domain-specific multi-stage query language for medical document repositories

to Advance Knowledge for Humanityto Advance Knowledge for Humanity

Aastha MadaanUniversity of Aizu

1

Domain Specific Multi-stage Query Language Domain Specific Multi-stage Query Language for Medical Document Repositoriesfor Medical Document Repositories

05/01/23 VLDB Phd Workshop 2013


IntroductionIntroduction Specialized Domains Biomedical, agriculture, medical/healthcare

Insufficient Search Engine-like Keywords based search

Require Effective search and query mechanisms

Medical domain

Medical professionals Specific technical articles (particular topic/sub-topics )General public General information (disease or medicine).

How to retrieve medical information effectively

05/01/23 2VLDB Phd Workshop 2013

Query ”general AIDS information” � medical search tool, such as PubMed Result 1000s of documents Different aspects of AIDS⊆ Such as , treatment, drug therapy, transmission, diagnosis, and history



Introduction (1)Introduction (1) Medical Information

Knowledge Evolved over 10s of yearsContains Well defined terms and processes

Available on the WebPatient Specific InformationKnowledge-based Information

Medical Literature

Web Documents

Patient-encounterRecordings


Complexity Knowledge-based Resources

Heterogeneous End-user groups Patients, researchers, doctors and other experts

Variation Information Requirements Patient-treatment, self -diagnosis, general health information

Structure Medical Documents Scientific papers, encyclopedias and other literature Unique, well-defined

05/01/23 VLDB Phd Workshop 2013 4

Introduction (2) Introduction (2)


Introduction (3): Specialized DocumentsIntroduction (3): Specialized DocumentsCase of medical encyclopedias

Comprehensive medical guide Patients and clinicians

Authoritative source NLM (National Library of Medicine)

Paper based resources Electronic format

Examples MedlinePlus, WebMD, ADAMS, Merriam-Webster Medical Dictionary



Introduction (4): Why QueryIntroduction (4): Why Query External knowledge base Clinicians

Evidence based medicine

During different stages of point-of-care

Patient Assessment plan

Treatment based on Patient diagnosis

Improve Quality of Care

Authoritative information required

Self Diagnosis Patients and their relatives During Early appearance of symptoms/post-checkups Enhance Personal Knowledge



The Underlying StructureThe Underlying Structure


The Hierarchical StructureTopic of the Document

SubtopicsMiscellaneous/Related

Content

Subtopic 1 Subtopic 2 Subtopic n Content topic 1 Content topic 2 Content topic n

Content Content Content Content Content Content

Flow of Contents Organized stages of point-of-care



Introduction (5): End UsersIntroduction (5): End Users Variable

a. Demographical Characteristicsb. Tasks/Purposec. Computer/Domain Expertise

Practitioners and Researchers Well-versed

Domain knowledge and terminologies Require

Precise, complete, accurate and timely results

Patients and their relatives NOT Well-versed

Domain knowledge and terminologies Require

General information

Healthcare Workers

SpecializedResearchers

Patients, their relatives


Evidence-based Queries Intent: Diagnostic Raised by: Clinicians/Experts Target resources: Online Medical Repositories (e.g. medical encyclopedia) Example: “Cases where helicobacter bacteria causes peptic ulcer”

Hypothesis-directed Queries Intent: Non-diagnostic Raised by: Novice users/patients Target resources: Online Medical Repositories (e.g. medical encyclopedia) Example: “Treatment in case of high fever and dizziness”


Medical QueriesMedical Queries


Query FlowsQuery Flows


Occurrence Evidence-based and hypothesis-directed queries Represent Stages of information seeking Comprise Varying levels of query complexity

1. 2. 3. 4.



Query: Find chances of "Cancer Risk" in patients showing symptom "Sleep Deprivation" and have been exposed to "Radiation" (but not "Environmental Toxins" and does not have "Genetic Disorder") .

Help Needed

Research GapResearch Gap

Results Large in number, irrelevantFailure Keyword search, domain-specific search toolsRequire Precise and easy-to-use database style query methods

Key steps:

1. Schema understandable by users 2. Identify Resources to query3. Identify Granularity of results

Healthcare Expert

Paper-based resources



Aim: Query Online Medical Information Effectively

Transform Document Repository User-Level Schema

Enable High-level Query Language

Target Audience Skilled and semi-skilled users

Utilize Query capabilities of a database query language

Facilitate In-depth Queries and Granular Results

Bridging the GapBridging the Gap


Query the New WayQuery the New Way


User-levelUser-levelSchemaSchema

High-level Query High-level Query language language

Traditional Method

Proposed Method

ResourceResource

KeywordSearch

Query Method

Medical Expert

Medical Expert

Results -Lack specificity-Long list of full documents-Trustworthiness of resources unknown

