The Bayes Clinical Statistical Accuracy Optimization Using Expectation-Maximization Iteration

Bayes classifier employs a statistical model to categorize data. It is regarded as the most effective classifier when its probabilistic models are able to extract sufficient information from reality. Nevertheless, it is unfeasible to obtain an adequate quantity of data to construct a precise statistical model in certain significant instances. The proposed algorithm aims to enhance the precision of the Bayes statistical model through the incorporation of the Expectation-Maximization (EM) algorithm. The simulation results indicate that the accuracy of the Bayes probabilistic model can be enhanced by approximately 10% under specific conditions. Furthermore, our findings revealed that each medical feature contributes a unique increment to the statistical model.