Abstract
People often concern the relationships between symptoms and diseases when seeking medical advices. In this paper, medical data are divided into three copies, records related to main disease categories, records related to subclass disease types, and records of specific diseases firstly; then two disease recognition methods only based on symptoms for the main disease category identification, subclass disease type identification, and specific disease identification are given. In the methods, a neural network and a support vector machine (SVM) algorithms are adopted, respectively. In the method validation part, accuracy of the two diagnosis methods is tested and compared. Results show that automatic disease prediction only based on symptoms is possible for intelligent medical triage and common disease diagnosis.
Highlights
There is shortage of per capita medical resources, and high-quality medical resources are concentrated in large cities and large hospitals
In this paper, such diagnosis methods are proposed. rough these methods, preliminary diagnoses can be provided for specialized diseases, and it can help medical workers in areas having underdeveloped medical resources implement medical triage and provide consultation services for people who will seek precise treatment in big hospitals
A corollary is that the accuracy of this kind of diagnosis methods is limited by the problem itself, and even another practicable machine learning method is adopted, and the performance is similar with the neutral network and support vector machine (SVM) method
Summary
There is shortage of per capita medical resources, and high-quality medical resources are concentrated in large cities and large hospitals. In China, many patients have strong health awareness, even if their symptoms are not serious; they flock to large hospitals to seek quality medical services. An intelligent information system, which can automatically perform prediagnosis based on the symptoms provided by patients, can alleviate the problem of medical resource shortage. In this paper, such diagnosis methods are proposed. Rough these methods, preliminary diagnoses can be provided for specialized diseases, and it can help medical workers in areas having underdeveloped medical resources implement medical triage and provide consultation services for people who will seek precise treatment in big hospitals. A common disease diagnosis service can be realized for people who can seek medical treatment by themselves
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
