Performance Evaluation of DQN, DDQN and Dueling DQN in Heart Disease Prediction

Abstract

Heart disease remains a significant global health concern, it is necessary to develop highly accurate diagnostic system. In recent years, reinforcement learning algorithms have gained attention in addressing medical classification problem. The proposed system is a comparative study of three deep reinforcement learning algorithms, namely Deep Q-Network (DQN), Double Deep Q-Network (DDQN) and Dueling Deep Q-Network, for the classification of heart disease. We examined the efficiency of the proposed system using the dataset of patient information and compare their accuracy, speed and ability.This includes the multi agent model for disease classification and it is widely used in robotics, economics and telecommunication. Our findings shows the effectiveness of deep reinforcement learning approaches with multi agent in heart disease classification.