Automated Healthcare Medical Imaging through EOA Optimized Hyperparameter in CNN

Medical imaging (MI) is the prefatory field of healthcare that plays a vital role in the patient’s clinical and medical intervention. Recent developments in MI help healthcare professionals such as doctors and radiologists diagnose patients with improved quality of MI methods such as X-ray, CT and MRI scans. The upsurge of the Convolutional Neural Network (CNN) offers the latent for timely detection and classification of the disease with deeper insights. It adapts the spatial features and learns the input data hierarchically, which makes well-suited for this task. To improve the robustness of the CNN, the parameters are optimized through a nature-inspired dynamic mass balance-based metaheuristic approach named the Equilibrium Optimizer Algorithm (EOA). As a real-life implication, the suggested EOA-CNN approach has been verified by utilizing two benchmark datasets. The competence of EOA-CNN is assessed by present-day methods in terms of parameters namely accuracy, error rate, min, avg, std, accuracy, sensitivity and specificity. By streamlining the workflow, the said approach is a valuable asset conforming to the reliability to resolve healthcare imaging complications with advanced diagnostics support. This framework is being used to optimize the constraints meanwhile bridging the gaps in healthcare convenience.