Predictive Analysis of Diseases Using Bayesian Inference and Markov Chain Monte Carlo Methods

Abstract

Predictive Analysis of Diseases Using Bayesian Inference and Markov Chain Monte Carlo Methods” focuses on applying Bayesian inference and Markov Chain Monte Carlo to make prediction and analyzation of certain disease, including a general case study and Alzheimer’s Disease. The first application shows how, in the event that a patient receives a false diagnosis, the Bayesian inference can be used to test the likelihood that the patient has a condition. This example demonstrates how Bayesian inference can lead to a closer objective reality by iteratively modifying the prior probability to the posterior probability depending on new information. In the second application on Alzheimer’s disease, the regression model parameters (intercept and slope) are estimated using the Gibbs sampling approach and two datasets including data on Alzheimer’s patients. The final model, tested on a separate dataset, achieved an accuracy of 62%. The study demonstrates the potential of Bayesian and MCMC approaches in disease prediction, suggesting a pathway to more robust models in medical analytics.