Breast Cancer Detection in the IoT Cloud-based Healthcare Environment Using Fuzzy Cluster Segmentation and SVM Classifier

Early-stage detection of breast cancer disease and its accurate diagnosis have been always challenging for the healthcare professional. An IoT healthcare system can play a vital role in this field. The existing diagnosis technique does not efficiently identify breast cancer in the beginning phases, and many of the women struggled from such a deadly illness. In this research paper, we introduce IoT cloud-based predictive analytics mainly based on fuzzy cluster-focused augmentation and optimal SVM classification for forecasting breast cancer infection via regular inspection and enhancing the health services by giving healthcare guidelines. In the proposed model, the fuzzy clustering algorithm is being used for efficient image segmentation that mainly focused on transition region filtration. Besides that, fuzzy C-means clustering and optimal SVM are also applied to characterize the transition period region attributes and feature extraction. The experimental phase is divided into three parts: parameter optimization testing, feature selection, as well as optimal SVM. The experimental phase results reveal that the proposed enhanced SVM method performed outstandingly in terms of feature selection, precision, TPR, FPR, F1-score over existing machine learning methods, i.e., logistic regression, decision tree, and SVM method.