andreas holzinger vo 444.152 medical informatics lecture –version ws science meets...

A. Holzinger 444.152 Med Informatics L11/65

Andreas HolzingerVO 444.152 Medical Informatics

Lecture 1 – Version WS 2012/13Introduction

Computer Science meets Life Sciences: Challenges and Future Directions

[email protected]


Reading on Paper or on any electronic device


1. Intro: Computer Science meets Life Sciences, challenges, future directions

2. Back to the future: Fundamentals of Data, Information and Knowledge

3. Structured Data: Coding, Classification (ICD, SNOMED, MeSH, UMLS)

4. Biomedical Databases: Acquisition, Storage, Information Retrieval and Use

5. Semi structured and weakly structured data (structural homologies)

6. Multimedia Data Mining and Knowledge Discovery

7. Knowledge and Decision: Cognitive Science & Human‐Computer Interaction

8. Biomedical Decision Making: Reasoning and Decision Support

9. Intelligent Information Visualization and Visual Analytics

10. Biomedical Information Systems and Medical Knowledge Management

11. Biomedical Data: Privacy, Safety and Security

12. Methodology for Info Systems: System Design, Usability & Evaluation

Schedule Winter course 2012/13


Science is to test ideas –Engineering is to put these ideas into practice

Your Lecturer: Andreas Holzinger

Head of Research Unit Human‐Computer Interaction Institute for Medical Informatics, Statistics and Documentation, Medical University Graz

Assoc. Prof. for Information Processing, Faculty of Informatics, Graz University of Technology

Since 1999 participation in leading positions in 30+ R&D multi‐national projects, budget 3+ MEUR;

Visiting Professor in Berlin, Innsbruck, London, Vienna and Aachen

300+ publications, 2800+ citations, h‐index = 25, g‐Index = 59 Research field: Computer and Information Science Topics: Informatics > Biomedical Informatics >

Knowledge Discovery > E‐Business > Usability

http://scholar.google.com/citations?user=BTBd5V4AAAAJ&hl=en


At the end of this first lecture you will … … have an overview about some topics in the field of medical informatics;

… have a rough overview of the historical roots of medical informatics;

… be familiar with some differences between Life Sciences and Computer Sciences;

… be are aware of some possibilities and some limits of Computer Science for the application in the medical domain;

… have some ideas of possible future directions of (bio‐)medical informatics;

Learning Goals


Keywords of the 1st Lecture

Big health data Biological information Health information Information quality Modeling artifacts Omics‐based medicine Performance (Computing vs Cognition) Personalized Molecular Medicine Pervasive computing Smart objects


Bioinformatics = discipline, as part of biomedical informatics, at the interface between biology and information science and mathematics; processing of biological data;

Biomedical data = compared with general data, it is characterized by large volumes, complex structures, high dimensionality, evolving biological concepts, and insufficient data modeling practices;

Biomedical Informatics = 2011 definition: similar to medical informatics but including the optimal use of biomedical data;

Classical Medicine = is both the science and the art of healing and encompasses a variety of practices to maintain and restore health;

Cognitive Performance = human capabilities, e.g. time to perform task, number of errors per task, attention etc.

Information Overload = difficulty of end users to make decisions in the presence of too much and too complex information;

Information Quality = can be a means to bring technology and medicine closer together. While some aspects of human factors of technology have already been incorporated into medical informatics, information quality combines aspects understood by both fields (medicine and technology);

Medical Informatics = 1970 definition: “… scientific field that deals with the storage, retrieval, and optimal use of medical information, data, and knowledge for problem solving and decision making“;

Advance Organizer 1/2


Molecular Medicine = emphasizes cellular and molecular phenomena and interventions rather than the previous conceptual and observational focus on patients and their organs;

Omics = field of studies including genomics and proteomics; Pervasive Computing = similar to ubiquitous computing (Ubicomp), a post‐

desktop model of Human‐Computer Interaction (HCI) in which information processing is integrated into everyday, miniaturized and embedded objects and activities; having some degree of “intelligence”;

Pervasive Health = unobtrusive, analytical, diagnostic, supportive, information and documentary functions to improve health care, e.g. remote, automated patient monitoring, diagnosis, home care, self‐care, independent living, etc.;

P‐Health Model = Preventive, Participatory, Pre‐emptive, Personalized, Predictive, Pervasive (=available to anybody, anytime, anywhere);

Technological Performance = machine “capabilities”, e.g. short response time, high throughput, low resources, high availability, etc.

