This repository is mainly dedicated for listing the recent research advancements in the application of Self-Supervised-Learning in medical images computing field. Inspired by awesome-self-supervised-learning
What is self-supervised learning?
Self-Supervised learning (SSL) is a hybrid learning approach that combines both supervised and unsupervised learning simultaneously. More clearly, SSL is an approach that aims at learning semantically useful features for a certain task by generating supervisory signal from a pool of unlabeled data without the need for human annotation. These representations is then used for subsequent tasks where the amount of labeled data is limited.
Self-Supervised Learning pipelines in computer vision
Why Self-Supervised learning in medical imaging ?
- Unlabeled medical imaging data is a abundant, but human annotated data is scarce.
- building a large enough human annotated medical imaging datasets is:
- Expensive.
- Time consuming.
- Requires experienced personnel.
- Prone to patients’ privacy preserving issues.
This repository is a continuation of our survey in the field, please read and consider citing it in your work:
Call for Contribution
Please help contribute this list by contacting me or add pull request
Markdown format: height
- Paper Name.
[[pdf]](link)
[[code]](link)
- Author 1, Author 2, and Author 3. *Conference Year*
Criteria
-
A list of recent self-supervised learning papers in medical imaging published since 2017.
-
Papers are collected from peer-reviewed journals and high reputed conferences. However, it might have recent papers on arXiv.
-
A meta-data is required along the paper, e.g. category.
List of Journals / Conferences (J/C):
- IEEE Access
- IEEE Transaction on Medical Imaging (IEEE-TMI)
- IEEE Transaction on Biomedical Engineering (IEEE-TBME)
- IEEE Journal of Biomedical and Health Informatics (IEEE-JBHI)
- Applied Soft Computing (ASC)
- Medical Image Analysis (MedIA)
- International Journal of Computer Assisted Radiology and Surgery (IJCARS)
- Nature Machine Intelligence (NMI)
- Pattern Recognition
- Expert Systems with Applications (ESA)
- Neurocomputing
- Proceedings of Machine Learning Research (PMLR)
- Annual Conference on Neural Information Processing Systems (NIPS)
- International Conference on Information Processing in Medical Imaging (IPMI)
- International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI)
- International Conference on Medical Imaging with Deep Learning (MIDL)
- IEEE International Symposium on Biomedical Imaging (ISBI)
- Joint European Conference on Machine Learning and Knowledge Discovery in Databases (JECMLKDD)
- IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)
- International Workshop on Deep Learning in Medical Image Analysis (DLMIA)
2021
Paper title | Journal/Conference | Category | Paper link | Code link |
---|---|---|---|---|
Multimodal Self-supervised Learning for Medical Image Analysis | IPMI | Predictive | Link | N/A |
Self-supervised multimodal reconstruction pre-training for retinal computer-aided diagnosis | ESA | Generative | Link | N/A |
MedAug: Contrastive learning leveraging patient metadata improves representations for chest X-ray interpretation | ArXiv | Contrastive | Link | N/A |
COVID-19 Prognosis via Self-Supervised Representation Learning and Multi-Image Prediction | ArXiv | Contrastive | Link | pytorch |
Momentum contrastive learning for few-shot COVID-19 diagnosis from chest CT images | Pattern Recognition | Contrastive | Link | N/A |
Big Self-Supervised Models Advance Medical Image Classification | ArXiv | Contrastive | Link | N/A |
Self-supervised Multi-task Representation Learning for Sequential Medical Images | JECMLKDD | Multiple-tasks/Multi-tasking | Link | N/A |
Self-path: Self-supervision for classification of pathology images with limited annotations | TMI | Multiple-tasks/Multi-tasking | Link | N/A |
Twin self-supervision based semi-supervised learning (TS-SSL): Retinal anomaly classification in SD-OCT images | Neurocomputing | Multiple-tasks/Multi-tasking | Link | tensorflow |
Rotation-oriented collaborative self-supervised learning for retinal disease diagnosis. | TMI | Multiple-tasks/Multi-tasking | Link | tensorflow |
Volumetric white matter tract segmentation with nested self-supervised learning using sequential pretext tasks | MedIA | Multiple-tasks/Multi-tasking | Link | N/A |
2020
Paper title | Journal/Conference | Category | Paper link | Code link |
