ABSTRACT
Medical Imaging incorporates various imaging modalities which generate virtual representations of the internal organs of the body, and it plays a vital role in clinical diagnosis for patients. With the advent of new technologies introduced in healthcare over the years, medical imaging has gained tremendous attention among researchers. Deep learning models have proved to be substantial in identifying and classifying the patterns in biomedical images. This chapter provides an overview of the models of deep learning that are implemented in medical images, and it can be viewed as an enhancement of the conventional approach to enhance the visualization of medical images. Deep learning techniques can be classified into two-dimensional and three-dimensional structures to analyze image modalities. Convolutional neural networks, recurrent neural networks, U-Net, etc., are basic models for deep learning which can be further categorized. This chapter also provides an overview of the advancements, current models and the respective architectures followed for deep learning in healthcare in recent times. Three-dimensional medical image processing is one of the emerging fields in serving medical applications. This chapter will cover the basic deep learning techniques applied to three-dimensional structures for various medical applications. The second part is the analysis of different deep learning models applied to medical images with their advantages, disadvantages and applications.
