding to chemotherapy.

“Suppose you are prescribed six roundsof chemo. It’s very expensive and very hardon the patient, but if you do another PETstudy after the first treatment and findthere is no more lesion, you can stop thetreatment,” he said.

That’s what happened with MontrealCanadiens captain Saku Koivu, who was di-agnosed with non-Hodgkin’s lymphoma inSeptember 2001. A PET scan performed inSherbrooke prior to the conclusion of theprescribed course of chemotherapy pro-nounced him cured, allowing him to returnto the ice the following April.

Having to endure more chemo thannecessary, said Dr. Lamoureux, “is almostunethical.”

“The technology has evolved so rapidlythat we can now use it to diagnose cardiacand neurodegenerative diseases likeAlzheimer’s, Parkinson’s and vascular de-mentia (in addition to cancer),” he noted.

“In Canada, neurodegenerative diseasescost us $33 billion per year, so you can seethe importance of making the right diagno-sis. With PET imaging, you can distinguishbetween different neurodegenerative dis-eases and offer the appropriate treatment.”

According to Dr. Lamoureux, the Mon-treal Heart Institute performs 60 PET/CTcardiac studies per week, but because ofthe shortage of PET imaging availability inthe other nine provinces, studies aremostly limited to oncology.

The CADTH’s 2020 Canadian MedicalImaging Inventory reveals up to 11.8 per-cent of the 128,121 PET studies performedin Canada during the most recent fiscalyear (or for 2018-2019 in Quebec) were forcardiology cases and only up to 5.7 percentwere for neurological investigations.

“The other nine provinces are 10 yearsbehind the rest of the world in PET use,”said Dr. Lamoureux. “I think the generalpublic should be aware of that because inmany situations, patients aren’t being of-fered the appropriate investigation.

“The situation in the rest of Canada iscompletely unacceptable. The technologyis there. The specialists are there, and it’sno longer an expensive tool. Every nuclearmedicine centre should have a SPECT/CTand a PET/CT. It should be the same for allof Canada.”

Dr. Lamoureux also disagrees with thepolicy in Ontario and other provinces ofrelying on local fundraising to cover thecost of PET/CTs.

“I’m completely against that,” he com-plained. “Doctors are like soldiers. We’re

Dr. Francois Lamoureux with the GE Healthcare Discovery MI PT/CT at l’Hôpital de Val d’Or. Formerly, chief ofthe nuclear medicine department at l’Hôpital Notre-Dame in Montreal, Dr. Lamoureux is the president of theCanadian Association of Nuclear Medicine and serves as a nuclear medicine specialist at l’Hôpital de Val d’Or.

The Legal Guide to Consent, Capacity & Substitute Decision Making

December 7, 2020 - Online, Live & InteractiveIncludes 120-day access to the program archive

A patient shows signs of diminished capacity. Do you know all your legal obligations to protect their rights?

Get the knowledge you need to navigate the complex issues you grapple with, including: issues of informed consent, COVID-19’s impact on capacity assessments, protecting patient privacy and more.

Register today at: osgoodepd.ca/capacity

PHO

TO C

OU

RTES

Y O

F D

OM

INIC

CH

AM

BERL

AN

D –

JO

URN

AL

LE C

ITO

YEN

/VA

L-D

’OR

Quebec continues leadership in PET/CT, deploys scanner in Val d’Or



cerner.ca @CernerCanada

Embracing proactive health for better outcomesPromoting prolonged wellbeing is something that benefits us all –citizens, communities, and health organizations alike.

Wellbeing relies on a holistic view of individuals and wider populations, and Cerner can help you achieve this. Our personalized, comprehensive suite of products and services allows you to join the dots of people’s health to form a complete circle of their lives and those of their community.

With Cerner’s HealtheIntent®, you get to know your population, engage its members and manage what happens to ensure the delivery of better outcomes.

HAMILTON, ONT. – McMaster Uni-versity’s McArthur Lab, whichconducts data-intensive researchon the genetic makeup of super-bugs – including the SARS-CoV 2virus that causes the COVID-19

disease – has installed high-powered flash storagefrom Pure Storage. The Pure FlashBlade system pro-vides the lab with 72 terabytes of digital storage anddelivers data to the desktop much faster than con-ventional systems.

“There’s no point in playing with traditional stor-age because it’s just not fast enough,” said Andrew G.McArthur, Ph.D., associate professor at McMasterUniversity, in Hamilton, Ont. “With Pure Storage wecan stay ahead of the curve as we fight global threatsto human health.”

The McArthur Lab is home to 30 investigators, in-cluding professors, students and postdoctoral fellowswith expertise in infectious diseases, human genetics,microbiology and drug discovery. They’re also col-laborating with numerous colleagues and partnersoutside the lab – across Canada, in the United Statesand around the world.

The lab has been deeply involved in decipheringthe transmission of COVID-19 using proprietarytechnology that it developed itself along with off-the-shelf equipment. Because the COVID virusmutates slowly, it’s possible to take meaningful “ge-netic fingerprints” as it works its way through thepopulation.

The McArthur lab can detect chains of transmis-sion by sequencing the genomes of the viruses foundin people who have tested positive. If the sequencesare similar, the patients were probably infected by

each other. If the sequences are different, the infec-tion likely came from elsewhere.

In this way, researchers and public health officialscan establish chains of transmission. Doing this kind ofdetective work – for COVID-19 and many other dis-eases – requires researchers to work quickly while us-ing enormous data sets. They’re sequencing the virus

in thousands of patients, and each virus consists of28,000 base pairs that must be analyzed. The im-mense volume of data that’s generated by geneticresearch prompted McMaster’s McArthur Labto opt for the Pure FlashBlade storage.

“FlashBlade speeds drug discovery andpathogen biology by analyzing select data sets24 times faster than before,” said McArthur.“What once took one to two days to conduct,now can be done in two to three hours.”

The improvement in speed and resultsbenefits not only the researchers, but health-care providers, policy makers and patients, as well.

“Data is critical to the daily insightsin the lab,” said McArthur. “How-

ever, the vast and growing amount of information re-lies on near-instant processing times. By replacingour legacy storage infrastructure with Pure, we havethe ability to gain insight into the virus through next-generation sequencing.”

As gene sequencing technology continues to ad-vance, rapid scalability and performance will remaincritical to the lab’s efforts.

“The need to quickly identify and respond to globalthreats to human health has never been clearer,” saidJosh Gluck, vice president of Global Healthcare Tech-nology Strategy at Pure Storage. “With FlashBlade, re-searchers at McArthur Lab can continue to conductresearch that has widespread global impact.”

