Efficient Anomaly Detection from Medical Signals and Images with CNNs for IoMT Systems.

Abstract

Deep learning falls in the machine learning family in the Artificial Intelligence (AI) field. It is one of the most prominent methods based on learning principles. The known traditional and Convolutional Neural Networks (CNNs) have been utilized in pattern recognition techniques based on the deep learning concepts on different images; due to the importance of Anomaly Detection (AD) in automatic diagnosis. It is an essential and vital tool in medical signal and image processing. In this paper, the AD is performed on medical EEG spectrograms and medical corneal images for IoMT systems. Deep learning based on the CNN models is employed in the processes of training and testing. Each input image passes through a series of convolution layers and kernels filters. For the classification, the pooling and Fully-Connected (FC) layers have been utilized for this purpose. Computer simulation experiments reveal the success and superiority of the presented proposed techniques in the automated medical diagnosis for Internet of Medical Things (IoMT) systems. This article is protected by copyright. All rights reserved.