Results-Specific, granular-Segments of documents query criteria-Trustworthy/Authoritative sources only



Proposed ApproachProposed Approach



Key Features Key Features User-Level Schema Offline Process Universal , concept-level schema Attributes

Understandable Domain experts and novice users Provide Granular, context-based results Use Web segmentation algorithm, Domain concepts

Multi-stage Query Language Online Process Map multi-stage diagnostic process Step-by-step Query Flow Interactive Querying Continuous query refinement

View Results Add concept View Results

Support Simple, Medium, Complex, Recursive Queries Use User-level schema



OutlineOutlineTwo-step Framework



Data ModelData Model


Data ModelData ModelTree Structured Repository


H1H1

f1f1

f2f2


Data Model (1)Data Model (1)



Data Model (2): SchemaData Model (2): Schema


Attributes Diagnostic concepts/terms Stages of point-of-care

Do not change frequently


Data Model (3): A XML DocumentData Model (3): A XML Document


Title CausesSymptoms

Treatment

Document corresponding to “Aarskog Syndrome”

MedlinePlus Medical Encyclopedia


Data Model (4): Query EffortData Model (4): Query Effort Query: Find if "Oxygen therapy" work for the treatment of "Chronic

Respiratory Failure" and symptoms are "Lethargy" OR "Shortness of breath”.


Advanced keyword search Proposed MethodSELECT attribute = “Treatment” WHEREAttribute “Disease_name” = “Chronic Respiratory Failure”ANDAttribute “Treatment” = “Oxygen therapy”ANDAttribute “Symptoms” =“Lethargy”OR Attribute “Symptoms” = “Shortness of breath”

Easy-to-UseNot PossibleResult segment

Context of user-query


Data Model (5): Granular ResultsData Model (5): Granular Results


Queried Attributes/Segments Query Results

Context Granular

Each result is a segment, combination of Concept/context in query Item of concern (content enclosed in a segment)


An ExampleAn ExampleQuery: Find other symptoms where “chronic kidney failure” is caused by “anemia”

Queried segment Symptoms

Segments in Query Causes = “anemia” and Disease_name = “Chronic kidney failure”

Result Segment Symptoms

Context disease_name = “chronic kidney failure” & causes = “anemia”




Next Step Next Step Multi-stage Query Language Multi-stage Query Language


Proposed Query Language (1)Proposed Query Language (1) XQBE [Braga, 2005] User level schema

Create queries Drag and drop interface

Query : “Find cases where a person is inflicted with “peptic ulcer” due to “helicobacter pylon bacteria”


Attributes understandable by end users

1. Case = disease_name Value = ??

2. Due to = Causes Value = “helicobacter pylon bacteria”

3. Inflicted with = Symptoms Value = “peptic ulcer”

Query Effort Minimal learning curve Computer-expertise not required


Multi-stage Query-by-Concept Concept Query-able attribute Topic, sub-topic, medical concept

Query Effort Dynamic selection of attributes No computer expertise

Query Process


Proposed Query Language (2)Proposed Query Language (2)

An Example: Cases where fever is caused due to infliction of Pneumonia and

Tuberculosis


Another ExampleAnother Example Query: Find cases where 3 clinical concepts (“cough”, “no

sore-throat”, and “no sterol injection”) occur in context of symptoms along with a sub-concept (“non sterol injection at the left side”)

XQBE on Specialized Medical Repositories

Multi-stage Query-by-concept Query Language



Evaluation Plan Evaluation Plan


Data Sets Document repository

MedlinePlus Health topics (900+) , encyclopedia (4000+), drugs (12000+) Set of Queries

50 test queries (multi-staged) Using literature survey and consultation with medical users

Quantitative Studies Evaluation Metrics

Accuracy of segment extraction (schema creation) Precision and Recall Reduction in search space Query Results

Qualitative Studies Usability Studies

Actual End-users Query Performance


Initial AchievementsInitial Achievements

HTML documents XML schema as per proposed model

XQuery on XML

Integration with XQBE

Query by concept Enumeration using paper and pencil




ChallengesChallenges

Scalability Schema extraction of similar repositories

Understand and Implement Query operations needed

Understand User characteristics to be considered

User Interface Query Language



Related WorkRelated Work Domain-specific Information Retrieval [Yan, 2011]

Similarity and popularity based models Insufficient for domain experts “Information granulation” needs to be considered in huge document repositories

Form-based Query Interfaces [Jayapandian, 2009] Easy-to-use Limited access to the database Complex queries large number of forms Varying medical concepts large number of fields in forms

Beyond single page web search results [Varadarajan, 2008] Provide granular results for user’s search Return segments from multiple or related web documents as results

