Literature Review On Metaheuristics Techniques In The Health Care Industry

Abstract

In recent times, machine learning has provided increasingly satisfying results in the field of medicine, providing results with very high accuracy while helping to reduce costs and diagnose the disease in real time. To achieve this, it is necessary to develop different deep machine learning techniques. Some of these are metaheuristic techniques that offer practical solutions for different types of chronic diseases. These types of algorithms have received the most attention in solving optimization problems. Therefore, this paper presents a wide review of the literature for solving the problems of feature selection using metaheuristic algorithms and selecting those that have had the highest performance compared to the results given by other algorithms. In this paper, a study of 71 articles from a research database was carried out, from which metaheuristic algorithms were analyzed and evidenced on the optimization and selection of features for the prediction of chronic diseases using numerical, binary, or even imaging data. The efficiency of the algorithms is measured based on the accuracy results, error rate, F-means, or other parameters or graphical representations found in this study. This work will help researchers to improve any of the methods, hybridize them, or even build applications for predicting diseases in the future. Gaps in this field have also been identified, and future studies should be conducted.