Bayesian network model for diagnosis of Social Anxiety Disorder

Abstract

Nowadays a lot of systems are developed to predict or suggest a diagnosis about the health level of a patient for helping physicians in their decisional process. Recent researches prove that decisional systems implemented by Bayesian networks represent an efficient tool for medical healthcare practitioners. Bayesian Networks (BNs) are graphical models with significant capabilities that can be used for medical predictions and diagnosis. Social Anxiety Disorder (SAD) is the third most common psychiatric disorder in America behind depression and alcohol abuse. This paper focuses on the use of Bayesian network in assisting SAD diagnosis, in which SAD is analyzed and modeled by Bayesian networks in two phases: construction of BN structure and Conditional Probability Tables (CPTs). This research provides a Bayesian network-based analysis of data, gathered from a number of engineering students.