Breast Cancer Prediction using Deep Network Model with Multi-Modal Data Fusion

Breast cancer is an illness that significantly influences women’s health. It is regarded as the leading cancer mortality. Accurate identification and pharmacological intervention are critical for lowering the mortality rate from breast cancer. In the latest days, deep learning methods have been widely used for reliable illness detection, with multi-model fusion being one of the most commonly used.Nevertheless, decision trees with lower categorization quality and superior similarities are created during training process affecting the model’s overall classification efficiency. A hierarchical multi-model is made in this research. The hierarchical clustering approach applies the concept to decision trees by assessing the resemblance among all the decision trees.Sample trees are chosen from separated groups to build the hierarchical randomization with lesser commonality and good accuracy.This report presents the use of the Wisconsin Diagnosis Breast Cancer (WDBC) database and samples from UCIrepository. The suggested method’s performance was measured using accuracy, precision, sensitivity, selectivity, and AUC (Area Under ROC Curve).The experimental results demonstrate that the segmentation depends on the multi-modelas a subset of features approach achieves the best accuracy of 97% and 97% when compared to others. This study’s strategy is a valuable tool for identifying prostate cancer.