Deep Learning Techniques for Computer-Aided Diagnosis of Various Cancers

Deep learning has emerged as a key machine learning technology in object detection, attracting attention for its results in medical imaging analysis. Convolutional neural networks (CNNs) are the most commonly used deep learning algorithms for this purpose, and they play an important role in the identification and potential early diagnosis of cancer. This chapter presents a survey on the advancements in deep learning techniques for cancer detection and diagnosis and hopes to provide an overview of the progress in this field. The chapter covers the progress in the diagnosis of the deadliest of cancers through computer-aided diagnosis (CAD) and discusses the challenges of deep learning techniques used for cancer detection and diagnosis such as segmentation, classification, analysis of region of interest (ROI), cancer, denoising, and filtering using popular state-of-the-art deep learning architectures such as convolutional neural networks, fully convolutional networks, and autoencoders. The chapter presents details of deep learning techniques based on the type of cancer and provides insight into the field’s progress.