الفهرس | Only 14 pages are availabe for public view |
Abstract The impor estimated simply because there seems to be a continuing advancement in the complexity and severity of many diagnosed medical maladies. Today, even the slightest inclination that there are symptoms, however minor, are already treated to a barrage of diagnostic medical steps to try to determine if there is an underlying problem that needs a more targeted medical solution. Doctors go to great lengths to try to discover the potential causes behind each symptom so that patients are always aware of their medical conditions at any given time. The medical field is the scientific discipline that deals with finding cure for every conceivable type of illness and disease so this work uses case based reasoning in medical field to help doctors diagnose diseases, to find the appropriate treatment for the patient and to analyze causes and/or treatments. Medical Diagnoses Decision Support system have been proposed to improve the quality of health care services. Knowledge-based systems, compared to conventional data-base systems, are talented to support medical diagnoses to be more accurate and efficient. However, knowledge acquisition is usually a bottleneck in the process of developing such systems. In machine learning, ambiguities rise when the machine tries to understand human language generating uncertainty in the inferencing process; the uncertainty is elucidated by using Semantic technology. One possibility for acquiring medical knowledge, particularly tacit knowledge, is to use data or historical cases in both syntactic and semantic ways. A case III based reasoning (CBR) system is a combination of processes and previous experiences. This thesis proposes a semantic medical CBR, called MedSDrive, as a decision support system that drives medical personal to take better decisions regarding health services from diagnosing to curing. The semantic reasoning mechanism allows deeper understanding of the medical knowledge for more accurate selection and matching of prediagnosed and verified cases and help in fixing most of the traditional CBR problems and limitations. The goal is to retrieve the most suitable medicines for new cases (patients) depending on an ontology based structure for diseases which is adapted automatically to cover all current and future diseases. The ontology amendment is an enhancement to CBR retrieval and update phases. In this work, we examine two object-oriented ontology based CBR frameworks jCOLIBRI and myCBR to compare the similarity output for same input query. During the implementation of the MedSDrive system using jCOLIBRI we found average similarity is higher than average similarity when using MedSDrive system with myCBR which give 99.3% and 93.8 respectively. |