Benchmarking DINOv3 for Multi-Task Stroke Analysis on Non-Contrast CT

• URL: http://arxiv.org/abs/2509.23132v1
• Date: Sat, 27 Sep 2025 05:33:46 GMT
• Title: Benchmarking DINOv3 for Multi-Task Stroke Analysis on Non-Contrast CT
• Authors: Donghao Zhang, Yimin Chen, Kauê TN Duarte, Taha Aslan, Mohamed AlShamrani, Brij Karmur, Yan Wan, Shengcai Chen, Bo Hu, Bijoy K Menon, Wu Qiu,
• Abstract summary: Non-contrast computed tomography (NCCT) is essential for rapid stroke diagnosis but is limited by low image contrast and signal to noise ratio.<n>We address this challenge by leveraging DINOv3, a state-of-the-art self-supervised vision transformer, to generate powerful feature representations for a comprehensive set of stroke analysis tasks.
• Score: 7.576806888399401
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Non-contrast computed tomography (NCCT) is essential for rapid stroke diagnosis but is limited by low image contrast and signal to noise ratio. We address this challenge by leveraging DINOv3, a state-of-the-art self-supervised vision transformer, to generate powerful feature representations for a comprehensive set of stroke analysis tasks. Our evaluation encompasses infarct and hemorrhage segmentation, anomaly classification (normal vs. stroke and normal vs. infarct vs. hemorrhage), hemorrhage subtype classification (EDH, SDH, SAH, IPH, IVH), and dichotomized ASPECTS classification (<=6 vs. >6) on multiple public and private datasets. This study establishes strong benchmarks for these tasks and demonstrates the potential of advanced self-supervised models to improve automated stroke diagnosis from NCCT, providing a clear analysis of both the advantages and current constraints of the approach. The code is available at https://github.com/Zzz0251/DINOv3-stroke.

