A system for model-based computer-aided diagnosis and therapy.
A system for model-based computer-aided diagnosis and therapy.
Design considerations of a database system in a clinical network environment
AFIPS '76 Proceedings of the June 7-10, 1976, national computer conference and exposition
| Hi-index | 0.00 |
Priming refers to the stocking of a knowledge base with inference rules derived from expert physicians. One approach to priming is demonstrated by KAMM, a program that accepts rules from an expert at a CRT, and integrates them into the systemic memory of a medical consultation system called MEDICO. Decomposition of inference rules, and storage in a relational database, provides for great flexibility. Easy reorganization of the knowledge base is facilitated by RAIN, a Relational Algebraic INterpreter which is available both to the designer and to KAMM. Verification of systems of inference rules, i.e., inference nets, is performed while the rules are still in decomposed form in the relational database. Systemic memory is generated subsequently by KAMM, from the verified network implicit in the relational database. A hashed index on an encyclopedia of propositions helps speed MEDICO's access to groups of inference rules during consultation sessions.