HMS-VesselNet: Hierarchical Multi-Scale Attention Network with Topology-Preserving Loss for Retinal Vessel Segmentation

• URL: http://arxiv.org/abs/2603.21891v1
• Date: Mon, 23 Mar 2026 12:16:45 GMT
• Title: HMS-VesselNet: Hierarchical Multi-Scale Attention Network with Topology-Preserving Loss for Retinal Vessel Segmentation
• Abstract summary: HMS-VesselNet is a hierarchical multi-scale network that processes fundus images across four parallel branches at different resolutions.<n>Tested on 68 images from DRIVE, STARE, and CHASE_DB1 using 5-fold cross-validation, the model achieves a mean Dice of 88.72 +/- 0.67%, Sensitivity of 90.78 +/- 1.42%, and AUC of 98.25 +/- 0.21%.
• Score: 0.0
• License: http://creativecommons.org/licenses/by-nc-nd/4.0/
• Abstract: Retinal vessel segmentation methods based on standard overlap losses tend to miss thin peripheral vessels because these structures occupy very few pixels and have low contrast against the background. We propose HMS-VesselNet, a hierarchical multi-scale network that processes fundus images across four parallel branches at different resolutions and combines their outputs using learned fusion weights. The training loss combines Dice, binary cross-entropy, and centerline Dice to jointly optimize area overlap and vessel continuity. Hard example mining is applied from epoch 20 onward to concentrate gradient updates on the most difficult training images. Tested on 68 images from DRIVE, STARE, and CHASE_DB1 using 5-fold cross-validation, the model achieves a mean Dice of 88.72 +/- 0.67%, Sensitivity of 90.78 +/- 1.42%, and AUC of 98.25 +/- 0.21%. In leave-one-dataset-out experiments, AUC remains above 95% on each unseen dataset. The largest improvement is in the recall of thin peripheral vessels, which are the structures most frequently missed by standard methods and most critical for early detection of diabetic retinopathy.

