A Review of AI-Based Diagnosis of Multiple Thoracic Diseases in Chest Radiography

Artificial intelligence (AI) researchers using computer-aided diagnosis (CAD) obtained significant results in different areas of medical image. The CAD of thoracic diseases using deep learning particularly convolution neural network (CNN) on chest radiography is one among them. They are expected to assist radiologists with diagnostic excellence during the treatment of thoracic diseases using chest X-ray images. The CNN’s performance in dealing with one or two pathologies on chest X-ray has motivated AI researchers to go beyond it. Today, we have deep learning algorithms that can detect and classify multiple thoracic pathology at a time. In this article, we will review some breakthrough applications built with deep learning models such as CNNs to detect and classify multiple pathology in one exam under chest radiography. Also, we will discuss important design factors and future trends in computer-aided diagnosis of multi-disease classification problems in chest radiology.