MSA-UNet3+: Multi-Scale Attention UNet3+ with New Supervised Prototypical Contrastive Loss for Coronary DSA Image Segmentation
Not on the map. Why?
Abstract (arXiv)
Accurate segmentation of coronary Digital Subtraction Angiography (DSA) images is essential for diagnosing and treating coronary artery disease (CAD). Despite advances in deep learning, challenges such as high intra-class variance and class imbalance limit precise vessel delineation. Existing approaches for coronary DSA segmentation cannot effectively address these issues. Furthermore, existing segmentation network encoders do not directly generate semantic embeddings, which could enable the decoder to reconstruct segmentation masks more effectively. We propose a Supervised Prototypical Contrastive Loss (SPCL) that combines supervised and prototypical contrastive learning to enhance coronary DSA image segmentation. The supervised contrastive loss enforces semantic embeddings in the encoder, improving feature differentiation. The prototypical contrastive loss enables the model to focus on the foreground class while alleviating high intra-class variance and class imbalance by concentrating only on hard-to-classify background samples. We implement the proposed SPCL within MSA-UNet3+, a Multi-Scale Attention-Enhanced UNet3+ architecture. The architecture integrates a Multi-Scale Attention Encoder (M-encoder), a Multi-Scale Dilated Bottleneck (MSD-Bottleneck) for multi-scale feature extraction, and a Contextual Attention Fusion Module (CAFM) to preserve fine-grained details while improving contextual understanding. Experiments on a private coronary DSA dataset demonstrate that MSA-UNet3+ outperforms state-of-the-art methods, achieving the highest Dice coefficient and F1-score while significantly reducing ASD and ACD. The framework provides precise vessel segmentation for accurate identification of coronary stenosis and supports informed diagnostic and therapeutic decisions. The code will be released at https://github.com/rayanmerghani/MSA-UNet3plus.
Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Builds on 27 of its 64 references that are in the map. First posted on arXiv 7 Apr 2025.
Builds on
27 of 64 references.
References · 27
- 2015U-Net: Convolutional Networks for Biomedical Image Segmentationcited by 19,713
- 2020Supervised Contrastive Learningcited by 2,829
- 2018UNet++: A Nested U-Net Architecture for Medical Image Segmentationcited by 1,525
- 2018Attention U-Net: Learning Where to Look for the Pancreascited by 1,435
- 2021Swin-Unet: Unet-like Pure Transformer for Medical Image Segmentationcited by 1,222
- 2020Image Segmentation Using Deep Learning: A Surveycited by 660
- 2020Contrastive learning of global and local features for medical image segmentation with limited annotationscited by 309
- 2018Recurrent Residual Convolutional Neural Network based on U-Net (R2U-Net) for Medical Image Segmentationcited by 180
- 2018A Novel Focal Tversky loss function with improved Attention U-Net for lesion segmentationcited by 172
- 2019Deep learning for cardiac image segmentation: A reviewcited by 131
- 2020Contrastive Learning for Label-Efficient Semantic Segmentationcited by 93
- 2021Unified Focal loss: Generalising Dice and cross entropy-based losses to handle class imbalanced medical image segmentationcited by 88
15 more
- 2019Bi-Directional ConvLSTM U-Net with Densley Connected Convolutionscited by 82
- 2022Learning with Limited Annotations: A Survey on Deep Semi-Supervised Learning for Medical Image Segmentationcited by 76
- 2022Towards Visual-Prompt Temporal Answering Grounding in Medical Instructional Videocited by 43
- 2021Analyzing Overfitting under Class Imbalance in Neural Networks for Image Segmentationcited by 34
- 2023CMUNeXt: An Efficient Medical Image Segmentation Network based on Large Kernel and Skip Fusioncited by 31
- 2022CMU-Net: A Strong ConvMixer-based Medical Ultrasound Image Segmentation Networkcited by 29
- 2023From CNN to Transformer: A Review of Medical Image Segmentation Modelscited by 22
- 2023Image Augmentation with Controlled Diffusion for Weakly-Supervised Semantic Segmentationcited by 19
- 2023DA-TransUNet: Integrating Spatial and Channel Dual Attention with Transformer U-Net for Medical Image Segmentationcited by 16
- 2024Adaptive Patch Contrast for Weakly Supervised Semantic Segmentationcited by 15
- 2024Image Augmentation Agent for Weakly Supervised Semantic Segmentationcited by 15
- 2024PMFSNet: Polarized Multi-scale Feature Self-attention Network For Lightweight Medical Image Segmentationcited by 6
- 2024Accurate Explanation Model for Image Classifiers using Class Association Embeddingcited by 6
- 2024A Weakly Supervised and Globally Explainable Learning Framework for Brain Tumor Segmentationcited by 6
- 2023Performance Analysis of UNet and Variants for Medical Image Segmentationcited by 1
Built on by
Citing papers per year: 2026: 1.
1 paper.