Title :
Clinical Decision Support System for heart diseases using Extended sub tree
Author :
Ravindranath, Kulkarni Rashmi
Author_Institution :
Dept. of Comput. Eng., P.E.S. Modern Coll. of Eng., Pune, India
Abstract :
Clinical Decision Support System (CDSS) is a tool which helps doctors to make better and uniform decisions. There are many existing systems present which are used for diagnosing the diseases. For different types of diseases the existing CDSS systems changes with different algorithmic approaches. Every approach has its pros and cons. Selecting the positive aspect and overcoming the problems is the main motive This paper focuses on comparative study of existing CDSS systems namely Mycin, DeDombal, Quick Medical Record (QMR), Internist 1. Also the paper focuses on different algorithmic approaches for CDSS. It also give comparative study for algorithmic approaches of heart diseases. The proposed system deals with the similarity matching function. In decision tree construction, the nodes are constructed on splitting attribute or the flag value. Hence if continuous value is to be handled then it can prove fatal. This kind of flaw is observed in ID3 algorithm. Hence to overcome this drawback Extended sub tree approach is implemented. The results show the comparative analysis between these two approaches.
Keywords :
decision support systems; decision trees; diseases; medical information systems; patient diagnosis; CDSS; DeDombal; ID3 algorithm; Internist 1; Mycin; QMR; algorithmic approaches; clinical decision support system; decision tree construction; disease diagnosis; extended sub tree; flag value; heart diseases; quick medical record; similarity matching function; splitting attribute; uniform decisions; Accuracy; Algorithm design and analysis; Data mining; Decision support systems; Decision trees; Diseases; Heart; Clinical Decision Support System (CDSS); Decision Tree; Electronic Medical Record (EMR); Extended Sub tree (EST); Patient health Information (PHI);
Conference_Titel :
Pervasive Computing (ICPC), 2015 International Conference on
Conference_Location :
Pune
DOI :
10.1109/PERVASIVE.2015.7087026