Applying Self-Supervised Learning to Medicine: Review of the State of the Art and Medical Implementations

(A Review of Self-Supervised Learning in Medicine)

Selected highlights from the 2021 Self-Supervised Learning Review [https://doi.org/10.3390/informatics8030059] that reviewed over 3,012 works related to the field of self-supervised learning.

Table of contents:

List of papers for Pixel-to-Scalar Self-Supervised Learning

Method Work Used By How Many Works
Context Prediction LINK 1487
Context Free Network LINK 1251
DeepPermNet LINK 69
RotNet LINK 1166
S4L LINK 310
Rotation Feature Decoupling LINK 74
Cross and Learn LINK 47
Egomotion LINK 513

List of papers for Pixel-to-Scalar Self-Supervised Learning

Method Work Used By How Many Works
Inpainting LINK 3210
Colorization LINK 635
Colorization LINK 2043
Split-Brain Autoencoder LINK 419
Multi-Task SSL LINK 418

List of papers for Adversarial Self-Supervised Learning

Method Work Used By How Many Works
Adversarial Feature Learning LINK 1353
InfoGAN LINK 3112
Artifact Detection LINK 76
Auxiliary Rotation Loss LINK 136

List of papers for Contrastive Self-Supervised Learning

Method Work Used By How Many Works
Local Aggregation LINK 197
CPC LINK 490
SimCLR LINK 2161
MoCo LINK 1720

List of papers applied to medicine

Application Work
Self-Supervised Learning for Spinal MRIs LINK
Brain Lesion Detection and Segmentation LINK
Cytoarchitectonic Segmentation of Human Brain Areas LINK
Dense Depth Estimation in Monocular Endoscopy LINK
Exploiting the potential of unlabeled endoscopic video data with self-supervised learning LINK
Motion Estimation and Segmentation for Cardiac MR Image Sequences LINK
Cardiac MR Image Segmentation by Anatomical Position Prediction LINK
Modelsgenesis:Genericautodidactic models for 3d medical image analysis LINK
Surrogate Supervision for Medical Image Analysis LINK
Deep learned tissue “fingerprints” classify breast cancers by ER/PR/Her2 status LINK
Learning Unsupervised Feature Representations For Single Cell Microscopy Images LINK
Segmentation of White Blood Cell Images by Self-Supervised Learning LINK
Self-Supervised Similarity Learning for Digital Pathology LINK
Automated acquisition of explainable knowledge from unannotated histopathology images LINK
Neural Image Compression for Gigapixel Histopathology Image Analysis LINK
Unsupervised Learning for Cell-level Visual Representation in Histopathology Images LINK
Semi-Supervised Histology Classification LINK

Complete details of all manuscripts that were reviewed

Database Keyword Details
Google scholar Self-supervised learning LINK
Google scholar Selfsupervised learning LINK
Google scholar Representation learning LINK
CrossRef Self-supervised learning LINK
CrossRef Selfsupervised learning LINK
CrossRef Representation learning LINK
Scopus Self-supervised learning LINK
Scopus Selfsupervised learning LINK
Scopus Representation learning LINK