DETDO: double exponential smoothing-Tasmanian devil optimization-based data routing algorithm for breast cancer detection

The Internet of Things (IoT) within healthcare systems is presently a feasible choice for delivering enhanced medical care within modern e-healthcare. Through the utilization of an IoT-based intelligent healthcare system, DL strategies are currently being extensively exploited for the detection of cancer. Breast cancer exerts a significant influence on women worldwide. Various computer-aided diagnosis (CAD) strategies have been generated to identify and categorize breast cancer, as recognition of breast cancer images by radiologists is inherently intricate and time-consuming. This research intends to develop a method for breast cancer identification in an IoT platform using deep learning. After the simulation of IoT, the nodes transfer the cancer images to the destination using Double Exponential Smoothing-Tasmanian Devil Optimization (DETDO), which is presented via integrating Double Exponential Smoothing (DES) and Tasmanian Devil Optimization (TDO). Pre-processing is executed by employing resizing and contrast enhancement, then data augmentation is done, and the segmentation process is executed using MIScnn. At last, for breast cancer identification, LeNet-ZFNet trained using proposed DETDO, is employed. Experimental analysis affirmed that the presented approach gained elevated results in an accuracy of 96.8%, sensitivity of 98.5% and specificity of 95.6%.