An Intelligent Approach for Accurate Prediction of Chronic Diseases

Abstract

Around the globe, chronic diseases pose a serious hazard to healthcare communities. The majority of the deaths are due to chronic diseases, and it causes burdens across the world. Through analyzing healthcare data and extracting patterns healthcare administrators, victims, and healthcare communities will get an advantage if the diseases are early predicted. The majority of the existing works focused on increasing the accuracy of the techniques but didn’t concentrate on other performance measures. Thus, the proposed work improves the early detection of chronic disease and safeguards the lives of the patients by increasing the specificity and sensitivity of the classifiers along with the accuracy. The proposed work used a hybrid optimization algorithm called the Hybrid Gravitational Search algorithm and Particle Swarm Optimization algorithm (HGSAPSO) to upgrade the detection of chronic diseases. Existing classifier parameters with their optimized parameters are compared and evaluated. Classifiers such as Artificial Neural Network (ANN), Support Vector Machines (SVM), K-Nearest Neighbor (Knn), and Decision tree (DT) are used. Health care data are obtained from the UCI machine learning repository to evaluate the proposed work. The proposed work is assessed on 6 benchmark datasets and the performance metrics such as Accuracy, Specificity, Sensitivity, F-measure, Recall, and Precision are compared. The experimental results exhibit that the proposed work attains better accuracy on Artificial Neural Network-Hybrid Gravitational Search algorithm and Particle Swarm Optimization algorithm (ANN-HGSAPSO) classifier compared to other classifiers. ANN-HGSAPSO provides 93% accuracy for Chronic Kidney Disease (CKD), Cardio Vascular Disease (CVD) 96%, Diabetes 82%, Hepatitis 94%, Wisconsin Breast Cancer (WBC) 91%, and for Liver disease dataset 96%.