Abstract:
Case-based reasoning (CBR) has become a
successful technique for developing medical diagnosis systems.
CBR is an approach to knowledge-based problem solving and
decision support. This paper presents CBR methodology and the
technical aspects of implementing a medical diagnosis system.
As a case study, Dengue Haemorrhagic Fever (DHF) Diagnosis
system is implemented by using the basic idea behind CBR and
classifies the severity of DHF. Nearest Neighbor techniques is
applied for case retrieval. If the result of input case is not
satisfied, the adaptation process is done. For the adaptation
process, the retrieved solution is adapted by adaption rules to
meet the input case’s constraints. This system has been tested
with DHF patient cases and overall system accuracy reaches
96%.
