CASE-BASED LEARNING OF PLANS AND GOAL STATES IN MEDICAL DIAGNOSIS

We introduce a case-based system, BOLERO; that learns both plans and g oal states. The major aim is that of improving the performance of a ru le-based diagnosis system by adapting its behavior using the most rece nt information available about a patient. On the one hand BOLERO gets knowledge from cases in the form of diagnostic plans that are represen ted as sequences of decision steps. The advantages of this representat ion include: (1) retrieval and adaptation of parts of plans (steps) ap propriate to the current problem state; (2) generation of new plans no t previously available in memory; and (3) learning from experience, bo th from successful or failed plans. On the other hand, since goal stat es are sets of final diagnosis likelihoods they are not known beforeha nd, i.e. goal states are not defined and the system has to learn to re cognize them. For this reason BOLERO has a case-based method that uses solutions of past cases to recognize a diagnostic state as a goal sta te of a new planning problem. BOLERO and a rule-based system are integ rated into a meta-level architecture in which we emphasize the collabo ration of both systems in solving problems. The rule-based system exec utes the plans generated by BOLERO. As a consequence of the execution of plans, the rule-based system furnishes BOLERO with new information with which BOLERO can generate a new plan to adapt the reasoning proce ss of the rule-based system into correspondence with the recent availa ble data. All the methods have been designed to be useful for medical diagnosis and have been tested in the domain of diagnosing pneumonia.