Medical imaging is essential in a variety of medical applications, like medical treatments had been used for early identification, tracking, prognosis, and diagnosis testing of different medical problems. Machine learning is critical in the field of image processing and computer vision. Machine learning (ML) techniques are used to analyze information from visual data in problems ranging from image segmentation and registration to formation, object classification, and scene understanding. Deep learning (DL) is being used to analyze medical images, which is a growing field. DL methodologies and their applications to computer-aided diagnosis involve standard machine learning techniques in the field of computer vision, deep learning ML models, and applications to medical image processing. Many of the most recent machine learning technologies in computer-aided diagnosis and medical image processing are the classification of objects such as lesions into specific classes associated with the input attributes such as contrast and area acquired from segmented object classes. Theoretically, an artificial neural network is influenced by neural structures. The Neocognitron, CNNs, and neural filters all are significant deep learning techniques. Image-based machine learning, which includes deep learning, is a useful and high-performing technology. In the upcoming years, deep learning will become the standard technology for medical image analysis. We present a review of recent machine learning and deep learning approaches for detecting four diseases, including tuberculosis, lung cancer, pneumonia, and COVID-19, in this study. We review the disease which are detected and classified from X-ray images. We intend to explore the most accurate technique for detecting various diseases as part of this study, which will be useful in the future.