For its part, Pure Storage (NYSE: PSTG), based inMountain View, Calif., is an IT pioneer that delivers

storage as-a-service in a multi-cloud world.Pure’s FlashBlade now serves as the digi-

tal backbone of the McArthur Lab, as itcontinues to combat the spread of su-perbugs and other major issues, suchas the rapid spread of antimicrobialresistance – the ability of pathogens todevelop resistance to antibiotics.“FlashBlade generates results in threehours or less for insights that cansave lives and lead to better public

health decision,” said McArthur. “Notonly can McMaster process more data,

it is processing better data, leading tobetter, faster outcomes. Since

moving to a modern data-centric infrastructure, the

lab’s results have becomeeven more accurate.”

VI

EW

PO

IN

T

Using flash storage, the lab is able to analyze huge data sets 24 times faster than before.

Modern storage infrastructure helpsMcArthur Lab speed up lifesaving results

Andrew G. McArthur, Ph.D.,associate professor at McMasterUniversity in Hamilton, Ont.

“What once took one to two days to conduct, now can be done in two to three hours,” commented AndrewMcArthur, lab director.

Technology can help with Canada’s chronic wait times problemBY DR . CHRIS SIMPSON

Never waste a good crisis”, thechange management expertssay. As the COVID-19

pandemic continues, and we reflecton the massive disruption it hascaused, I have found myself searchingfor opportunities for healthcaretransformation amongst the chaos.

Many have said that this crisis hasnot really generated new problemsso much as it has exacerbated oldones. One good example of this isthe Canadian healthcare system’schronic wait time problem. Waitingfor medical care is as Canadian asmaple syrup and poutine. But now,thanks to COVID-19, things havegotten a whole lot worse.

Even before the pandemic,Canada was a poor wait timesperformer; finishing at or near thebottom of the list of peer countrieswhen it comes to access to primarycare, specialist care, tests, proceduresand surgeries.

Now, as we look back over thepast few months, we see that theproblem has worsened, withbacklogs in surgeries and procedures(let alone the backlog of other kindsof medical care) estimated to be inthe hundreds of thousands acrossthe country.

As with any complex issue, theanswer is never easy, nor is there asingle “silver bullet”. One highlyeffective but underutilized solutionmay just be ready for prime time. Wehave always known it works, butmaybe we just needed a crisis tomove to widespreadimplementation. I’m speaking, ofcourse, of single entry models(SEM), also known as central intake.

Recall your last trip to Walmart.When you go to pay for yourpurchases, you’ll notice that therearen’t 12 lineups for 12 cashiers.There is one line up for 12 cashiers.Whoever gets to the front of the linefirst goes to the next availablecheckout. That’s a single-entry

model. And Walmart does it this wayfor a reason: you can get morepeople through faster.

The same principles apply forqueues of patients waiting for amedical service, like hipreplacement, a consultation with apsychiatrist, access to addiction

services, or adiagnostic test,like an MRI. Itis far moreefficient forpatients towait in onequeue, and goto the firstavailableprovider, thanif they formnumerous

queues for those same providers. Common intake is cost effective,

more equitable, and it drives downwait times. The goal is simply to getthe right patient in front of the rightprovider at the right time.

Too often, still, referring doctorsare not aware of the referral optionsout there for the myriad of clinicalservices available, and they default tothe people and services they knowand have known, usually with littleidea what the wait times are.

The result is that a patient waitingfor a medical service may findthemselves waiting for monthslonger than another patient downthe street for the same service,simply because the referringprovider doesn’t have the toolsavailable to “comparison shop”.

But there are some areas that aredoing great work in this space. TheMississauga Halton region ofOntario has perhaps one of thewidest implementations of a singlepoint-of-entry system. Managed bythe Mississauga Halton CentralIntake program, the programreviews, triages and routes referralsfor diabetes, foot care, mental health,addictions and hip and knee


Dr. Chris Simpson


“

V I E W P O I N T

BY ROBERT KAUL

It’s not news that COVID-19 has signif-icantly impacted the healthcare systemand the way services/programs are de-

livered. What is worth noting is that thepandemic has spurred the introduction ofnew models of care and patient interac-tions that could have long-lasting, positiveoutcomes.

At Cloud DX we recently joined an ini-tiative led by Curatio Inc., a digital healthcompany headquartered in Vancouver, tosupport the urgent need of patients whoare no longer able to take part in face-to-face outpatient programs due to the pan-demic. The goal of the aptly namedStronger Together program is to safelysupport outpatients at home during thepandemic, especially those who are frail,vulnerable and isolated.

BC Interior Health, The University ofBritish Columbia,Pacific Blue Crossand United HealthNetwork are amongthe partners in-volved in the pro-gram. Stronger To-gether is supportedby a $1.4 million co-investment from theDigital TechnologySupercluster, an in-dustry-led consor-

tium investing in innovation projects withfunding from Canada’s federal Ministry ofInnovation, Science and Industry.

People living with chronic disease, thedisabled and elderly typically visit hospi-tals and community centers for rehabilita-tion and disease management programs ingroup settings.

These programs deliver education, in-formation, exercises, coaching and criticalpeer support, that address both the med-ical and psychosocial needs of patients andfamilies. Often, these programs also havesome form of monitoring and clinicalevaluation to catch issues before they be-come acute and thus prevent complica-tions and hospitalizations.

It’s not considered safe for these individ-uals to participate in these types of pro-grams during the pandemic, so a virtualcare solution makes sense. However, theproviders who typically deliver these pro-grams lack the technology to deliver themvirtually and don’t have the remote moni-toring tools to safely track patients at home.

Similarly, there is no commonly acceptedstandard for this type of delivery and noplatform on which multiple offerings can beintegrated. In fact, use of mainstream ‘con-sumer’ technologies and platforms are oftenproblematic for healthcare organizations asthey may not have the privacy and regula-tory requirements needed, nor have theygenerally been designed for clinical use.

The Stronger Together initiative was de-veloped to fill these gaps and create sup-port for patients. Stronger Together lever-ages Curatio’s secure, private social net-work to deliver peer support, coachingfrom nurses and experts, evidence-basedhealth literacy programs and daily check-ins alongside Cloud DX’s award-winningremote patient monitoring technology.

A multi-phased project, Phase I ofStronger Together was deployed within BCInterior Health to help patients prepare forjoint replacement surgery. Together, pa-tients worked through weekly evidence-based educational programs and partici-pated in daily group discussions led by

Community Coaches, all while checkingand monitoring their vitals each morningfrom the comfort of their homes.