---|---|---|---|---|
Self-Loop Uncertainty: A Novel Pseudo-Label for Semi-supervised Medical Image Segmentation | MICCAI | Predictive | Link | N/A |
Rubik’s Cube+: A self-supervised feature learning framework for 3D medical image analysis | MedIA | Predictive | Link | N/A |
Self-Supervised Learning Based on Spatial Awareness for Medical Image Analysis | IEEE Access | Predictive | Link | N/A |
Self-supervised Skull Reconstruction in Brain CT Images with Decompressive Craniectomy | MICCAI | Generative | Link | pytorch |
Learning the retinal anatomy from scarce annotated data using self-supervised multimodal reconstruction | ASC | Generative | Link | N/A |
Multimodal Transfer Learning-based Approaches for Retinal Vascular Segmentation | ArXiv | Generative | Link | N/A |
Multi-modal self-supervised pre-training for joint optic disc and cup segmentation in eye fundus images | ICASSP | Generative | Link | N/A |
Self-supervised retinal thickness prediction enables deep learning from unlabelled data to boost classification of diabetic retinopathy | NMI | Generative | Link | tensorflow |
Leveraging Self-supervised Denoising for Image Segmentation | ISBI | Generative | Link | tensorflow |
Self-Supervised Pretraining with DICOM metadata in Ultrasound Imaging | PMLR | Generative | Link | N/A |
Revisiting rubik’s cube: Self-supervised learning with volume-wise transformation for 3d medical image segmentation | MICCAI | Generative | Link | N/A |
Semi-supervised breast cancer histology classification using deep multiple instance learning and contrast predictive coding | ArXiv | Contrastive | Link | N/A |
Embedding Task Knowledge into 3D Neural Networks via Self-supervised Learning | ArXiv | Contrastive | Link | N/A |
PGL: Prior-Guided Local Self-supervised Learning for 3D Medical Image Segmentation | ArXiv | Contrastive | Link | pytorch |
Self-Supervised Feature Learning via Exploiting Multi-Modal Data for Retinal Disease Diagnosis | TMI | Contrastive | Link | pytorch |
MoCo Pretraining Improves Representation and Transferability of Chest X-ray Models | PMLR | Contrastive | Link | pytorch |
Contrastive learning of global and local features for medical image segmentation with limited annotations | ArXiv | Contrastive | Link | tensorflow |
Self-Supervised Representation Learning for Ultrasound Video | ISBI | Multiple-tasks/Multi-tasking | Link | N/A |
A Multi-Task Self-Supervised Learning Framework for Scopy Images | ISBI | Multiple-tasks/Multi-tasking | Link | N/A |
3D Self-Supervised Methods for Medical Imaging--update references | NIPS | Multiple-tasks/Multi-tasking | Link | tensorflow |
Retinal Image Classification by Self-Supervised Fuzzy Clustering Network | IEEE Access | Multiple-tasks/Multi-tasking | Link | N/A |
Learning semantics-enriched representation via self-discovery, self-classification, and self-restoration | MICCAI | Multiple-tasks/Multi-tasking | Link | pytorch |
SAR: Scale-Aware Restoration Learning for 3D Tumor Segmentation | ArXiv | Multiple-tasks/Multi-tasking | Link | N/A |
2019
Paper title | Journal/Conference | Category | Paper link | Code link |
---|---|---|---|---|
Self-Supervised Learning for Cardiac MR Image Segmentation by Anatomical Position Prediction | MICCAI | Predictive | Link | N/A |
Self-supervised Feature Learning for 3D Medical Images by Playing a Rubik’s Cube | MICCAI | Predictive | Link | N/A |
Self-supervised learning for medical image analysis using image context restoration | MedIA | Generative | Link | N/A |
Models Genesis: Generic Autodidactic Models for 3D Medical Image Analysis | MICCAI | Generative | Link | tensorflow |
Surrogate Supervision for Medical Image Analysis: Effective Deep Learning From Limited Quantities of Labeled Data | ISBI | Multiple-tasks/Multi-tasking | Link | N/A |
2018
Paper title | Journal/Conference | Category | Paper link | Code link |
---|---|---|---|---|
Exploiting the potential of unlabeled endoscopic video data with self-supervised learning | IJCARS | Generative | Link | N/A |
Improving Cytoarchitectonic Segmentation of Human Brain Areas with Self-supervised Siamese Networks | MICCAI | Predictive | Link | N/A |
2017
Paper title | Journal/Conference | Category | Paper link | Code link |
---|---|---|---|---|
Self-supervised Learning for Spinal MRIs | DLMIA | Contrastive | Link | N/A |
Self supervised deep representation learning for fine-grained body part recognition | ISBI | Predictive | Link | N/A |