AIME'03 Workshop: Model-based and Qualitative Reasoning in Biomedicine

Also in the medical management of disorders in patients, qualitative and model-based approaches are being used. For example, many methods used for diagnosing disease rely on explicit models of normal or abnormal structure and behaviour (often referred to as 'first-principles models') of the underlying disease process. Examples are causal disease models, and physiological models of regulatory mechanisms in the human body. Model-based approaches have the potential of facilitating the development of systems because they can be partially based on models described in the medical literature. Thus, the dependence of a developer on the availability of expert physicians can be lessened. Furthermore, a model-based approach promotes the reuse of models. Finally, model-based systems have the advantage that generated advice to users can be explained in terms of the used model. Qualitative approaches play a role in the modelling of disease and treatment processes, including the handling of the uncertainty involved in these processes.

Techniques used in the construction of qualitative and model-based systems ranges from logical techniques, such as used in model-based diagnosis, qualitative simulation techniques, such as those related to QSIM, to qualitative probabilistic networks which can be used to handle uncertainty in a purely qualitative fashion.

Topics of interest

Model-based reasoning

• Model-based diagnosis using abduction and consistency checking
• Model-based forecasting and prediction
• Model-based treatment selection

Qualitative reasoning

• Causal modelling (qualitative models of cause-effect relationships)
• Qualitative simulation, and function-based modelling

Qualitative representation of uncertainty

• Qualitative probabilistic networks
• Fuzzy qualitative reasoning

Hybrid approaches (mixed qualitative/quantitative models)

• Bayesian networks
• Multi-model approaches

Modelling in biomedicine

• Logical formalisation of biomedical knowledge, clinical guidelines, and so on
• Bayesian modelling (development of Bayesian networks based on biomedical expert knowledge)
• Mathematical modelling and simulation (development of mathematical simulation models of cells, e.g. nerve cells, and organs, e.g. the heart and vessels)

Modelling of biological structures and function

• Models of the structure of and interaction between biomolecules
• Models of genetic regulation and other biological mechanisms
• Qualitative simulation of human (patho)physiology, and use of control theory

Final Scientific Programme

---

Session I: 9:45 - 10:45

• [09:45 - 10:00] Peter Lucas - Welcome
• [10:00 - 10:45] Invited Talk - Ivan Bratko: Automated modelling with qualitative representations

---

10:45 - 11:00 Coffee Break

---

Session II: 11:00 - 12:30

• [11:00 - 11:30] Riccardo Bellazi, Raffaella Guglielmann, and Liliana Ironi: Qualitative reasoning for modeling metabolic systems: a case study
• [11:30 - 12:00] Liliana Ironi and Stefania Tentoni: Challenging QR research issues in electrocardiology
• [12:00 - 12:30] Elisa Fromont, Marie-Odile Cordier, Rene Quiniou, and Alfredo Hernandez: Kardio and Calicot: a comparison of two cardiac arrhythmia classifiers

---

12:30 - 14:00 Lunch

---

Session III: 14:00 - 14:45

• [14:00 - 14:45] Invited Talk - Linda van der Gaag: Model-based reasoning with qualitative probabilistic networks

---

14:45 - 15:00 Break

---

Session IV: 15:00 - 16:00

• [15:00 - 15:30] Robert Trelease and Jack Park: Qualitative reasoning in integrative biology: evolving a network experiment modeling system for collaborative multilevel systems research
• [15:30 - 16:00] Jung-Wook Bang, Raphael Chaleil, and Jon Williamson:Two-stage Bayesian network for metabolic network prediction

---

16:00 - 16:30 Refreshments Break

---

Session V: 16:30 - 17:45

• [16:30 - 17:00] Anand Kumar, Barry Smith, Mario Stefanelli, Silvana Quaglini, and Matteo Piazza: Implementing clinical guidelines in an organizational setup
• [17:00 - 17:30] Agnieszka Onisko and Marek J. Druzdzel: Effect of imprecision in probabilities on Bayesian network models: an empirical study
• [17:30 - 17:45] Closing

---

Registration

Workshop price is only 50 Euros; workshop participants must also register for the main AIME'03 conference.

Workshop organisation

Chair

Peter Lucas, University of Nijmegen, Nijmegen, The Netherlands

Programme committee

Klaus-Peter Adlassnig, Austria
Pedro Barahona, Portugal
Ivan Bratko, Slovenia
George Coghill, UK
Enrico Coiera, Australia
Carlo Combi, Italy
Marie-Odile Cordier, France
Marek Druzdzel, USA
John Fox, UK
Raffaella Guglielmann, Italy
Werner Horn, Austria
Liliana Ironi, Italy
Hidde de Jong, France
Elpida Keravnou, Cyprus
Rudibert King, Germany
Benjamin Kuipers, USA
Casimir Kulikowski, USA
Peter Lucas, Netherlands
Simon Parsons, USA
Rene Quiniou, France
Silja Renooij, Netherlands
Steffen Schulze-Kremer, Germany
Yuval Shahar, Israel
Robert Trelease, USA
Stefania Tentoni, Italy

For more information about the workshop please contact the chair

---

Peter Lucas  Institute for Computing and Information Sciences University of Nijmegen Toernooiveld 1 6525 ED Nijmegen, The Netherlands Tel: +31 24 365 2611/3456; Fax: +31 24 3653366 peterl@cs.kun.nl

Peter Lucas | AIME'03 | AIME'03 Workshop