Conference Program

3rd International Workshop on Knowledge Representation for Health Care (KR4HC 2011)
July 6, 2011. Golf Hotel. Bled, Slovenia.

Session 1: Health Processes (chair: Silvia Miksch)
» David Riaño:
A Systematic Analysis of Medical Decisions. How to Store Knowledge and Experience in Decision Tables.
» Juan Fernández-Olivares, Inmaculada Sanchez-Garzón, Arturo González Ferrer, Juan A. Cózar, Ana Fernández-Teijeiro, Rodrigo Cabello and Luis Castillo:
Task Network based modeling, dynamic generation and adaptive execution of patient-tailored treatment plans based on Smart Process Management technologies
» Kathrin Dentler, Annette Ten Teije, Ronald Cornet, Nicolette De Keizer:
Towards the Automated Derivation of Quality Indicators
Coffe Break
Session 2: Patient Records,  Medical Costs, and Clinical Trials (chair: David Riaño)
» Jose M. Juarez, Manuel Campos, Antonio Gomariz and Antonio Morales:
Computing Problem Oriented Medical Records: Problem Flow Model
» Joan Albert López-Vallverdú, David Riaño and Antoni Collado:
Detecting Dominant Alternative Interventions to Reduce Treatment Costs
» Krystyna Milian, Annette Ten Teije, Anca Bucur and Frank van Harmelen:
Patterns of clinical trial eligibility criteria
Invited Talk (chair: Silvia Miksch) - Prof. Yuval Shahar:
The Human Cli-Knome Project: Building a Universal, Formal, Procedural and Declarative Clinical Knowledge Base for the Automation of Therapy and Research
Poster Session (Tea Break) (chair: Annette ten Teije)
1. Iliya Fridman, Erez Shalom and Yuval Shahar:
Evaluation of guideline-application engines by longitudinal simulation: A Position Paper
2. Sarah Lim Choi Keung, Lei Zhao, Ire Ogunsina, Theodoros Arvanitis and Edward Tyler:
Vocabulary Controlled Linkage of Electronic Health Record Data
3. Nicolas Griffon, Saoussen Sakji, Ahmed-Diouf Dirieh Dibad, Julien Grosjean, Philippe Massari and Stefan Darmoni:
A Model for Information Retrieval in Electronic Health Records
4. Hajar Kashfi and Jairo Jr. Robledo:
Towards a Case-Based Reasoning Method for an openEHR-Based Clinical Decision Support System for Dry Mouth
5. Paul Taylor and Igor Toujilov:
Mammographic Knowledge Representation in Description Logic
6. Dennis Wegener, Alberto Anguita and Stefan Rueping:
Enabling the reuse of data mining processes in healthcare by integrating data semantics
7. Raphael Bahati, Stacey Guy and Femida Gwadry-Sridhar:
Analysis of Treatment Compliance of Patients with Diabetes
8. Katharina Kaiser and Andreas Seyfang:
Supporting Knowledge Modelling by Multi-modal Learning
Session 3: Clinical Guidelines (chair: Annette ten Teije)
» Mor Peleg, Samson Tu, Giorgio Leonardi, Silvana Quaglini, Paola Russo, Giovanni Palladini and Giampaolo Merlini:
Reasoning with Effects of Clinical Guideline Actions using OWL for the management of AL amyloidosis
» Erez Shalom, Iliya Fridman, Yuval Shahar, Avner Hatsek and Eitan Lunenfeld:
Towards a realistic clinical-guidelines application framework: Desiderata, Applications, and lessons learned
» Rodrigo Bonacin, Cédric Pruski and Marcos Da Silveira:
Careflow Personalization Services: Concepts and Tool for the Evaluation of Computer-Interpretable Guidelines
» Reinhard Hatko, Joachim Baumeister, Volker Belli and Frank Puppe:
DiaFlux: A Graphical Language for Computer-Interpretable Guidelines

1. PRESENTATIONS are expected 20 min presentation +  5 min for questions.
2. POSTERS are expected
to exceed 90 cm wide by 180 cm tall.
3. Both PRESENTATION and POSTER papers will appear in the AIME workshop proceedings.

Invited Talk

Yuval Shahar, M.D., Ph.D.