Prior to the program launch, researchand interviews identified a gap in patients’understanding of the power and controlthey have over their own health outcomes.

Also highlighted was the need for informa-tion around the importance of a patient’srole in healthcare. There turned out to be areal potential for improving health out-comes by assisting with patient educationand process transparency and to provide

BC launches new form of virtual care for patients with chronic illnesses

• Turn-key and customizable

solutions for your hospital, clinic,

or long-term care home.

• Built-in clinical features like

digital charting, e-prescriptions,

integration with peripheral

diagnostics, and more.

• Proven partner for seamless

implementation, onboarding, and

ongoing operational support.

Learn more at getmaple.ca/hospitals

Discover how our virtual care solutions can help you better

serve your patients and modernize your healthcare facility.

Unlock new care delivery models with Canada’s

most trusted telemedicine provider.

Improve patient outcomes and access to care through telemedicine.

Robert Kaul


h t t p : / / w w w . c a n h e a l t h . c o m N O V E M B E R / D E C E M B E R 2 0 2 0 C A N A D I A N H E A L T H C A R E T E C H N O L O G Y 1 7

FE

AT

UR

E

RE

PO

RT

BY DIANNE DANIEL

onsider the following urgent situations: • Concerned parents rush their childto The Hospital for Sick Children(SickKids) emergency department inToronto. After checking in with the

triage nurse, they go to the waiting area and a shortwhile later, they receive a message that a urine testand X-ray have been ordered and are told what to donext. By the time they see the physician, she’s alreadyarmed with the results and makes an informed deci-sion about their child’s care on the spot.

• The condition of an inpatient at St. Michael’s Hos-pital in Toronto starts to deteriorate, putting him athigh risk for an adverse event. A secure email is im-mediately sent to the nurse’s desk and the attendingphysician receives a page, allowing for early care in-tervention and preventing a sudden trip to the ICU.

• A critically ill infant in the neonatal intensive careunit (NICU) at Southlake Regional Health Centre inNewmarket, Ontario, starts to develop sepsis, a con-dition that appears suddenly and can turn fatalwithin hours. Clinicians are alerted and quickly in-tervene with life-saving measures.

The common thread between all three of these re-search scenarios, currently under way in Ontario? Abranch of artificial intelligence called machine learn-ing that is poised to reshape medicine, one predictivemodel at a time.

“This is a major data and computing revolution ofour time,” said Dr. Amol Verma, an internist and sci-entist at St. Michael’s. “If we’re going to figure outhow to harness it for healthcare, to really improvehow we deliver care for our patients, it’s going to takea lot of hard work.”

Dr. Verma is part of a multi-disciplinary team atthe hospital working on an early warning systemcalled CHARTwatch, designed to reduce mortalityand improve the quality of care of patients on the gen-eral internal medicine ward. One of the first Canadianhospitals to establish an in-house data science and ad-vanced analytics team – including a multimillion-dol-lar infrastructure investment and the creation of avice-president of Data and Analytics position – St.Michael’s recognizes machine learning as a “key areaof healthcare development and innovation,” he said.

In simple terms, machine learning is the processof using algorithms to teach a computer to make ac-curate decisions and predictions based on data. Thegoal of CHARTwatch is to improve real-time clinicaldecisions by automating the process of rapidly col-lecting and analyzing data from the hospital’s elec-tronic medical record (EMR).

Whereas similar predictive models, widely used inthe U.K., were designed to analyze five to 10 datapoints and use simple statistical algorithms,CHARTwatch analyzes more than 100 data elementsstored in the hospital’s EMR. Its complex algorithmsrun like a digital assistant in the background, operat-ing in parallel to the hospital staff ’s existing work-flow and making accurate predictions based on real-time patient information, including length of stay.

It took two years, a great deal of input from clini-cians and analysis of 10 years’ worth of historicaldata to develop the model, which is designed to pro-vide 24 to 48 hours advance notice when a patient isat risk of deteriorating. Clinicians are alerted by se-

cure email, page or through the hospital’s physiciansign-out tool and patient risk scores are assigned ac-cording to three categories: green, yellow or red.

Before launching their model in a clinical settingthis past summer, the St. Michael’s team conducted areal-time comparison between the model and clini-cians. After comparing more than 3,000 predictioninstances over a four-month period, they determinedtheir model was 20 percent more accurate than clin-icians at identifying patients who were either goingto die or end up in the ICU, said Dr. Verma.

“I think what’s really exciting about this is onceyou deploy to the real world, it’s actually not that themodel is better than the clinician. Ultimately, it’s thecombined intelligence: How does the model informthe clinical judgement of the doctor, who can thenhelp the patient,” he said.

Now that the model is live, researchers are carefullymonitoring results through weekly team meetings and

a full year of qualitative and quantitative evaluation isplanned. Anecdotal evidence shows that CHARTwatchis accurately identifying outlier patients, promptingclinicians to reassess patients they may have otherwisemissed. At the same time, the model sometimes tellsclinicians what they already know and in some in-stances, the clinicians override the model’s decision.

An added benefit is that the model is being used tostrengthen communication between patients, theirfamilies and clinicians. “We had patients who had avery long and complicated hospital stay and theirfamily members were very anxious about leavingthem overnight on the ward,” said Dr. Verma. “Themodel predicted the patient was low risk and we wereable to share that with family and reassure them.”

Improving communication is also one of the ob-jectives of researchers at SickKids who are working todevelop early warning systems to detect sepsis andcardiac arrest, as well as predictive models to bettermanage patient flow and patient volumes.

The hospital’s multidisciplinary research teamfollows a structured pipeline to advance machinelearning projects, starting with clinical use case de-sign, followed by data acquisition and preparation,model development, model and user validation, andending with clinical integration. Legal, privacy andethical considerations – including steps to identifymodel bias and ensure fairness and equity regardingwho will benefit from the model – are emphasizedthroughout the process.

The end goal is to translate academic studies intoreal-world, meaningful action, said Dr. Devin Singh,physician lead for Clinical AI and Machine Learningin Pediatric Emergency Medicine at SickKids.

