A Proposed Architecture Based on Deep Learning and Optimization Techniques for Medical Diagnostic Imaging

Abstract

The COVID 19 pandemic is much more than a health crisis; it is an unparalleled socioeconomic problem that is placing strain on every country it affects. It has catastrophic social, economic, and political consequences that will take time to be repaired. COVID-19 has been diagnosed using a variety of methods, including body temperature measurement, reverse transcription polymerase chain reaction (RT-PCR), chest computed tomography (CT), and chest x-ray. The application of artificial intelligence technologies, namely deep learning tools, to the diagnosis of covid from x-ray images of patients is the focus of this research. In bigdata contexts, deep learning has already proven to be an effective and quick tool for processing and classifying images. On several occasions, it has also been utilized to diagnose medical images. In this paper, we propose a new model for classifying X-ray images based on the Chaotic-MRFO optimization algorithm for feature selection, CNN for feature extraction, and both of SVM and KNN for classification. In addition to the pre-processing step, we used the Matlab tool to convert the images from RGB to grayscale. We have demonstrated the efficacy of our model as an optimal model for diagnosing X-ray images with high precision in the covid context, as well as the ability to generalize and apply it to other medical images.