In this repo I try to collect a list of papers that focus on the segmentation of MRI in various anatomies. This is still a early list and will be actively expanded.
This repo will mainly focus on the papers that utilizes Machine Learning techniques to perfrom MRI segmentation, so most of the papers in the list will be after 2019.
If you find any interesting works feel free to create pull requests or email me to make the list more comprehensive.
Title | Year | Anatomy | Model | Comments |
---|---|---|---|---|
SynthSeg: Segmentation of brain MRI scans of any contrast and resolution without retraining [Paper] [Code] |
2023/05 | brain | U-Net | Utilizes generative model to generate synthetic scans during training |
Spinal Cord MRI Segmentation Techniques and Algorithms: A Survey [Paper] | 2021/04 | - | - | 2021 Survey paper |
Shape-Aware Meta-learning for Generalizing Prostate MRI Segmentation to Unseen Domains [Paper] [Code] | 2020/10 | prostate | Mix-residual-UNet | Focus on domain generalization, using 6 different T2 MRI dataset |
SAU-Net: Efficient 3D Spine MRI Segmentation Using Inter-Slice Attention [Paper] | 2020/09 | spine | Spatial Attention-based U-Net (SAU-Net) | Introduce inter-slice attention module to refine inter-slice segmentation, evaluation is limited |
Spine Magnetic Resonance Image Segmentation Using Deep Learning Techniques [Paper] | 2020/07 | spine | - | 2020 Survey paper on spine sementation for detection of vertrbrae malalignment |
From classical to deep learning: review on cartilage and bone segmentation techniques in knee osteoarthritis research [Paper] | 2020/03 | knee | 2020 Survey paper on knee bone segmentation | |
Fully Automated Patellofemoral MRI Segmentation using Holistically Nested Networks: Implications for Evaluating Patellofemoral Osteoarthritis, Pain, Injury, Pathology, and Adolescent Development [Paper][Code] | 2020/01 | knee bone | Holistically nested networks (HNN) | Segments bone and bone with active groth plates. 9-fold cross-validation. Tri-planar segementation |
DeepAtlas: Joint Semi-Supervised Learning of Image Registration and Segmentation [Paper] | 2019/10 | knee + brain | DeepAtlas (backbone 3D U-Net) | Includes regsitration in the pipeline. Joint learning of weakly supervised registration and semisupervised segmentation. Combines segmentation loss, intensity + anatomy similarity loss |
Automatic segmentation of the spinal cord and intramedullary multiple sclerosis lesions with convolutional neural networks [Paper] | 2019/01 | spinal cord | 2D CNN + 3D CNN | 2D CNN for spinal cord detection, 3D CNN for segmentation of spinal cord and intramedullary lesions. Trained multiple networks for different MRI protocols (T1, T2, T2*) |