Modeling a Fuzzy System for Diagnosis of Syndromes with Integrated Eastern and Western Medicine Using the Importance of Symptoms

In this study, we present an approach to include the importance of symptoms for the diagnosis of syndromes with integrated eastern and western medicine. We also focus on knowledge representation and inference engine of our proposed system using the importance of symptoms. The innovative point of this study is combining the degree of diagnosis of syndrome of Eastern medicine with that of disease of Western medicine when both medicines are associated to a common “disease” name to obtain more accurate diagnosis. Moreover, the importance of symptoms in the inference rules in medical expert systems still has an important role in the diagnosis of syndromes. Based on this approach, the system can adapt more with real clinical practice of integrated eastern and western medicine diagnosis. Finally, examples are provided to demonstrate the advantage of this approach.

Case Base Reasoning (CBR) and Density Based Spatial Clustering Application with Noise (DBSCAN)-based Indexing in Medical Expert Systems

Case-based Reasoning (CBR) has been widely applied in the medical expert systems. CBR has computational time constraints if there are too many old cases on the case base. Cluster analysis can be used as an indexing method to speed up searching in the case retrieval process. This paper propose retrieval method using Density Based Spatial Clustering Application with Noise (DBSCAN) for indexing and cosine similarity for the relevant cluster searching process. Three medical test data, that are malnutrition disease data, heart disease data and thyroid disease data, are used to measure the performance of the proposed method. Comparative tests conducted between DBSCAN and Self-organizing maps (SOM) for the indexing method, as well as between Manhattan distance similarity, Euclidean distance similarity and Minkowski distance similarity for calculating the similarity of cases. The result of testing on malnutrition and heart disease data shows that CBR with cluster-indexing has better accuracy and shorter processing time than non-indexing CBR. In the case of thyroid disease, CBR with cluster-indexing has a better average retrieval time, but the accuracy of non-indexing CBR is better than cluster indexing CBR. Compared to SOM algorithm, DBSCAN algorithm produces better accuracy and faster process to perform clustering and retrieval. Meanwhile, of the three methods of similarity, the Minkowski distance method produces the highest accuracy at the threshold ≥ 90.