Evaluating the Efficiency of Task Offloading in Mobile Edge Cloud Computing for Deep Learning-Based Brain Tumor Detection

Abstract

The work investigates how to improve deep learning (DL) model performance for brain tumor classification by utilizing Mobile Edge Computing (MEC) and Task Offloading. The system takes into account the computing power of edge servers and mobile devices, as well as the communication expenses related to job offloading. The extent of learning parameters in these designs is determined by the hyperparameters for width and depth. The findings demonstrate that task offloading keeps accuracy while cutting down on the amount of time and effort required for DL models. The study highlights that while selecting task offloading, computation and communication costs must be balanced. Future medical solutions may be more easily available and effective as a result of these developments.