“You could run these models as decision support,where all that’s happening is an alert is firing butthere’s no impact directly on patient care,” said Dr.Singh, who recently launched a health technologystart-up lab called Hero AI to help translate leading-

edge research at SickKids into the clinical set-ting. “The more interesting step is to avoid thataltogether and go straight into action,” he added.What’s different about their approach is that theyare circumventing the challenge of alert fatigue –which occurs when a predictive model generatesfalse positive results – by setting their modelthresholds high. For example, a model could beset to react only when it is 99 percent certain it

has identified a positive case or that a wait-ing ER patient requires a specific test.“The sheer act of programming yourmodel that way means you’re going toend up missing a bunch of cases that

come through, but when the model doesfire, you’re so confident the result is correct,you can add automation to that workflow,”explained Dr. Singh.One project, expected to go live in the ER thisyear, is a model that predicts downstream eventsafter a patient arrives. The model collects and an-

alyzes triage data from the electronic health record,such as demographic details, vital signs and pre-ferred language, as well as nursing notes, to accu-rately detect common ailments that are most likelyto require routine tests such as urinalysis, X-ray orultrasound. It then triggers the automatic genera-tion of orders and notifies patients to complete thetests while they wait to be seen.“Those tests can now be done ahead of time, dur-

ing the period of time that you’re waiting, adding ef-ficiency and value to your stay, and by the time thedoctor sees you, they can go straight to decisionmaking,” he said. “That’s really powerful.”

Instead of trying to program a predictive modelthat will be right for every patient, every time, the ap-proach focuses on automating care for a subset ofpatients with confidence. The end result is a two-pronged ER workflow where testing is automated forthose patients identified by the model, while remain-ing cases follow the normal pathway of care.

“Our hypothesis is by adding efficiency for 20 per-cent of patients, we’re actually adding efficiencies toeverybody, said Dr. Singh.

Other leading-edge research being explored atSickKids focuses on using available data to predictpatient volumes, including weather and traffic pat-terns, COVID case numbers, prescription usage andpossibly even Google search hits. Researchers are alsolooking at developing a chatbot to serve as a virtual

AI is getting better at detecting problems in patients, and at alerting the right clinicians.

Machine learning makes progress in clinical care at hospitals across Ontario

ILLU

STRA

TIO

N:

LIN

DA

WEI

SS

C


A I A N D M A C H I N E L E A R N I N G

assistant for parents at home, helpingthem to know when they need to take achild to the ER.

“If you come too early and you’re tooreflexive in your response to unwell chil-dren, then you start to overcrowd andoverwhelm a publicly funded healthcaresystem and that leads to adverse out-comes,” he explained, noting that a late tripto the ER can also be devastating.

Carolyn McGregor, the Canada Re-search Chair (Alumni) in Health Informat-ics based at Ontario Tech University in Os-hawa, said the simplest way to think aboutmachine learning is to call it automated de-cision making. Known for her break-through work to detect early onset of sepsisin infants in neonatal intensive care units(NICUs), Dr. McGregor is now workingwith a variety of stakeholders to further ad-vance machine learning in healthcare.

Her patented approach to monitoringinfants in the NICU – which resulted inthe Artemis Cloud Health Analytics-as-a-Service (HAaaS) platform – examines theinterplay between respiration rate variabil-ity and heart rate variability. The model es-sentially watches for changes in heart rateor breathing that are signs a child is deal-ing with infection, and then alerts physi-cians to intervene and decide next steps.

The cloud environment supportingArtemis is provisioned by the Centre forAdvanced Computing at Queen’s Univer-sity in Kingston, Ontario. Artemis runscontinuously in the background of theNICU, processing roughly three milliondata points per infant per hour, as it ana-lyzes changes in infant physiology.

The approach is unique in that it allowstreating physicians to ‘see’ changes in be-haviour they might not pick up at bedside.“The second by second numbers (on amonitor) help them to see in the moment ifthe patient is still alive, but it doesn’t answerquestions about how the patient’s physiol-ogy is changing such that they are showingsigns of infection, a hemorrhage or someother condition,” explained Dr. McGregor.

Based on a study of Artemis deploy-ments at McMaster Children’s Hospi-tal in Hamilton, Ontario, and South-

lake Regional Health Centre in Newmarket,Ontario, the platform was proven to main-tain service availability as high as 99.7 per-cent and McGregor’s team was on the cuspof releasing a decision support protocol forearly detection of sepsis when the COVID-19 pandemic launched them into a holdingpattern. “We’re looking forward to when wecan restart the studies,” she said.

One of the challenges to the approachof using real-time physiological monitor-ing to inform predictive models is signalquality. In the NICU, the advantage is thatbabies sleep for long periods, allowing foruninterrupted data flow. The data fromMcMaster’s NICU is collected from PhilipsIntellivue monitors, while Southlake usesGE Dash monitors. As Dr. McGregor’steam now works to apply the platform inother real-world environments, they needaccess to low-cost, easy-to-wear sensorscapable of transmitting data at a fast fre-quency to a public cloud.

“We’re trying to create the digital twinof the human and then do analysis on theinformation of that digital twin,” she ex-plained. “At the moment, we’re still havingtrouble getting that exact replica and en-

suring the digital twin is an actual repre-sentative of the human.”

Last year, Ontario Tech University part-nered with University of Technology Sydney(Australia) to launch a Joint Research Centrein Artificial Intelligence for Health and Well-ness. As a co-director of the centre, Dr. Mc-Gregor is applying her expertise to improvehealth, wellness, resilience and adaption inseveral different populations. The extended

technology platform is called Athena Cloud,and incorporates environmental and activitydata in addition to physiological data.

The centre is collaborating with theCanadian Space Agency to predict how as-tronauts adapt to weightlessness, as well aswith firefighters, tactical officers andmembers of the Department of NationalDefence to monitor physiological changesthat occur during intense training scenar-

ios, such as running into a burning build-ing. Another project is working to developan early indicator of aggressive behaviourin mental health patients.

“As much as I realize the possibilities areinfinite, I’ve been very fortunate that otherpeople are coming to me and saying, ‘Canwe use it to do this?’” said McGregor. “Theanswer is yes you can and the beauty ofthat is we’re solving real problems.”


BY ALAN SARDANA AND DR. JOSHUA LIU

n analysis published in theSeptember 2020 issue of theCanadian Medical AssociationJournal estimates that it

could take one and a halfyears to clear the backlog of almost150,000 surgeries created between Marchand June 2020, due to the pandemic. It isimperative for our healthcare system tonot only help Canadians get surgery, butto do so safely and in a manner thatminimizes risks associated with hospitalcare during the pandemic.

Digital patient engagement is apromising virtual care technology thatpresents an effective way to deliver safersurgery during COVID-19. Digitalpatient engagement involves empoweringpatients to be guided through their pre-and post-surgery care via theirsmartphone, tablet or computer.

