Deep Learning Models for Digital Pathology

Abstract

Histopathology images; microscopy images of stained tissue biopsies contain fundamental prognostic information that forms the foundation of pathological analysis and diagnostic medicine. However, diagnostics from histopathology images generally rely on a visual cognitive assessment of tissue slides which implies an inherent element of interpretation and hence subjectivity. Access to digitized histopathology images enabled the development of computational systems aiming at reducing manual intervention and automating parts of pathologists' workflow. Specifically, applications of deep learning to histopathology image analysis now offer opportunities for better quantitative modeling of disease appearance and hence possibly improved prediction of disease aggressiveness and patient outcome. However digitized histopathology tissue slides are unique in a variety of ways and come with their own set of computational challenges. In this survey, we summarize the different challenges facing computational systems for digital pathology and provide a review of state-of-the-art works that developed deep learning-based solutions for the predictive modeling of histopathology images from a detection, stain normalization, segmentation, and tissue classification perspective. We then discuss the challenges facing the validation and integration of such deep learning-based computational systems in clinical workflow and reflect on future opportunities for histopathology derived image measurements and better predictive modeling.

Advancements in medical diagnosis and treatment through machine learning: A review

“…The next class of DL algorithms often used to evaluate feature representations are RNNs (Sanchez-Casanova et al, 2020). RNNs are particularly effective at processing time-series data, like the levels of gene expression in pathology (BenTaieb & Hamarneh, 2019). To comprehend the dynamics of gene regulation and how it links to the emergence of diseases, gene expression data, which offers information on the activity of genes within a cell, is frequently gathered over time.…”

Current trends of artificial intelligence and applications in digital pathology: A comprehensive review

“…We can also use these machines for deep learning algorithm training later. The images created by scanning are analyzed on computer systems by object detection algorithms or AI-based algorithms such as CNNs (Khosravi et al, 2018) or even deep learning algorithms (BenTaieb and Hamarneh, 2019). The results obtained from the algorithms give information about the disease, if any, and the prognosis of the disease, such as the progression of cancer (He et al, 2021).…”

Whole Slide Pathology Image Patch Based Deep Classification: An Investigation of the Effects of the Latent Autoencoder Representation and the Loss Function Form

“…Computer-aided diagnostic systems are a major area of development in digital pathology. The volume and complexity of analysis tasks often make unassisted human interpretation inefficient: apart from being time-consuming, human interpretation uses only a small fraction of morphological information presented on pathology slides [6]. Therefore, a more flexible and robust automated approach is required.…”