Towards label-efficient automatic diagnosis and analysis: a comprehensive survey of advanced deep learning-based weakly-supervised, semi-supervised and self-supervised techniques in histopathological image analysis

Abstract

Histopathological images contain abundant phenotypic information and pathological patterns, which are the gold standards for disease diagnosis and essential for the prediction of patient prognosis and treatment outcome. In recent years, computer-automated analysis techniques for histopathological images have been urgently required in clinical practice, and deep learning methods represented by convolutional neural networks have gradually become the mainstream in the field of digital pathology. However, obtaining large numbers of fine-grained annotated data in this field is a very expensive and difficult task, which hinders the further development of traditional supervised algorithms based on large numbers of annotated data. More recent studies have started to liberate from the traditional supervised paradigm, and the most representative ones are the studies on weakly supervised learning paradigm based on weak annotation, semi-supervised learning paradigm based on limited annotation, and self-supervised learning paradigm based on pathological image representation learning. These new methods have led a new wave of automatic pathological image diagnosis and analysis targeted at annotation efficiency. With a survey of over 130 papers, we present a comprehensive and systematic review of the latest studies on weakly supervised learning, semi-supervised learning, and self-supervised learning in the field of computational pathology from both technical and methodological perspectives. Finally, we present the key challenges and future trends for these techniques.

Harnessing artificial intelligence for prostate cancer management

“…Different training strategies require different levels of annotation information. 24 Supervised learning requires pathologists to manually draw every gland on the whole-slide images (WSIs) and provide Gleason pattern (GP) information, which is time consuming. Weakly supervised learning typically uses slide- or patient-level Gleason scores recorded in pathology reports as inputs, without the need for additional annotations.…”

Artificial Intelligence in Endoscopic Ultrasonography-Guided Fine-Needle Aspiration/Biopsy (EUS-FNA/B) for Solid Pancreatic Lesions: Opportunities and Challenges

“…In particular, the algorithm's operation relies on the graphics processing unit (GPU); but, the current storage capacity of GPUs is limited, making it challenging to fully utilize all the information in whole slide images (WSIs) or other image formats, which can result in the loss of some useful information [87,88]. Additionally, supervised learning is a common approach for most CNNs used in deep learning, which requires pathologists to accurately label ROIs in the images, adding to the cost [89]. Furthermore, AI analysis relies on highquality training datasets, which require a substantial number of training images and can be time-consuming to prepare [90].…”

MAMILNet: advancing precision oncology with multi-scale attentional multi-instance learning for whole slide image analysis

“…( 1 ) Qu et al. ( 2 ). Whole Slide Image (WSI) analysis, enabled by deep learning algorithms, shows promise in tumor detection, typing, and drug treatment response prediction, heralding a new era of precision medicine in oncology Cheplygina et al.…”