Patients receive electronic remindersabout key steps in their care plan (e.g.when to stop eating or drinking beforesurgery), access interactive education(e.g. exercise videos for rehab) and tracksymptoms or progress (e.g. pain scores,vital signs, etc.). Healthcare providers atthe hospital can get alerts and accessdashboards to monitor patient progress

(e.g. photos of surgical incisions) andcatch problems earlier.

While this may sound futuristic, aCanadian health technology companycalled SeamlessMD has been pioneeringdigital patient engagement for severalyears. Prior to the pandemic, Seam-lessMD’s platform was being used byCanadian healthcare organizations toguide patients through surgery, such asat Sunnybrook Health Sciences Centreand The Ottawa Hospital.

In fact, Trillium Health Partnerslaunched the SeamlessMD platform forCardiac and Orthopedic surgery duringthe pandemic, and when in-person, pre-surgery education classes had to becancelled, the hospital innovated quicklywith SeamlessMD to deliver pre-recorded classes virtually to patientsthrough the platform.

Leveraging Patient-ReportedOutcomes (PRO) data collected onplatforms such as SeamlessMD (e.g. painscores, vital signs data, etc.), with A.I.and Machine Learning to better predictadverse health outcomes and deliversafer surgery has always been a vision forthe Canadian healthcare system. Now,the COVID-19 pandemic appears to bethe turning point needed to not onlyaccelerate the adoption of digital patient

engagement platforms, but also the useof AI & Machine Learning to delivermore predictive care.

In August 2020, the Digital TechnologySupercluster announced the investmentof about $30 million in federal fundingfor 17 projects to help tackle COVID-19.One of these projects is being led bySeamlessMD to help hospitals use digitalpatient engagement technology andmachine learning to deliver safer surgery

during COVID-19, thereby helping toreduce the backlog of surgeries.

A consortium of seven hospitals inCanada, including Trillium HealthPartners and Michael Garron Hospital,have already signed up to participate,and this number continues to grow.Besides SeamlessMD, industry partnersinvolved in the initiative include AltaML,Xerus Medical and Excelar Technologies.

Through this project, Canadianhospitals will integrate the SeamlessMDplatform with their existing Electronic

Medical Record systems, and enablepatients to be digitally screened for pre-surgery risk factors, COVID-19symptoms and post-surgery symptomsof complications (e.g. fever, surgicalincision photos, etc.). This will allowhealthcare providers to remotely assesspatient readiness for surgery, avoidunnecessary in-person follow up visitsand catch problems earlier in the post-surgery recovery process.

In addition, the collaboration willinvolve training machine learningmodels with both historical surgery dataand real-time symptom data to betterpredict the risk of cancellations,readmissions and emergency roomvisits. This project represents a paradigmshift in healthcare, from a historicallypaper-based model of patient educationto a digital model of intelligent,personalized patient care.

While several of the hospitals involvedin the initiative are already using digitalpatient engagement technology, many willbe implementing this type of technologyfor the first time. While we do not knowhow long the impacts of the pandemic willlast, one thing we know for sure is that thepandemic has accelerated Canadianhealthcare into a far more digital future,and there is no turning back.

How machine learning can reduce the backlog of surgical procedures

BY RITA WILSON RN, MN, M ED.

new era in healthcare and in nurs-ing has begun in Canada. COVID-

19 has accelerated the adoptionof virtual care delivery models

across the country. And health technolo-gies powered by artificial intelligence (AI)such as predictive analytics, robotics, vir-tual care apps, and smart hospitals are be-ginning to appear on the Canadian health-care landscape.

As demand for artificially intelligenthealth technologies (AIHTs) grows, clini-cal settings in Canada will undergo signif-icant changes that will have a ripple effectacross all health sectors, impacting allhealthcare professionals. The nursing-AI-HTs interface will require new skills andexpertise to ensure the integrity and ethicsof the nurse-patient relationship.

This emerging trend gives rise to im-portant questions, including: How willthese technologies influence thepatient/family/caregiver’s experience ofcompassionate care? What impact willthese technologies have on nurses, pa-tients, families and caregivers? And, howwill nurses safeguard compassionate carein an age of AI?

The Registered Nurses’ Association ofOntario (RNAO), in partnership with AMSHealthcare, is raising awareness of theemerging AI and nursing interface to stim-ulate critical reflection and fuel discussionabout actions that nurses and other stake-holders need to take to shape AIHTs.

“This new world for our profession, inwhich nurses, AIHTs and compassionate

care co-exist, can be seen as threatening orexciting, depending on how we choose toembrace it, shape it and prepare for it,” saidRNAO CEO Dr. Doris Grinspun. “Thereare endless possibilities with AIHTs to en-hance care and advance best practices usingpredictive analytics that enrich nurses’ de-cision-making power.

In addition, the potential to increase ac-cess to health services, and to support peo-ple at home with virtual care apps is real,especially if leveraged for hard to reachpopulations.”

Research suggests that between eight and16 percent of nursing time is spent on non-nursing activities and tasks that could bemanaged by advanced technology. “The im-plementation of AIHTs to perform routineactivities would provide nurses with the op-portunity to spend more quality time oncompassionate patient care,” said AMS CEOGail Paech.

“Technology could play an enormousrole in the future, providing solutions forclinical activities including patient vitalsigns monitoring and alerting, or falls mon-itoring and prevention,” she said.

In the age of AI, an increase in virtualnursing roles is likely. The deployment ofAIHTs to support these roles will requirenurses with the knowledge, judgment andskills to effectively use the new technologies.Future nurses will need adequate educationand training, as well as decision supporttools to identify patients that are appropri-ate for virtual care, interventions that cansafely be completed virtually, and, clinicalconditions that are conducive to patients’use of AIHTs, such as virtual care apps.

A thorough review and reform ofnursing curricula to ensure congruencyof the nursing role with present needsand future demands of the emerging AI-HTs is also critical.

The implementation of these complextechnologies will necessitate governancestructures that integrate nursing informat-ics roles (e.g., chief nursing informatics of-ficer) to assist nursing and other executiveleaders to make informed clinical andbusiness decisions and to lead the techni-cal aspects of the implementation. It is im-portant for executive leaders to advocatefor technology-related governance struc-tures that include these roles and affordnurses decision-making authority.

The emerging future constitutes a ma-jor paradigm shift that necessitates strongand proactive nursing leadership – in all

roles and sectors. This leadership is vital toactively assess and guide the nursing/AI in-terface, encourage and support the contri-bution of nurse leaders in all roles and sec-tors, and ensure fulsome patient engage-ment throughout the technology life cycle.