Translational medicine = based on interventional epidemiology; advancement of Evidence‐Based Medicine (EBM), integrates research from the basic sciences aiming for patient care and prevention;

Von‐Neumann‐Computer = a 1945 architecture, which still is the predominant machine architecture of today (opp.: Non‐Vons, incl. analogue, optical, quantum computers, cell processors, DNA and neural nets (in silico);

Advance Organizer 2/2


AAL = Ambient Assisted Living CPG = Clinical Practice Guideline CPOE = Computerized physician order entry DEC = Digital Equipment Corporation (1957‐1998) DNA = Deoxyribonucleic Acid EBM = Evidence Based Medicine ECG = Electrocardiogram EEG = Electroencephalogram EMG = Electromyogram EPR = Electronic Patient Record EPR = Electronic Patient Record GBM = Genome Based Medicine GPM = Genetic Polymorphism GPM = Genetic Polymorphism HCI = Human‐Computer Interaction LNCS = Lecture Notes in Computer Science NGS = Next Generation Sequencing PDP = Programmable Data Processor (mainframe) RFID = Radio‐frequency identification SNP = Single Nucleotide Polymorphism

Glossary


The focus in this lecture is on three issues …

DataInformation

Knowledge

A. Holzinger 444.152 Med Informatics L112/65 ESO, Atacama, Chile (2011)


Wiltgen, M. & Holzinger, A. (2005) Visualization in Bioinformatics: Protein Structures with Physicochemical and Biological Annotations. In: Central European Multimedia and Virtual Reality Conference. Prague, Czech Technical University (CTU), 69‐74

… to microscopic atomistic structures


Our quest: Gaining out of knowledge from this data

Wiltgen, M., Holzinger, A. & Tilz, G. P. (2007) Interactive Analysis and Visualization of Macromolecular Interfaces Between Proteins. In: Lecture Notes in Computer Science (LNCS 4799). Berlin, Heidelberg, New York, Springer, 199‐212.


Jeong, H., Mason, S. P., Barabasi, A. L. & Oltvai, Z. N. (2001) Lethality and centrality in protein networks. Nature, 411, 6833, 41‐42.

First yeast protein‐protein interaction network (2001)

Nodes = proteinsLinks = physical interactions (bindings)Red Nodes = lethalGreen Nodes = non‐lethalOrange = slow growthYellow = not known


First human protein‐protein interaction network (2005)

Stelzl, U. et al. (2005) A Human Protein‐Protein Interaction Network: A Resource for Annotating the Proteome. Cell, 122, 6, 957‐968.

A. Holzinger 444.152 Med Informatics L117/65 Photo by Department of Dermatology, Graz (2012)


Human Disease Network ‐> Network Medicine

Barabási, A. L., Gulbahce, N. & Loscalzo, J. 2011. Network medicine: a network‐based approach to human disease. Nature Reviews Genetics, 12, 56‐68.

A. Holzinger 444.152 Med Informatics L119/65Egerstedt, M. 2011. Complex networks: Degrees of control. Nature, 473, 158‐159.


Hurst, M. (2007), Data Mining: Text Mining, Visualization and Social Media. Online available: http://datamining.typepad.com/data_mining/2007/01/the_blogosphere.html, last access: 2011‐09‐24

Non‐Natural Network Example: Blogosphere


Complex data in socio‐technical systems …

Vespignani, A. (2012) Modelling dynamical processes in complex socio‐technical systems. Nature Physics, 8, 1, 32‐39.


Information object

Example: Social Behaviour Contagion Network

Aral, S. (2011) Identifying Social Influence: A Comment on Opinion Leadership and Social Contagion in New Product Diffusion. Marketing Science, 30,2, 217‐223.


Clouds of data – unordered sequence of points in n‐dim

Zomorodian, A. J. 2005. Topology for computing, Cambridge (MA), Cambridge University Press.

Let us collect ‐dimensional observations:


How do you visualize a four‐dimensional object?”

How do you visualize a three‐dimensional object?”


Clouds of data – unordered sequence of points in n‐dim


Point cloud in topological space metric space

Let us collect ‐dimensional observations:


This is Marty – our intelligent house rabbit


More rabbits …


Sampled point set from a surface

Recovered Topology

Linear surface approximation


Dissimilarity and distance measures as basis

A set S with a metric function d is a metric space

Doob, J. L. 1994. Measure theory, Springer New York.


Computational space

Machine intelligence

Cognitive Space Perception

Human intelligence

Human ComputerInteraction

Data – Information – Knowledge – this is a HCI task!

Visualization

Human intelligence harnesses machine intelligenceHolzinger, A. 2012. On Knowledge Discovery and interactive intelligent visualization of biomedical data. In: DATA ‐ International Conference on Data Technologies and Applications.


The sexiest job of the 21st Century …


Who knows this person?

