DINO-LG: Enhancing Vision Transformers with Label Guidance for Coronary Artery Calcium Detection
Not on the map. Why?
Abstract (arXiv)
Coronary artery disease (CAD), one of the leading causes of mortality worldwide, necessitates effective risk assessment strategies, with coronary artery calcium (CAC) scoring via computed tomography (CT) being a key method for prevention. Traditional methods, primarily based on UNET architectures implemented on pre-built models, face challenges like the scarcity of annotated CT scans containing CAC and imbalanced datasets, leading to reduced performance in segmentation and scoring tasks. In this study, we address these limitations by introducing DINO-LG, a novel label-guided extension of DINO (self-distillation with no labels) that incorporates targeted augmentation on annotated calcified regions during self-supervised pre-training. Our three-stage pipeline integrates Vision Transformer (ViT-Base/8) feature extraction via DINO-LG trained on 914 CT scans comprising 700 gated and 214 non-gated acquisitions, linear classification to identify calcified slices, and U-NET segmentation for CAC quantification and Agatston scoring. DINO-LG achieved 89% sensitivity and 90% specificity for detecting CAC-containing CT slices, compared to standard DINO's 79% sensitivity and 77% specificity, reducing false-negative and false-positive rates by 49% and 57% respectively. The integrated system achieves 90% accuracy in CAC risk classification on 45 test patients, outperforming standalone U-NET segmentation (76% accuracy) while processing only the relevant subset of CT slices. This targeted approach enhances CAC scoring accuracy by feeding the UNET model with relevant slices, improving diagnostic precision while lowering healthcare costs by minimizing unnecessary tests and treatments.
Key facts. Cited by 1 paper in the map (data through 7 Oct 2026). Builds on 15 of its 50 references that are in the map. First posted on arXiv 12 Nov 2024.
Builds on
15 of 50 references.
References · 15
- 2015U-Net: Convolutional Networks for Biomedical Image Segmentationcited by 19,604
- 2023DINOv2: Learning Robust Visual Features without Supervisioncited by 7,764
- 2021Emerging Properties in Self-Supervised Vision Transformerscited by 6,893
- 2018Attention U-Net: Learning Where to Look for the Pancreascited by 1,420
- 2021Swin-Unet: Unet-like Pure Transformer for Medical Image Segmentationcited by 1,218
- 2023Evaluating General Purpose Vision Foundation Models for Medical Image Analysis: An Experimental Study of DINOv2 on Radiology Benchmarkscited by 30
- 2023Pretrained ViTs Yield Versatile Representations For Medical Imagescited by 13
- 2024Comparative Analysis of ImageNet Pre-Trained Deep Learning Models and DINOv2 in Medical Imaging Classificationcited by 13
- 2022Anatomy-aware Self-supervised Learning for Anomaly Detection in Chest Radiographscited by 12
- 2023Self-supervision for medical image classification: state-of-the-art performance with ~100 labeled training samples per classcited by 7
- 2023Shifting to Machine Supervision: Annotation-Efficient Semi and Self-Supervised Learning for Automatic Medical Image Segmentation and Classificationcited by 5
- 2023DINO-CXR: A self supervised method based on vision transformer for chest X-ray classificationcited by 3
3 more
- 2023USL-Net: Uncertainty Self-Learning Network for Unsupervised Skin Lesion Segmentationcited by 2
- 2022Automated Coronary Calcium Scoring using U-Net Models through Semi-supervised Learning on Non-Gated CT Scanscited by 1
- 2024Coronary artery segmentation in non-contrast calcium scoring CT images using deep learningcited by 1
Built on by
Citing papers per year: 2026: 1.
1 paper.