Case-Based Reasoning Approach For Thalassaemia Diagnosis

This thesis proposes a Case-based Reasoning model for medical diagnosis, particularly for Thalassaemia diagnosis. The model is designed and a prototype is developed to test
the diagnosis accuracy of the model. Two local similarity combined with two global similarity is evaluated to identify the suitable similarity function for the model. The
appropriate K value for Nearest Neighbour also identified for this model. Finally, the testing done using leave one out method on Thalassaemia cases from Alor Star General
Hospital and the testing demonstrates 88% of diagnosis accuracy. The results show that case-based reasoning model has a great potential to be implemented in diagnosing thalassaemia cases.