Yuval Shahar is a professor and previous chair of Ben Gurion University (BGU)’s Information Systems Engineering department. He holds a B.Sc. and an M.D. degree from the Hebrew University, an M.Sc. in computer science from Yale University, and a Ph.D. in Medical Information Sciences from Stanford University.  After a decade at Stanford as a researcher and full-time faculty member, he has joined BGU to found and head its Medical Informatics Research Center.      

Prof. Shahar’s research focuses on temporal reasoning, temporal data mining, therapy planning, decision analysis, information visualization, knowledge acquisition, knowledge representation, and knowledge-based systems, mostly in biomedical domains.  He is a past chair of the AI in Medicine (AIME) conference (2009) and a many-time scientific program committee member of the American Medical Informatics Association’s (AMIA) Fall Symposium.  Among multiple awards, Prof. Shahar won an NIH 5-year career award and an NSF award to explore the theoretical and practical implications of the temporal-reasoning methodology he had developed, an IBM Faculty Award, and an HP Worldwide Innovation Program award. He was elected in 2005 as an International Fellow of the American College of Medical Informatics (ACMI).

Prof. Shahar serves on the editorial board of the Journal of Biomedical Informatics, Artificial Intelligence in Medicine, Methods of Information in Medicine, and Applied Ontology.

The Human Cli-Knome Project: Building a Universal, Formal, Procedural and Declarative Clinical Knowledge Base for the Automation of Therapy and Research

Clinical Guidelines are a major tool in improving the quality of medical care.  However, currently most guidelines are in free text, not in a formal, executable format, and are not easily accessible to clinicians at the point of care.  Automated support to the application of guidelines at the point of care has a significant potential for assessing and improving the quality of care, and for reducing its costs, especially in the case of management of chronic patients. Furthermore, the required knowledge consists not only of the guideline’s procedural knowledge, such as a flowchart, but also its underlying declarative knowledge, such as context-sensitive interpretations of longitudinal patterns of raw clinical data accumulating from several sources. 

Thus, a major grand challenge for medical informatics is the creation of a distributed, universal, formal, sharable, reusable, and computationally accessible medical knowledge base. Much of the collaboration among areas such as clinical medicine, clinical research, public health, and medical informatics, can and should be distilled into a set of formal representations of declarative and procedural knowledge that would be stored in a universally accessible (to humans and machines) knowledge base. Furthermore, the declarative medical knowledge base can also be exploited to support the tasks of interpretation, summarization, visualization, explanation, temporal data mining, and interactive exploration, so as to discover further clinical knowledge from the time-oriented raw data of patient populations and from the multiple levels of higher-level concepts and patterns that can be abstracted from these data. One might refer to such a continuously changing library of formal declarative and procedural clinical knowledge “the Human Clinome Project”, in homage to the Human Genome project - or perhaps “The Human Cli-Knowme project,” since it would encompass all currently known declarative and procedural human knowledge that can be represented and accessed by computational means.

One example of an effort striving towards this objective over the past decade is the Digital Electronic Guideline Library (DeGeL), a Web-based, modular, distributed architecture that facilitates gradual conversion of clinical guidelines from text to a formal representation in chosen target guideline ontology. The architecture supports guideline classification, semantic markup, context-sensitive search, browsing, run-time application to a specific patient at the point of care, and retrospective quality assessment by a clinical organization. The DeGeL procedural-knowledge hybrid meta-ontology includes elements common to all guideline ontologies, such as semantic classification and domain knowledge; it also includes four content-representation formats: free text, semi-structured text, semi-formal representation, and a formal representation.  These formats support increasingly sophisticated computational tasks. The procedural knowledge of each guideline is represented using multiple specific guideline ontologies, such as Asbru. The declarative knowledge in the DeGeL library is represented using the knowledge-based temporal-abstraction (KBTA) ontology. The DeGeL library uses a role-based knowledge-modification authorization model to determine which operations can be applied to the knowledge base, and by whom. The accuracy and completeness of the DeGeL process and tools for specification of both types of knowledge by clinical editors and knowledge engineers were evaluated in several rigorous studies, with encouraging results.

Thus, I argue that building a distributed, multiple-ontology architecture that caters for the full life cycle of a significant portion of current clinical guidelines has become a feasible task for a joint, coordinated, international effort involving clinicians and medical informaticians.