Related papers

• Multi-View Stenosis Classification Leveraging Transformer-Based Multiple-Instance Learning Using Real-World Clinical Data [76.89269238957593]
  Coronary artery stenosis is a leading cause of cardiovascular disease, diagnosed by analyzing the coronary arteries from multiple angiography views.<n>We propose SegmentMIL, a transformer-based multi-view multiple-instance learning framework for patient-level stenosis classification.
  arXiv  Detail & Related papers  (2026-02-02T13:07:52Z)
• Leveraging Unlabeled Scans for NCCT Image Segmentation in Early Stroke Diagnosis: A Semi-Supervised GAN Approach [4.199320411821769]
  Ischemic stroke is a time-critical medical emergency where rapid diagnosis is essential for improving patient outcomes.<n>Non-contrast computed tomography (NCCT) serves as the frontline imaging tool, yet it often fails to reveal the subtle ischemic changes present in the early, hyperacute phase.<n>We introduce a semi-supervised segmentation method using generative adversarial networks (GANs) to accurately delineate early ischemic stroke regions.
  arXiv  Detail & Related papers  (2025-11-24T18:14:53Z)
• Improve Retinal Artery/Vein Classification via Channel Couplin [12.219907282789285]
  We propose a novel loss named Channel-Coupled Vessel Consistency Loss to enforce the coherence and consistency between vessel, artery and vein predictions.<n>We also introduce a regularization term named intra-image pixel-level contrastive loss to extract more discriminative feature-level fine-grained representations.<n>SOTA results have been achieved across three public A/V classification datasets including RITE, LES-AV and HRF.
  arXiv  Detail & Related papers  (2025-07-31T18:43:02Z)
• Depth-Sequence Transformer (DST) for Segment-Specific ICA Calcification Mapping on Non-Contrast CT [38.85617601239779]
  Conventional 3D models are forced to process downsampled volumes or isolated patches.<n>We reformulate the 3D challenge as a textbfParallel Probabilistic Landmark localization task along the 1D axial dimension.<n>We propose the textbfDepth-Sequence Transformer (DST), a framework that processes full-resolution CT volumes as sequences of 2D slices.
  arXiv  Detail & Related papers  (2025-07-10T23:12:12Z)
• Bayesian Deep Learning Approaches for Uncertainty-Aware Retinal OCT Image Segmentation for Multiple Sclerosis [0.0]
  Optical Coherence Tomography ( OCT) provides valuable insights in ophthalmology, cardiology, and neurology.<n>One critical task for ophthalmologists using OCT is delineation of retinal layers within scans.<n>Previous efforts to automate delineation using deep learning face challenges in uptake from clinicians and statisticians.
  arXiv  Detail & Related papers  (2025-05-17T15:56:17Z)
• Beyond the Eye: A Relational Model for Early Dementia Detection Using Retinal OCTA Images [43.73298205923969]
  We present a novel PolarNet+ that uses retinal optical coherence tomography angiography ( OCTA) to discriminate early-onset Alzheimer's disease (AD) and mild cognitive impairment (MCI) subjects from controls.<n>Our method first maps OCTA images from Cartesian coordinates to polar coordinates, allowing approximate sub-region calculation.<n>We then introduce a multi-view module to serialize and analyze the images along three dimensions for comprehensive, clinically useful information extraction.
  arXiv  Detail & Related papers  (2024-08-09T15:10:34Z)
• Dual-Task Vision Transformer for Rapid and Accurate Intracerebral Hemorrhage CT Image Classification [9.058538258023383]
  Intracerebral hemorrhage (ICH) is a medical condition caused by the rupture of blood vessels in the brain. We propose a neural network structure, dual-task vision transformer (DTViT) for the automated classification and diagnosis of ICH images.
  arXiv  Detail & Related papers  (2024-05-10T21:17:31Z)
• Predicting risk of cardiovascular disease using retinal OCT imaging [40.71667870702634]
  Cardiovascular diseases (CVD) are the leading cause of death globally.<n>Optical coherence tomography ( OCT) has gained recognition as a potential tool for early CVD risk prediction.<n>We investigated the potential of OCT as an additional imaging technique to predict future CVD events.
  arXiv  Detail & Related papers  (2024-03-26T14:42:46Z)
• Affinity Feature Strengthening for Accurate, Complete and Robust Vessel Segmentation [48.638327652506284]
  Vessel segmentation is crucial in many medical image applications, such as detecting coronary stenoses, retinal vessel diseases and brain aneurysms. We present a novel approach, the affinity feature strengthening network (AFN), which jointly models geometry and refines pixel-wise segmentation features using a contrast-insensitive, multiscale affinity approach.
  arXiv  Detail & Related papers  (2022-11-12T05:39:17Z)
• CACTUSS: Common Anatomical CT-US Space for US examinations [36.45569352490318]
  Abdominal aortic aneurysm (AAA) is a vascular disease in which a section of the aorta enlarges, weakening its walls and potentially rupturing the vessel. Recent abdominal CT datasets have been successfully utilized to train deep neural networks for automatic aorta segmentation. CACTUSS acts as a virtual bridge between CT and US modalities to enable automatic AAA screening sonography.
  arXiv  Detail & Related papers  (2022-07-18T14:05:25Z)
• Cross-Site Severity Assessment of COVID-19 from CT Images via Domain Adaptation [64.59521853145368]
  Early and accurate severity assessment of Coronavirus disease 2019 (COVID-19) based on computed tomography (CT) images offers a great help to the estimation of intensive care unit event. To augment the labeled data and improve the generalization ability of the classification model, it is necessary to aggregate data from multiple sites. This task faces several challenges including class imbalance between mild and severe infections, domain distribution discrepancy between sites, and presence of heterogeneous features.
  arXiv  Detail & Related papers  (2021-09-08T07:56:51Z)
• Assignment Flow for Order-Constrained OCT Segmentation [0.0]
  The identification of retinal layer thicknesses serves as an essential task be done for each patient separately. The elaboration of automated segmentation models has become an important task in the field of medical image processing. We propose a novel, purely data driven textitgeometric approach to order-constrained 3D OCT retinal cell layer segmentation
  arXiv  Detail & Related papers  (2020-09-10T01:57:53Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.