Deep Learning with Backtracking Search Optimization Based Skin Lesion Diagnosis Model

Nowadays, quality improvement and increased accessibility to patient data, at a reasonable cost, are highly challenging tasks in healthcare sector. Internet of Things (IoT) Cloud Computing (CC) architectures utilized the development smart systems. These entities can support real-time applications by exploiting massive volumes produced wearable sensor devices. The advent evolutionary computation algorithms Deep Learning (DL) models has gained significant attention diagnosis, especially decision making process. Skin cancer is deadliest disease which affects people across globe. Automatic skin lesion classification model important application due its fine-grained variability presence lesions. current research article presents new diagnosis i.e., with Evolutionary Algorithm based Image Segmentation (DL-EAIS) for IoT cloud-based environments. Primarily, dermoscopic images captured using devices, then transmitted cloud servers further diagnosis. Besides, Backtracking Search optimization (BSA) Entropy-Based Thresholding (EBT) BSA-EBT technique applied image segmentation. Followed by, Shallow Convolutional Neural Network (SCNN) as feature extractor. In addition, Deep-Kernel Extreme Machine (D-KELM) employed determine class labels images. An extensive set simulations was conducted validate performance presented method benchmark dataset. experimental outcome infers that proposed demonstrated optimal over compared techniques under diverse measures.