RNAO and AMS recently published areport titled Nursing and CompassionateCare in the age of Artificial Intelligence: En-gaging the Emerging Future.

It outlines key recommendations thatserve as urgent calls to action on the inter-face of Nursing and AI. RNAO and AMSwant the report to spark vigorous discus-sion on what is needed to shape the emerg-ing future. The report is available atRNAO.ca

Rita Wilson RN, MN, M Ed. is the eHealthProgram Manager at RNAO.

Patients receive electronicreminders about key steps intheir care plans, and they cantrack symptoms and progress.

Gail Paech, CEO of AMS Doris Grinspun, CEO of the RNAO

As AI works its way into healthcare, nursing roles must be integrated



A

A

BY SHIRLEY FENTON

We live in the information ageand an era of Big Data.However, without analytics,that data can be meaning-

less. Organizations urgently need peoplewho know how to explore, analyze, and ex-tract value from data – and deliver the rightinsight at the right time to the right people.

A new micro-certification in Data An-alytics is now available, thanks to a part-nership between McMaster UniversityContinuing Education (McMaster CE)and National Institutes of Health Infor-matics (NIHI).

In June 2020, eCampusOntario, aprovincially-funded, non-profit organiza-tion, awarded a $15,000 grant to McMas-ter CE to pilot this program, as part oftheir initiative to broaden micro-certifica-tion in Ontario by working with universi-ties and colleges in collaboration with in-dustry partners.

Micro-credentials are a hot new ap-proach to education, a response to theskills gap caused by new technologies,rapid digital transformation in industry,and the need to continuously upskill. Es-sentially, a micro-credential is official cer-tification of acquisition of a specific skillor set of skills or capabilities.

It is awarded in a digital form, which isverified, secure, and shareable with peers,employers. and educational providers.Digital credentials can be curated, anno-tated, and distributed over digital net-works under the earner’s control.

According to RMIT University in Aus-tralia, “Micro-credentials also certify an in-

dividual’s achieve-ments in specificskills and differ fromtraditional educationcredentials, such asdegrees and diplo-mas, in that they areshorter, can be per-sonalized and pro-vide distinctive valueand relevance in thechanging world ofwork.”

“Micro-certifications are a fantasticcomplement to our traditional Certificateand Diploma programs,” said Dan Piedra,assistant director, Program and OnlineDevelopment at McMaster CE. “Pairingtheoretical knowledge with micro-certifi-cations allows individuals to clearlydemonstrate their specific skills and capa-bilities to future employers. The funda-mental skills and competencies neededfor today’s jobs are constantly evolving –

micro-certifications are a great way forpost-secondary institutions to stay agileand meet industry demands.”

Why data analytics?Skills in data analytics are one of the

most sought-after skills by employers.LinkedIn’s 2020 Global Talent Trend Re-port identified data analytics as one of themost in-demand skills for employment.

“Data analytics is a strategic asset to anyorganization to support evidence-baseddecision-making,” said Trevor Strome, in-structor for the NIHI-McMaster CE pro-gram, and one of Canada’s leading dataanalytics experts.

“Data overload can risk impeding, notimproving, the decision-making ability ofleaders, managers, and quality Improve-ment teams,” added Strome, who is alsothe author of Healthcare Analytics forQuality and Performance Improvement.He emphasizes that it is not only the vol-ume, but also the variety of data that hasgrown exponentially with the inclusion ofmultimedia and image files.

Strome said the courses in the NIHI-McMaster CE program are applicable toprofessionals in all industries. For his part,Strome works in two different fields: he isthe director of Digital Airport Solutions atthe Winnipeg Airports Authority, as well asassistant professor in the Department of

Emergency Medicine at the University ofManitoba.

The Data Analytics program is designedto help professionals develop their skills inthe fundamental elements of data analyt-ics. A digital credential will be awardedupon successful completion of the threecourses: Data Analytics I, II and III, whichare offered through the NIHI-McMasterCE partnership. All courses are online,with more than 30 hours of instruction,including both live sessions and recordedcontent.

The courses were developed by NIHI inclose collaboration with the dedicated In-structional Design team at McMaster CE.The micro-credential will be issued by Mc-Master CE.

This program is aimed at anyone inter-ested in learning how data analytics cansupport evidence-based decision-makingthat reduces costs, improves outcomes,and provides better service to clients, cus-tomers, and constituents. The programwill be of specific interest to executives,clinicians, technicians, e-health teams,quality improvement teams, data scien-tists, business analysts, database adminis-trators, and solution providers.

All three courses are now open forenrolment. For further information in-cluding how to register, please see:www.nihi.ca.

Shirley Fenton is Vice-President, National In-stitutes of Health Informatics; Co-Founder,Waterloo MedTech.

Transform decision-making with a micro-certification in Data Analytics

BY THOMAS HOUGH CMC

In every healthcare technology event,and in every trade journal, you cannow expect a barrage of information

on how Artificial Intelligence will changeDiagnostic Imaging and the entire health-care delivery system. The big question iswhen will we witness the appearance ofproducts or services that will improvehealthcare delivery and what part ofhealthcare delivery will be first affected?

Diagnostic Imaging is one of the areasfront of mind for us; to date, GE Health-care, for example, has Health Canada ap-proval of a system that can better iden-tify cases of pneumothorax and priori-tize them for readings by radiologists.For its part, Intelerad has been workingwith Zebra Medical to make AI-poweredtools available to Canadian radiologists,as well.

At the same time, there are many otherhealthcare initiatives that could benefitfrom AI in healthcare, including but notlimited to: Screening; Psychiatry, PrimaryCare, Disease Diagnosis, Telehealth, Elec-tronic Health Records, Drug Interactionsand Creation of New Drugs and workflowoptimization.

St. Michael’s Hospital, Toronto, hasbeen using AI for improved schedulingand workflow optimization. As we note inour feature section, St. Mike’s is also devel-oping an AI system that will detect patientdeterioration. Unlike many Early WarningSystems that use six or seven variables todetect patient decline, the St. Mike’s appli-cation uses more than 100! No wonder ar-tificial intelligence is required!

