A Case-Based Intelligent Diagnosis Support System

With the rapid development of medicine and the growing number of subfields,experts focus on their own specialized subfields for deep study of expertise,which also leads to knowledge narrowness at the same time.During the diagnosis of complicated diseases such as malignant tumors,medical institutions gather experts of various disciplines to jointly formulate the perfect treatment plans for patients.It not only provides patients with better quality of medical services,but also promotes the expertise communication of doctors among different disciplines,which has a positive impact on the development of medicine.However,multi-disciplinary treatment requires a lot of manpower and material resources,and due to the lack of standard electronic system for recording the inexplicit knowledge of the diagnosis,it is hard to provide help for future diagnosis of similar cases.To solve these problems,this paper attempts to apply the machine learning technologies to the medical field.Based on a high-quality database of breast cancer cases developed before,we develop an intelligent diagnosis support system to recommend treatment plans accurately and efficiently.In order to optimize the performance of the system and to provide patients with a better treatment plan,we look deep into the data and recommendation results,and propose a novel algorithm to merge recommendation results of different methods.Furthermore,we propose an adaptive recommendation algorithm in this paper,which can work in the context of concept drift.