Related Experiment Videos
Modelling of a case-based retrieval system for oncology
Delphine Rossille1, Jean-François Laurent, Anita Burgun
1Laboratoire d'Informatique Médicale, Université de Rennes 1, France. rossille.22006017@etudiant.univ-rennes1.fr
Studies in Health Technology and Informatics
|December 11, 2003
Summary
This study introduces a novel case-based reasoning decision-support system to aid oncologists. The system retrieves similar patient cases to guide treatment decisions when patients do not comply with standard guidelines.
Area of Science:
- Oncology
- Medical Informatics
- Artificial Intelligence in Medicine
Background:
- Clinical decision-making relies on past similar patient cases when guidelines are not followed.
- Existing patient records are not optimized for direct use in decision support systems.
Purpose of the Study:
- To propose a case-based reasoning (CBR) decision-support system for oncology.
- To automatically compare new patient cases against guidelines and historical data.
- To retrieve and present similar cases to aid therapeutic decisions.
Main Methods:
- Development of an object-oriented model for patient records, incorporating prognosis factors and illness episodes.
- Designing a system to compare new cases with structured guidelines and a database of similar cases.
- Utilizing case-based reasoning for decision support in oncology.
Main Results:
- A novel system structure for a CBR decision-support system in oncology is presented.
- An object-oriented model is proposed to structure patient data for improved usability.
- The system aims to bridge the gap between unstructured patient data and clinical decision-making.
Conclusions:
- The proposed system offers a unique approach to clinical decision support in oncology.
- The system can be viewed as a specialized data warehouse for oncology cases.
- Future work includes refining similarity measures and database integration.