Gastric Ulcer Detection in Endoscopic Images Using MobileNetV3-Small

In modern medicine, endoscopy plays a very important role as it allows physicians to detect severe diseases in early development stages. However, diagnosing patients is a challenging duty, as it requires many years of experience from doctors. In this study we proposed a new algorithm based on MobilenetV3-Small architecture to detect lesion areas of the gastrointestinal tract and peptic ulcers by the implementation of deep learning algorithms, including R-FCN, Resnet101, Yolov5, and MobilenetV3-Small, and discuss the possible results of MobilenetV3-Small algorithms in detection of ulcers. The MobilenetV3-Small architecture surpasses other tested algorithms when it comes to the speed of inference. Its efficient utilization of memory resources makes it particularly suitable for deployment on devices with limited resources, thereby improving its operational efficiency. This kind of computer aided systems may potentially help to save time, lower the cost of endoscopic procedures, and lower the risk of such procedures for the patients. To underpin the advantages of MobilenetV3-Small this paper includes a detailed overview and comparison of metrics of other models popular within the two past decades, targeting this field of research.