Related papers

• Unsupervised Anatomical Feature Learning via Diffusion Models: Enhanced Medical Image Segmentation with Denoising Diffusion Probabilistic Models [0.8590539335179757]
  This research paper proposes utilizing unsupervised Denoising Diffusion Probabilistic Models (DDPMs) to extract anatomical features.<n>We train a DDPM on 21 unlabeled abdominal CT scans to learn structural representations, transferring the encoder weights to a downstream segmentation task.<n>Using unlabeled images for diffusion-based pretraining successfully embeds robust anatomical features prior to human supervision.
  arXiv  Detail & Related papers  (2026-08-26T12:10:56Z)
• ESRVS: Extreme Semi-Supervised Retinal Vessel Segmentation with a Single Annotated Image [6.883947770846664]
  We study retinal vessel segmentation in an extreme semi-supervised setting with one annotated image and a pool of unlabeled images.<n>We propose ESRVS, which selects a representative reference image for manual annotation and transfers vessel cues using target-domain-adapted DINOv3 features.<n>With Mask2Former, ESRVS retains on average 93.7% of fully supervised Dice and 95.1% of fully supervised clDice.
  arXiv  Detail & Related papers  (2026-07-27T13:57:07Z)
• Robust automatic brain vessel segmentation in 3D CTA scans using dynamic 4D-CTA data [9.67851559401684]
  We develop a novel methodology for annotating the brain vasculature using dynamic 4D-CTA head scans.<n>We train deep learning models on our ground truth annotations by using the same segmentation for multiple phases from the dynamic 4D-CTA collection.<n>Our dataset comprises 110 training images from 25 patients and 165 test images from 14 patients.
  arXiv  Detail & Related papers  (2026-01-30T23:10:48Z)
• AortaDiff: A Unified Multitask Diffusion Framework For Contrast-Free AAA Imaging [7.769952833857425]
  We propose a unified deep learning framework that generates synthetic CECT images from non-contrast CT scans.<n>Our approach integrates conditional diffusion models (CDM) with multi-task learning, enabling end-to-end joint optimization of image synthesis and anatomical segmentation.
  arXiv  Detail & Related papers  (2025-10-01T22:19:27Z)
• A Novel Attention-Augmented Wavelet YOLO System for Real-time Brain Vessel Segmentation on Transcranial Color-coded Doppler [49.03919553747297]
  We propose an AI-powered, real-time CoW auto-segmentation system capable of efficiently capturing cerebral arteries.<n>No prior studies have explored AI-driven cerebrovascular segmentation using Transcranial Color-coded Doppler (TCCD)<n>The proposed AAW-YOLO demonstrated strong performance in segmenting both ipsilateral and contralateral CoW vessels.
  arXiv  Detail & Related papers  (2025-08-19T14:41:22Z)
• Novel Extraction of Discriminative Fine-Grained Feature to Improve Retinal Vessel Segmentation [18.968955885340797]
  We propose a novel Attention U-shaped Kolmogorov-Arnold Network named AttUKAN.<n>We implement Attention Gates into Kolmogorov-Arnold Networks to enhance model sensitivity.<n>We also design a novel Label-guided Pixel-wise Contrastive Loss to supervise our proposed AttUKAN.
  arXiv  Detail & Related papers  (2025-05-06T18:03:41Z)
• SMILE-UHURA Challenge -- Small Vessel Segmentation at Mesoscopic Scale from Ultra-High Resolution 7T Magnetic Resonance Angiograms [60.35639972035727]
  The lack of publicly available annotated datasets has impeded the development of robust, machine learning-driven segmentation algorithms. The SMILE-UHURA challenge addresses the gap in publicly available annotated datasets by providing an annotated dataset of Time-of-Flight angiography acquired with 7T MRI. Dice scores reached up to 0.838 $pm$ 0.066 and 0.716 $pm$ 0.125 on the respective datasets, with an average performance of up to 0.804 $pm$ 0.15.
  arXiv  Detail & Related papers  (2024-11-14T17:06:00Z)
• KaLDeX: Kalman Filter based Linear Deformable Cross Attention for Retina Vessel Segmentation [46.57880203321858]
  We propose a novel network (KaLDeX) for vascular segmentation leveraging a Kalman filter based linear deformable cross attention (LDCA) module. Our approach is based on two key components: Kalman filter (KF) based linear deformable convolution (LD) and cross-attention (CA) modules. The proposed method is evaluated on retinal fundus image datasets (DRIVE, CHASE_BD1, and STARE) as well as the 3mm and 6mm of the OCTA-500 dataset.
  arXiv  Detail & Related papers  (2024-10-28T16:00:42Z)
• 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)
• Automated segmentation of microvessels in intravascular OCT images using deep learning [0.0]
  We developed an automated deep learning method for detecting microvessels in intravascular optical coherence tomography (IV OCT) images. A total of 8,403 IV OCT image frames from 85 lesions and 37 normal segments were analyzed. Our method produced 698 image frames with microvessels present, compared to 730 from manual analysis, representing a 4.4% difference.
  arXiv  Detail & Related papers  (2022-10-01T02:14:14Z)
• Adaptation to CT Reconstruction Kernels by Enforcing Cross-domain Feature Maps Consistency [0.06117371161379209]
  We show a decrease in the COVID-19 segmentation quality of the model trained on the smooth and tested on the sharp reconstruction kernels. We propose the unsupervised adaptation method, called F-Consistency, that outperforms the previous approaches.
  arXiv  Detail & Related papers  (2022-03-28T10:00:03Z)
• Retinal Vessel Segmentation with Pixel-wise Adaptive Filters [47.8629995041574]
  We propose two novel methods to address the challenges of retinal vessel segmentation. First, we devise a light-weight module, named multi-scale residual similarity gathering (MRSG), to generate pixel-wise adaptive filters (PA-Filters) Second, we introduce a response cue erasing (RCE) strategy to enhance the segmentation accuracy.
  arXiv  Detail & Related papers  (2022-02-03T14:40:36Z)
• Vision Transformers for femur fracture classification [59.99241204074268]
  The Vision Transformer (ViT) was able to correctly predict 83% of the test images. Good results were obtained in sub-fractures with the largest and richest dataset ever.
  arXiv  Detail & Related papers  (2021-08-07T10:12:42Z)
• Classification of COVID-19 in CT Scans using Multi-Source Transfer Learning [91.3755431537592]
  We propose the use of Multi-Source Transfer Learning to improve upon traditional Transfer Learning for the classification of COVID-19 from CT scans. With our multi-source fine-tuning approach, our models outperformed baseline models fine-tuned with ImageNet. Our best performing model was able to achieve an accuracy of 0.893 and a Recall score of 0.897, outperforming its baseline Recall score by 9.3%.
  arXiv  Detail & Related papers  (2020-09-22T11:53:06Z)

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.