Evolutionary Wavelet Neural Network Ensembles for Breast Cancer and Parkinson's Disease Prediction

Abstract

This chapter introduces a novel approach for the prediction of breast cancer and Parkinson's disease. The authors propose an ensemble of E-WNNs to enhance the accuracy and robustness of predictive models. To predict breast cancer from complicated medical data, the E-WNN ensemble uses wavelet transforms in neural networks. The ensemble's networks' structure and attributes are adjusted using evolutionary algorithms to develop a powerful forecasting framework. To predict Parkinson's disease, they employ E-WNN to study clinical assessments and patient history. They fine-tune ensemble members to discover small patterns that reflect disease progression, leading to a more accurate diagnosis. They evaluate the ensemble's performance in terms of classification accuracy, sensitivity, and specificity, highlighting its potential as a valuable tool for early detection and diagnosis of breast cancer and Parkinson's disease. In this study of medical predictive modeling, evolutionary algorithms and wavelet modification transformations are used to make disease prediction systems more accurate and reliable.