A hint for those who do not know this currency …


5 m

From Data to Information: Life is complex information

Lane, N. & Martin, W. (2010) The energetics of genome complexity. Nature, 467, 7318, 929‐934.


Living things are able …

to grow …

to reproduce …

to evolve …

to self‐replicate …

to generate/utilize energy …

to process information …Schrödinger, E. (1944) What Is Life? The Physical Aspect of the Living Cell. Dublin Institute for Advanced Studies.


The central Hypothesis of Information Quality in e‐Health

Modern information management can bridge the hiatus theoreticus, the gap between scientific and theoretical knowledge in medicine on the one hand and its application within medical practice on the other hand.

Holzinger, A. & Simonic, K.‐M. (Eds.) (2011) Information Quality in e‐Health. Lecture Notes in Computer Science LNCS 7058, Heidelberg, New York, Springer.


The Roadmap of Information Quality in e‐Health

http://www.medunigraz.at/imi/usab2011


From Clinical Medicine to Molecular Medicine

Yapijakis, C. (2009) Hippocrates of Kos, the Father of Clinical Medicine, and Asclepiades of Bithynia, the Father of Molecular Medicine. In Vivo, 23, 4, 507‐514.


What is biomedical informatics?

Biomedical informatics (BMI) is the interdisciplinary field that studies and pursues the effective use of biomedical data, information, and knowledge for scientific problem solving, and decision making, motivated by efforts to improve human health

Shortliffe, E. H. (2011). Biomedical Informatics: Defining the Science and its Role in Health Professional Education. In A. Holzinger & K.‐M. Simonic (Eds.), Information Quality in e‐Health. Lecture Notes in Computer Science LNCS 7058 (pp. 711‐714). Heidelberg, New York: Springer.


Computational Sciences meets Life Sciences

http://www.bioinformaticslaboratory.nl/twiki/bin/view/BioLab/EducationMIK1‐2


Information and Medicine: From Atom to Population

Altman et al. (2008) Commentaries on "Informatics and Medicine: From Molecules to Populations". Methods of Information In Medicine, 47, 4, 296‐317.


Future p‐Health Model – A 6 P’s paradigm

Zhang, Y. T. & Poon, C. C. Y. (2010) Editorial Note on Bio, Medical, and Health Informatics. Information Technology in Biomedicine, IEEE Transactions on, 14, 3, 543‐545.


Computer: Von‐Neumann Architecture

CPU

Internal Memory

Controller(BIOS, OS, AP)

Internal MemoryShort term: RAMLong term: ROM

External MemoryLong term:

HDD, CD, Stick etc.

MonitorPrinterModemNetworketc.

KeyboardMouse

Graphic PadMicrophoneModemNetworketc.

INPUT

OUTPUT

Processor

Holzinger (2002), 90‐134


Gordon E. Moore (1965, 1989, 1997)

Computer: Technological Performance / Digital Power

Holzinger, A. 2002. Basiswissen IT/Informatik Band 1: Informationstechnik. Das Basiswissen für die Informationsgesellschaft des 21. Jahrhunderts, Wuerzburg, Vogel Buchverlag.


Beyond Moore’s Law ‐> biological computing

Cavin, R., Lugli, P. & Zhirnov, V. 2012. Science and Engineering Beyond Moore's Law. Proc. of the IEEE, 100, 1720‐49 (L=Logic‐Protein; S=Sensor‐Protein; C=Signaling‐Molecule, E=Glucose‐Energy)


Va

st re

duct

ion

in c

ost,

but

enor

mou

s c

apab

ility

Cf. with Moore (1965), Holzinger (2002), Scholtz & Consolvo (2004), Intel (2007)

.

Computer cost/size versus Performance


… using technology to augment human capabilities for structuring, retrieving and managing information

Old Dream of Mankind …

Harper, R., Rodden, T., Rogers, Y. & Sellen, A. (2008) Being Human: Human‐Computer Interaction in the Year 2020. Cambridge, Microsoft Research.

A. Holzinger 444.152 Med Informatics L147/65Photo by Institute of Medical Informatics, Graz (1970)

1970turning knowledge into data


1970+ Begin of Medical Informatics Focus on data acquisition, storage, accounting (typ. “EDV”) The term was first used in 1968 and the first course was set up 1978

1985+ Health Telematics Health care networks, Telemedicine, CPOE‐Systems etc.

1995+ Web Era Web based applications, Services, EPR, etc.

