The domain of medical diagnosis and treatment planning has undergone a significant transformation due to the integration of deep learning in the analysis of medical images. In fact, deep learning outperforms conventional image processing techniques in terms of accuracy and efficiency by using the capacity of neural networks, in particular convolutional neural networks, to extract complex information from medical images such as MRIs, CT scans and X-rays. This chapter explores the important role of deep learning models in detecting and diagnosing a wide range of medical conditions, including cancers, neurological disorders, and cardiovascular diseases. To do so, we highlight significant advances of deep learning in main medical image analysis frameworks, such as segmentation, anomaly detection, and classification. These advances increasingly improve the performance of the deep learning models while reducing the time and effort required by clinicians to perform outcomes. Furthermore, we address main challenges encountered in medical image analysis, particularly the lack of annotated and balanced data and the non-interpretability of deep learning models. Moreover, we discuss the potential for improving deep learning models with other emergent fields such as radiomics and computer vision. Thus, through concrete applications and real-life case studies, this chapter aims to highlight the impact of deep learning on improving clinical decision-making and patient outcomes to overcome current challenges and reshape the future of medicine.