Getting pictures of inside organs for therapeutic reasons, such identifying or researching disorders, is known as medical imaging. Deep CNN comeback and the availability of computing power are the main drivers behind this. Expertise in diagnosis may be achieved by physicians by using deep learning systems that can detect hidden patterns in photos. It is now the most successful technique for computer-assisted diagnosis, organ segmentation, cancer detection, and illness classification. Numerous deep learning techniques have been described for the analysis of medical pictures for different types of diagnosis. The work utilizing the contemporary cutting-edge deep learning approaches for medical image processing is reviewed in this publication. A summary of convolutional neural network-based medical imaging research efforts is presented at the outset of this survey. We discuss how popular pretrained models and RNN improve the performance of convolutional networks in our second part. Lastly, we aggregate the performance indicators of deep learning models concentrating on lung illness diagnosis for easier direct assessment.