mlsanityData to 14 Oct 2026

ASTRA-Net: Anatomy-Specific Transfer and Representation Alignment for Drug-Induced Sleep Endoscopy Segmentation

arXiv 2607.21370 ↗First posted 23 Jul 2026cs.CVcited by 0

Not on the map. Why?

Abstract (arXiv)

Quantitative drug-induced sleep endoscopy (DISE) requires reliable airway boundaries at specific anatomical levels. Pixel-level DISE annotations are scarce, and manual contouring limits the scalability of quantitative assessment. To address this limitation, we developed ASTRA-Net for known-plane DISE segmentation with limited real annotations. Stage 1 aligned intermediate ConvNeXt-Base representations from 14,250 unlabeled virtual endoscopy frames derived from computed tomography and real DISE frames. Virtual images were used only for feature alignment. Stage 2 fine-tuned four independent UNet++ decoders on 401 real annotated frames. Structured zero-mask supervision constrained incompatible plane outputs and invalid frames. Six alignment configurations used maximum mean discrepancy, domain adversarial learning, or both objectives. On a hold-out evaluation set of 100 frames, the five-model MMD-only segmentation ensemble achieved a mean Dice of 0.8927, with a 95% image-level bootstrap interval of 0.8631 to 0.9160. The mean intersection over union was 0.8239. A classification- enabled variant of the same alignment configuration reached a restricted four-plane top-1 accuracy of 0.92 on the same hold-out frames. These results indicate that ASTRA-Net can support frame-level, plane-specific DISE boundary delineation when real annotations are limited.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 18 of its 33 references that are in the map. First posted on arXiv 23 Jul 2026.

Builds on

18 of 33 references.

References · 18

  1. 2015U-Net: Convolutional Networks for Biomedical Image Segmentationcited by 19,713
  2. 2020A Simple Framework for Contrastive Learning of Visual Representationscited by 12,049
  3. 2019Momentum Contrast for Unsupervised Visual Representation Learningcited by 7,513
  4. 2021Masked Autoencoders Are Scalable Vision Learnerscited by 7,403
  5. 2022A ConvNet for the 2020scited by 3,579
  6. 2015Domain-Adversarial Training of Neural Networkscited by 3,044
  7. 2021TransUNet: Transformers Make Strong Encoders for Medical Image Segmentationcited by 1,558
  8. 2018UNet++: A Nested U-Net Architecture for Medical Image Segmentationcited by 1,525
  9. 2017Domain Randomization for Transferring Deep Neural Networks from Simulation to the Real Worldcited by 1,498
  10. 2020PraNet: Parallel Reverse Attention Network for Polyp Segmentationcited by 460
  11. 2020Self-Supervision with Superpixels: Training Few-shot Medical Image Segmentation without Annotationcited by 128
  12. 2016Unsupervised domain adaptation in brain lesion segmentation with adversarial networkscited by 125
6 more
  1. 2019Synergistic Image and Feature Adaptation: Towards Cross-Modality Domain Adaptation for Medical Image Segmentationcited by 121
  2. 2022Video Polyp Segmentation: A Deep Learning Perspectivecited by 82
  3. 2021Progressively Normalized Self-Attention Network for Video Polyp Segmentationcited by 75
  4. 2020What Can Be Transferred: Unsupervised Domain Adaptation for Endoscopic Lesions Segmentationcited by 50
  5. 2023MAPSeg: Unified Unsupervised Domain Adaptation for Heterogeneous Medical Image Segmentation Based on 3D Masked Autoencoding and Pseudo-Labelingcited by 25
  6. 2024SimuScope: Realistic Endoscopic Synthetic Dataset Generation through Surgical Simulation and Diffusion Modelscited by 8

Built on by

Not cited yet.