High-level Graphical Query Languages [Braga, 2005] Easy-to-use and understand Little or no programming effort required by the user Common languages QBE, XQBE


Summary and Conclusions Summary and Conclusions


Proposed Multi-stage Query Language

1. Aim Making Online medical information usable

2. Transformation User-Level Schema

3. Facilitates Granular/Context-based Results

4. Support Healthcare Experts

5. Minimize Learning curve for novice users

6. Reduce Dependency on keyword based searches

Provide Web user level activity Healthcare experts no or little programming effort


References (1)References (1)


[1] D. Braga, A. Campi, and S. Ceri. Xqbe (xquery by example): A visual interface to the standard xml query language. ACM Trans. Database Syst., 30(2):398–443, June 2005.[2] D. Cai, S. Yu, J.-R. Wen, and W.-Y. Ma. Extracting content structure for web pages based on visual representation. In Proceedings of the 5th APWeb, pages 406–417. Springer-Verlag, 2003.[3] M.-A. Cartright, R. W. White, and E. Horvitz. Intentions and attention in exploratory health search. In Proceedings of the 34th Intl. ACM SIGIR conference, pages 65–74, New York, NY, USA, 2011.ACM.[4] S. Cohen, Y. Kanza, Y. Kogan, W. Nutt, Y. Sagiv, and A. Serebrenik. Equix-a search and query language for xml. Journal of the American Society for Information Science and Technology, 53:2002, 2000.[5] S. M. Freire, E. Sundvall, D. Karlsson, and P. Lambrix. Performance of XML Databases for Epidemiological Queries in Archetype-Based EHRs. In Proceedings Scandinavian Conference on Health Informatics 2012 , volume 70 of Linkping ElectronicConference Proceedings, pages 51–57. Linkping University Electronic Press, 2012.[6] M. Gschwandtner, M. Kritz, and C. Boyer. Requirements of the health professional research. In Technical Report D8.1.2. Khresmoi Project, 2011.[7] A. Hanbury. Medical information retrieval, an instance of domain. In SIGIR'12. ACM, August 2012.[8] S. Hunt, J. J. Cimino, and D. E. Koziol. A comparison of clinicians’s access to online knowledge resources using two types of information retrieval applications in an academic hospital setting. J Med Libr Assoc, 101(1):26–31, 2013.[9] http://www.who.int/classifications/icd/en/, 2011.[10] M. Jayapandian and H. V. Jagadish. Automating the design and construction of query forms. ICDE, page 125, 2006.[11] F. Li and H. V. Jagadish. Usability, databases, and hci. IEEE Data Eng. Bull., 35(3):37–45, 2012. [12] http://loinc.org/, 2011.[13] A. Marian and W. Wang. Flexible querying of personal information. IEEE Data Eng. Bull., 32(2):20–27, 2009.[14] http://www.nlm.nih.gov/bsd/pmresources.html, 2011.[15] http://www.nlm.nih.gov/medlineplus/, 2009.[16] http://www.linkedin.com/groups/ Choice-OpenEHR-persistence-layer-144276.S.208531138?qid=208adbca-fc26-4ada-bf02-7efe5a9e5661&trk=group_most_recent_rich-0-b-ttl&goback=%2Egmr_144276, 2013.

http://www.who.int/classifications/icd/en/

http://www.nlm.nih.gov/bsd/pmresources.html

http://www.linkedin.com/groups/


References (2)References (2)[17] http://www.ncbi.nlm.nih.gov/pubmed, 2011.[18] S. A. Rahman, S. Bhalla, and T. Hashimoto. Query-by-object interface for information requirement elicitation in m-commerce. Int. J. Hum. Comput. Interaction, 20(2):135–160, 2006.[19] X. Y. Raymond, Y. Lau, D. Song, X. Li, and J. Ma. Toward a semantic granularity model for domain-specific information retrieval. ACM Trans. On Information Systems., 29(3), July 2011.[20] S. Sachdeva and S. Bhalla. Implementing high-level query language interfaces for archetype-based electronic health records database. In COMAD, 2009.[21] http://www.ihtsdo.org/snomed-ct/, 2011.[22] R. Varadarajan, V. Hristidis, and T. Li. Beyond single-page web search results. IEEE Transactions on Knowledge and Data Engineering, 20(3):411–424, 2008.[23] A. Yasir, M. Kumara Swamy, P. Krishna Reddy, and S. Bhalla. Enhanced query-by-object approach for information requirement elicitation in large databases. In Big Data Analytics, volume 7678 of Lecture Notes in Computer Science, pages 26–41. Springer, 2012.[24] M. Jayapandian, H.V. Jagadish : Automating the Design and Construction of Query Forms. IEEE Trans. Knowl. Data Eng. 21(10) :1389-1402 (2009).



QuestionsQuestions