When we think about AI, most of us are

familiar with the terms expert system andmachine learning. But there are additionalcategories that data scientists are workingwith, and that laypersons are now becom-ing more familiar with. Here are a few ofthese categories:

• Artificial General Intelligence (AGI)refers to a type of artificial intelligence thatis broad in the way that human cognitivereasoning can be broad; which can do dif-ferent kinds of tasks well, and that reallysimulates the breadth of the human intel-lect, rather than focusing on more specific

or narrower types oftasks. • Narrow ArtificialIntelligence (NarrowAI) is a specific typeof artificial intelli-gence in which analgorithm outper-forms humans in anarrowly definedtask. Unlike GeneralAI, Narrow AI fo-cuses on a single

subset of cognitive abilities and advances inthat spectrum.

• Strong Artificial Intelligence (Strong AI)is an artificial intelligence paradigm thathas mental capabilities and functions thatmimic the human brain. In the philosophyof Strong AI, there is no essential differ-ence between the piece of software, whichis the AI, exactly emulating the actions ofthe human brain, and actions of a humanbeing, including its power of understand-ing and even its consciousness. Strong AI isalso known as Full AI.

• Artificial Super Intelligence (ASI) isused to create computer applications

which surpass humans, when comparedto what humans can provide. In otherwords, ASI cognitive ability is superior toa human. ASI is what is required for com-plex clinical diagnostic algorithms capa-ble of developing and publishing a diag-nostic report that will direct a patienttreatment plan.

In conclusion, Narrow AI applicationswill be the fastest and first to be developedfor actual healthcare enterprises to adopt.Some of these Narrow AI apps are cur-rently adopted and employed in Canadianhealthcare facilities today.

General AI Apps are well along the de-velopment path and – like humans – cando a broad range of human cognitive tasksand will be implemented within the nextfive years. Artificial Super Intelligence is byfar the most complex challenge; applica-tions of this sore will be the most sought-after, certainly in the DI sphere. I predictthat we’ll see pilot versions within fiveyears, while Super Intelligence Apps capa-ble of workflow orchestration with com-plex and complete diagnostic and reportcapabilities will appear some 10 years inthe future.

Just think, only three years ago, therewere 70 Diagnostic Imaging AI compa-nies; two years ago there were 200 compa-nies, and as of this year, there are 700 com-panies racing to develop AI technology.Add that to the activity going on in hospi-tals, universities and government labs, andI’d say the next five to 10 years are going tobe interesting.

Thomas Hough is founder and President ofTrue North Consulting, based in Missis-sauga, Ont.


Shirley Fenton

Thomas Hough

AI is quickly developing and taking different forms

fighting disease. We don’t rely onlocal fundraising to buy tanks forthe military.”

Reliance on local fundraising alsodiscriminates against smaller, remotecentres of population like Val d’Orthat don’t have the ability to raiselarge sums of money.

“Radiotherapy for six weeks costs$30,000,” he said. “It’s expensive andtime-consuming, and it’s a very localtreatment. If you give chemo to apatient who doesn’t need it or youdo extensive surgery on a patientand it’s not the right treatment, it’svery expensive.”

PET imaging, he claims, can savemoney and result in better patientoutcomes.

Until now, Abitibi-Temiscaminguepatients prescribed a PET/CT scanhad to travel to Montreal or Gatineau,but 60 percent of them declined theopportunity because of the cost, thetime required, the distance and thedanger of traveling by car – especiallyin winter, said Dr. Lamoureux.

The recent acquisition of PET/CTs in Val d’Or, Sudbury andKelowna will allow patients to remainclose to home and to take advantageof the same advanced imagingmodalities available to patients inlarger communities.

PET/CT in Val d’OrCONTINUED FROM PAGE 14


patients with immediate positive feedbackfor helpful behaviours.

Phase I of the program included the par-ticipation of a clinical counsellor, a socialworker and a physiotherapist, with whompatients could book virtual appointments.The pre-surgical optimization informationtraditionally provided in person to patientswas woven into the four-week program thatwas provided to patients in-app.

Programs consisted of various steps in-cluding videos, worksheets, links to exter-nal references and daily vital sign checkswith the Cloud DX Connected Health Kit,provided to patients free of charge. Pa-

tients had the opportunity to discuss anyof the steps with the peers and coacheswithin their groups and with the in-appcommunity at large.

The four-week pilot program beganwith 22 patients who were awaiting jointreplacement surgery at Penticton RegionalHospital (PRH). Nineteen continued tomonitor their vitals remotely over thefour-week period, and 16 were onboardedinto the Stronger Together app to partici-pate in the four-week in-app program.

Upon onboarding, patients were di-rected to complete a clinically validatedbaseline questionnaire to assess theirhealth and wellness prior to beginning theprogram. At the end of the four-week pe-riod, patients were prompted to completethe questionnaire again to help assess any

changes in responses and overall wellness. Early results from the program are pos-

itive. A notably improved response in out-come measures was observed, and partici-pants who have not yet gone for surgery

appear to have increased their physical ac-tivity after the four-week program. Thegreatest change noted was in self-efficacyfor managing chronic disease. Patient sur-vey feedback was positive and highlightedthe ability to support patients safely and

remotely across social, physical, and men-tal dimensions of their healthcare journey.

Several participating patients reportedfeeling motivated and actively interested intheir own self-care. Others reported feelingthat they received personalized care andenjoyed the feedback from other patientparticipants. Providers reported feelingthat the program helped patients to take agreater interest in their own health.

The Stronger Together team is nowpreparing to enter into Phase II of the pro-ject. This next phase will include a numberof additional programs to support diversepatient populations across Canada as theywork to stay “healthy at home” during theongoing COVID-19 pandemic and beyond.

Opportunities and programs like thesehave propelled us into a new remote world,in spite of pandemic-related disruptions.With careful planning and consideration,new care modalities such as Stronger To-gether can create enduring improvementsin patient outcomes, rather than beingmerely a stopgap solution. We at Cloud DXlook forward to helping create a healthcaresystem where technology continues to beleveraged in unique and exciting ways tocreate better patient outcomes.

Robert Kaul is Founder, President & CEO atCloud DX. The company is headquarteredin Kitchener, Ont.

family physician and co-leader, DigitalHealth, East Toronto Family Practice Net-work. “We’re using it for urgent communi-cation, from physician to physician, nursepractitioner to physician, physician to di-etician, specialist to family doctor.”

Dr. Shah said that she has made good useof Hypercare to reach specialists at MichaelGarron Hospital when questions come upthat can be quickly resolved. The Hypercaremessaging system allows physicians to com-municate with specialists, often enablingtheir patients to avoid trips to the ER.

During the current COVID-19 crisis,keeping patients away from the hospitalwhen possible has been a big plus for in-fection control.

