A review of deep learning in medical imaging: Imaging traits, technology trends, case studies with progress highlights, and future promises

Abstract

Since its renaissance, deep learning has been widely used in various medical imaging tasks and has achieved remarkable success in many medical imaging applications, thereby propelling us into the so-called artificial intelligence (AI) era. It is known that the success of AI is mostly attributed to the availability of big data with annotations for a single task and the advances in high performance computing. However, medical imaging presents unique challenges that confront deep learning approaches. In this survey paper, we first highlight both clinical needs and technical challenges in medical imaging and describe how emerging trends in deep learning are addressing these issues. We cover the topics of network architecture, sparse and noisy labels, federating learning, interpretability, uncertainty quantification, etc. Then, we present several case studies that are commonly found in clinical practice, including digital pathology and chest, brain, cardiovascular, and abdominal imaging. Rather than presenting an exhaustive literature survey, we instead describe some prominent research highlights related to these case study applications. We conclude with a discussion and presentation of promising future directions.

Deep Learning-based Biological Anatomical Landmark Detection in Colonoscopy Videos

“…According to the literature [10] [11], biological anatomical landmark detection plays an essential role in various medical image analysis assignments, which can help achieve registration [12] and segmentation [13] tasks of medical images.…”

Medical Image Segmentation With Limited Supervision: A Review of Deep Network Models

“…Medical image segmentation involves delineating the anatomical or pathological structures from medical images of various modalities. As pointed out in [17], medical images are heterogonous with imbalanced classes and have multiple modalities with sparse annotations. Thus, it is complicated and challenging to analyze various medical images.…”

“…Several comprehensive surveys exist about the deep learning methods [15], [17] or its subcategories such as generative adversarial networks (GAN) [18] for general medical image Fig. 1: A taxonomy of medical image segmentation under limited supervision.…”

“…1: A taxonomy of medical image segmentation under limited supervision. analysis (including classification, reconstruction, detection, registration, and segmentation) [15], [17], [19] or a specialized topic [1], [4], [9], [12], [13], such as digital pathology image analysis [12], [20], and incorporating domain knowledge [21]. While several reviews have reviewed the application of deep learning for a specific application of the segmentation, such as cardiac image segmentation [4], brain tissue segmentation [1], brain tumor segmentation [22], and segmentation for covid-19 [13], the surveys by [16], [23], [24] review advancement of deep network architectures, losses, and training strategies for medical image segmentation.…”

Distributional Gaussian Process Layers for Outlier Detection in Image Segmentation

“…Deep learning methods have achieved state-of-the-art results on a plethora of medical image segmentation tasks Zhou et al (2020). However, their application in clinical settings is very limited due to issues pertaining to lack of reliability and miscalibration of estimated confidence in predictions.…”