Using health data repositories for developing clinical system software: a multi‐objective fuzzy genetic approach

Abstract

Evolution of technology has brought a revolution in various fields of sciences and amongst them, healthcare is one of the most critical and sensitive areas because of its connection with common masses' quality of life. The notion of integrating the healthcare system with the latest data repositories is to make disease prediction efficient, transparent, and reusable. Due to data heterogeneity, data repositories along with optimum classifiers help stakeholders to predict the disease more accurately without compromising the interpretability. Evolutionary algorithms have shown great efficacy, accuracy, and interpretability in improving disease prediction for several datasets. However, the quest for the best classifier is still in evolution. In this research, a state-of-the-art medical data repository has been developed to give researchers of medical domain great ease of use in utilizing different datasets governed by a multi-objective evolutionary algorithm using fuzzy genetics. The proposed model called 'MEAF' is evaluated on various public repositories. A subset of these repositories includes breast cancer, heart, diabetes, liver, and hepatitis datasets. The results have been analyzed, which show competitive accuracy, sensitivity, and interpretability as compared to relevant research. A customised software application named 'MediHealth' is developed to supplement the proposed model that will facilitate the domain users.

Disease Classification in Health Care Systems With Game Theory Approach

“…The graphical comparison of the proposed model highlights the phenomena of improved accuracy, interpretability, sensitivity and specificity as compared to its counterpart. Our technique achieved 98.97% accuracy with Wisconsin Breast Cancer Dataset, 83.52% accuracy with PIMA Indian Diabetes Dataset and 91.1% accuracy with Cleveland Heart Disease Dataset depicted in Figure 4, 5 and 6 respectively [33]. The improvement in the accuracy is due to the diversity in the selection process of the parent population by ingesting the penalization mechanism and retaining the offspring population in an archive to cover the complete solution space.…”