She mentioned the case of a patientwith very poor kidney function and nourologist. Normally, the patient would besent to the emergency department for fur-ther tests and treatment. However, Dr.Shah contacted a urologist at the hospitalusing Hypercare and arranged an officeappointment for the patient the next day.

She said the application has also beenuseful for patients with post-operative com-plications. Instead of sending them immedi-ately to the ER, she has messaged surgeons

using the Hypercare directory and app, get-ting fast responses to clinical questions.

The East Toronto Family Practice Net-work has opened up the Hypercare appli-cation to its 200 family physicians. Not allare using it yet, but many have started, saidDr. Shah. As well, an initiative is in theworks to extend the application to alliedhealthcare professionals in home care, in-

cluding wound care, so that visiting nursesand PSWs could send questions to clini-cians when needed.

“It’s so much better than the phone,”said Dr. Shah. “The phone just isn’t theway people function today. As a doctor, Idon’t have time to put in a call and to beput on hold – not with four or five patientsin my office waiting for me.”

have the option to turn Assist off and gothrough the procedures as a surgeonwould, actually doing it. In this mode thereis no skipping back and forth.

“From there, you are presented with anevaluation that can analyze angles, screwplacements, success of cuts, the amount oftime you took,” said Oliveira. Participantscan experience VR training on their ownor in a classroom setting where the in-structor can compare students’ perfor-mance data. Anyone using it on their ownhas the option to turn off Assist, and nothave the data sent to the dashboard.

Asked if there was any resistance to us-

ing the VR platform, Romero said that atfirst, “Like with anything new, it took a bitof getting used to, pushing the buttons, butonce users get used to it, it is easy to use.”

Romero likes that the evaluation at theend assesses key performance measures,

such as how the user held a drill or made acut in the tibia or femur, as well as the de-grees or angles used. “At the end, it will

thoroughly tell you how you did. A big ad-vantage – I thought that was really cool,”said Romero.

Oliveira said the company is currentlyworking on gathering information fromthis pilot initiative to further adapt the VRtraining experience to nurses. “We willcustomize it and get it right back into thehands of nurses,” said Oliveira.

According to Dr. Goel, VR training’s keybenefits include the ability to increaselearning speed while decreasing costs: “Wedemonstrated that you can learn 570%faster with VR. In a second randomizedcontrolled trial, our platform was identi-fied as being 37 times more cost-effectivethan traditional models of simulation andtraining,” Dr. Goel said.

Speaking about traditional costs versusthose for a “cost-effective, scalable plat-form,” Oliveira said, “For example, it willcost between $5,000 and $10,000 to get asurgeon to a course. A cadaver can costupwards of $5,000, as well, and only pro-vides a single opportunity to learn. Wecan educate an unlimited number ofhealthcare providers, with unlimited rep-etitions, at a streamlined price for theheadset of $1,000.”

It is important to note that VR trainingdoesn’t exist in a vacuum. “It has to becombined with mentorship, didactic andcurriculum training,” said Dr. Goel. Hav-ing VR training in that mix, he observed,“tends to push people beyond their cur-rent level. It enables collaboration and ele-vates critical thinking through a problem-based approach.”

Currently, with COVID-19, “we arelimiting the number of people in the (op-erating) room. So there are fewer learningopportunities,” said Dr. Goel. “VR offers acompletely different approach to learning– experiential, no risk, highly convenient.It’s fully immersive and designed for ac-tive learning.”

referrals for a population of 1.2 million.While the benefits of single-entry

models and common intake seem intu-itive enough, it is also readily apparentthat this can’t be done with papers piledon the corner of the physician’s desk.And certainly, a regional model intend-ing to serve many patients can’t be pa-per-based, fax-based or spreadsheet-based. The clerical requirements wouldbe huge and unmanageable. An en-abling technology is needed.

Novari eRequest is one suchenabling technology – a 3-in-1 systemwith referral management, centralintake, and workflow managementcapabilities. Importantly, it iscustomizable to each client and eachclinical service’s unique needs.

It can be implemented in anorganization like a hospital, across aregion or even province wide, as it

recently was in Saskatchewan, whereNovari eRequest is serving as thetechnology behind the province’ssuccessful COVID-19 testing andassessment program.

For individual hospitals and regions,setting up a single entry for all inboundreferrals, then electronically routingthem to the appropriate clinic or

provider has many benefits, like real-timetracking of referrals, real-time data oninternal wait times, identification of bot-tlenecks, no lost paper/fax referrals andautomated feedback to the sender of thereferral on the status on the referral.

The deferral of care experienced byhundreds of thousands of patientsacross Canada as a result of the

COVID-19-associated slowdown inservices has exposed in the starkest waypossible our country’s chronic access-to-care problem.

Clearing the backlog and getting waittimes down to acceptable levels will takea lot of work by a lot of people. Manysolutions will need to be devised andimplemented. But some solutions arealready ready to go.

Central intake, and the enablingtechnology to make it happen are al-ready in play and have been proven towork. The transparency and efficiencyof a professionally managed wait list fa-cilitated by the appropriate enablingtechnology will go a long way todemonstrating that our healthcare sys-tem truly is worthy of Canadians’ confi-dence and trust.

Dr. Chris Simpson is a Cardiologist atthe Kingston Health Sciences Centre, andVice Dean, Queen’s University School of Medicine, and Chief MedicalInformation Officer, Novari Health.

Wait times problemCONTINUED FROM PAGE 16

BC Interior HealthCONTINUED FROM PAGE 17

Central intake and theenabling technology itrequires are in play and havebeen proven to work.

VR is much more cost-effectivethan sending professionals tocourses, or obtaining single-usecadavers for surgical training.

The Stronger Together programis helping diverse patientpopulations stay healthy athome during the pandemic.



Orthopedic surgery nurses accelerate learning using VR

Michael Garron Hospital deploys enterprise messaging


Healthcare that’s smarter, more connected and alive with possibilities.Our solutions are built for greater connectivity, using cutting-edge software and technology, open architecture for scalability—and delivering the most innovative and responsive tools. With services that provide the critical support clients need to optimize performance and productivity.

Explore your possibilities at ca.allscripts.com

@[email protected]

ca.medical.canon

At Canon Medical, we have created an entire ecosystem of solutions that foster greater collaboration across multiple teams, departments, and technologies, including advanced AI and healthcare IT, to help you deliver a new level of patient-centred care.

Because we understand that when you collaborate, anything can be made possible.

Book your Virtual Experience Today [email protected]

Reimagine RSNA with Canon Medical

artificial intelligence and machine learning - canadian … · 2020. 11. 2. · hospital, in...

Documents