2005+ Ambient Era Pervasive & Ubiquitous Computing

2010+ Quality Era – Biomedical Informatics Information Quality, Patient empowerment, individual molecular medicine, End‐User Programmable Mashups

Four decades from Medical to Biomedical Informatics


four decades later …

turning data into knowledge


Example: Pervasive Computing – Smart Objects (1/5)

Holzinger, A., Schwaberger, K. & Weitlaner, M. (2005) Ubiquitous Computing for Hospital Applications: RFID‐Applications to enable research in Real‐Life environments 29th Annual IEEE International Computer Software & Applications Conference (IEEE COMPSAC), 19‐20.

EPR


Holzinger et al. (2005)




Holzinger, A., Schaupp, K. & Eder‐Halbedl, W. (2008) An Investigation on Acceptance of Ubiquitous Devices for the Elderly in an Geriatric Hospital Environment: using the Example of Person Tracking In: Lecture Notes in Computer Science (LNCS 5105). Heidelberg, Springer, 22‐29.



Holzinger, A., Nischelwitzer, A., Friedl, S. & Hu, B. (2010) Towards life long learning: three models for ubiquitous applications. Wireless Communications and Mobile Computing, 10, 10, 1350‐1365.



Alagoez, F., Valdez, A. C., Wilkowska, W., Ziefle, M., Dorner, S. & Holzinger, A. (2010) From cloud computing to mobile Internet, from user focus to culture and hedonism: The crucible of mobile health care and Wellness applications. 5th International Conference on Pervasive Computing and Applications (ICPCA). IEEE, 38‐45.


Example: The medical world is mobile (Mocomed)

Holzinger, A., Kosec, P., Schwantzer, G., Debevc, M., Hofmann‐Wellenhof, R. & Frühauf, J. 2011. Design and Development of a Mobile Computer Application to Reengineer Workflows in the Hospital and the Methodology to evaluate its Effectiveness. Journal of Biomedical Informatics, 44, 968‐977.


Next Generation Concepts for Medical Informatics

Standardized Medicine

EBM CPG

Person‐alized

Medicine

GBM GPM

Pervasive Healthcare

Preventive Health Integration

Tanaka, H. (2010)


Example: Concept of Omics‐based medicine

Tanaka, H. (2010) Omics‐based Medicine and Systems Pathology A New Perspective for Personalized and Predictive Medicine. Methods of Information In Medicine, 49, 2, 173‐185.

A. Holzinger 444.152 Med Informatics L158/65 Holzinger(2005)

User‐centredSystem‐

centred

Process‐centred

Information Quality as the hiatus theoreticus …

Holzinger, A. & Simonic, K.‐M. (Eds.) (2011) Information Quality in e‐Health. Lecture Notes in Computer Science LNCS 7058, Heidelberg, New York, Springer.


Thank you!


What is the difference between classical medicine and molecular medicine?

What is still the key topic of medical informatics and why? Why was it called “Medical informatics” in 1970 and why is it called

“Biomedical Informatics” today? How did Shortliffe (2011) describe biomedical informatics in

perspective and especially the interrelation between basic and applied science?

Which factors determine the p‐Health Model? What is typical health information, what is typical biomedical

information – what is the difference? What is the synergistic relationship between translational medicine and

biomedical informatics? What is the typical role of a biomedical informatics specialist? How can we augment human capabilities for structuring, retrieving and

managing information? Why is the Von‐Neumann Architecture interesting?

Sample Questions (1/2)


What does Moore’s Law tell you? What is the difference between technological performance

and human performance? What is the idea of “Pervasive Computing”? What does Pervasive Health mean? What are the four most important vital sensor data? What is a smart object? What does personalized medicine mean? What is genomic polymorphism? Is the term “medical free text” correct? What does level of structure mean? What is the difference between data, information and

knowledge? What is still the biggest risk in medical informatics? Why should information quality be important?

Sample Questions (2/2)


Interactive Visualization

Holzinger, A., Kickmeier‐Rust, M. D., Wassertheurer, S. & Hessinger, M. (2009) Learning performance with interactive simulations in medical education: Lessons learned from results of learning complex physiological models with the HAEMOdynamics SIMulator. Computers & Education, 52, 2, 292‐301.


Data for health use …

Holzinger, A., Dorner, S., Födinger, M., Valdez, A. C. & Ziefle, M. (2010) Chances of Increasing Youth Health Awareness through Mobile Wellness Applications. In: Lecture Notes in Computer Science LNCS 6389. Berlin, Heidelberg, Springer, 71‐81.


Example: Cyberinfrastructure for Network Science Center

Yildriim, Muhammed A., Kwan‐II Goh, Michael E. Cusick, Albert‐László Barabási, and Marc Vidal. (2007). Drug‐target Network. Nature Biotechnology 25 no. 10: 1119‐1126.

http://cishell.org/gallery.html

andreas holzinger vo 444.152 medical informatics lecture –version ws science meets...

Documents