This survey provides an overview of deep learning applications in medical image analysis, summarizing over 300 contributions, most published in the last year. It covers key concepts, techniques, and architectures used in image classification, object detection, segmentation, registration, and other tasks. The review highlights major application areas such as neuro, retinal, pulmonary, digital pathology, breast, cardiac, abdominal, and musculoskeletal imaging. It discusses the current state-of-the-art, challenges, and future research directions. The paper emphasizes the shift from handcrafted features to deep learning models that automatically learn features. It reviews major deep learning techniques, including supervised and unsupervised learning, neural networks, convolutional neural networks (CNNs), recurrent neural networks (RNNs), and unsupervised models like auto-encoders and restricted Boltzmann machines. The survey also discusses the use of deep learning in medical imaging tasks such as classification, detection, segmentation, and registration, highlighting the effectiveness of CNNs in medical image analysis. It notes the increasing use of CNNs in medical imaging, with recent studies using architectures like AlexNet, VGG, Inception, and ResNet. The paper also discusses the challenges of applying deep learning to medical imaging, including data scarcity, class imbalance, and the need for efficient processing. It concludes that deep learning has become a standard in medical image analysis, with ongoing research addressing open challenges and future directions.This survey provides an overview of deep learning applications in medical image analysis, summarizing over 300 contributions, most published in the last year. It covers key concepts, techniques, and architectures used in image classification, object detection, segmentation, registration, and other tasks. The review highlights major application areas such as neuro, retinal, pulmonary, digital pathology, breast, cardiac, abdominal, and musculoskeletal imaging. It discusses the current state-of-the-art, challenges, and future research directions. The paper emphasizes the shift from handcrafted features to deep learning models that automatically learn features. It reviews major deep learning techniques, including supervised and unsupervised learning, neural networks, convolutional neural networks (CNNs), recurrent neural networks (RNNs), and unsupervised models like auto-encoders and restricted Boltzmann machines. The survey also discusses the use of deep learning in medical imaging tasks such as classification, detection, segmentation, and registration, highlighting the effectiveness of CNNs in medical image analysis. It notes the increasing use of CNNs in medical imaging, with recent studies using architectures like AlexNet, VGG, Inception, and ResNet. The paper also discusses the challenges of applying deep learning to medical imaging, including data scarcity, class imbalance, and the need for efficient processing. It concludes that deep learning has become a standard in medical image analysis, with ongoing research addressing open challenges and future directions.