A content-based image retrieval system (CBIR) is required as a result of the increased use of digital imaging data and the difficulty of locating the information that hospitals need within the enormous database. A content-based medical image retrieval (CBMIR) system may be a powerful method for enhancing the detection and treatment of numerous diseases as well as a cutting-edge device for managing vast amounts of data. Accessing, maintaining, and removing useful material from these enormous databases is quite challenging in the absence of such solutions. Medical picture retrieval that depends on textual information, such as tags and manual annotation, has a low efficiency because it requires labour, medical expertise, and time. Radiologists must work hard and take their time to accurately diagnose any condition. As a result, radiologists frequently struggle to make an accurate illness diagnosis. Computer-aided diagnosis (CAD) systems can analyse CT scans to automatically identify and diagnose the symptoms, freeing up radiologists' time. By using content-based medical image retrieval (CBMIR), diagnosis and recognition can be done more quickly and accurately. A deep convolutional neural network (CNN)-based novel intelligent CBMIR strategy for extracting CIS that aids in recognising and classifying diseases is badly needed in today's society. In order to automatically identify and retrieve images using feature representations obtained from the images themselves, medical image retrieval systems are needed. An individually created deep convolutional neural network (CNN) can be used to do this. Building the system is made simpler by CNN's minimal pre-processing and lack of additional feature extraction approaches.