mlsanityData to 7 Oct 2026

Unpaired cross-modality educed distillation (CMEDL) for medical image segmentation

arXiv 2107.07985 ↗First posted 16 Jul 2021eess.IV cs.CVcited by 2

Not on the map. Why?

Abstract (arXiv)

Accurate and robust segmentation of lung cancers from CT, even those located close to mediastinum, is needed to more accurately plan and deliver radiotherapy and to measure treatment response. Therefore, we developed a new cross-modality educed distillation (CMEDL) approach, using unpaired CT and MRI scans, whereby an informative teacher MRI network guides a student CT network to extract features that signal the difference between foreground and background. Our contribution eliminates two requirements of distillation methods: (i) paired image sets by using an image to image (I2I) translation and (ii) pre-training of the teacher network with a large training set by using concurrent training of all networks. Our framework uses an end-to-end trained unpaired I2I translation, teacher, and student segmentation networks. Architectural flexibility of our framework is demonstrated using 3 segmentation and 2 I2I networks. Networks were trained with 377 CT and 82 T2w MRI from different sets of patients, with independent validation (N=209 tumors) and testing (N=609 tumors) datasets. Network design, methods to combine MRI with CT information, distillation learning under informative (MRI to CT), weak (CT to MRI) and equal teacher (MRI to MRI), and ablation tests were performed. Accuracy was measured using Dice similarity (DSC), surface Dice (sDSC), and Hausdorff distance at the 95th percentile (HD95). The CMEDL approach was significantly (p < 0.001) more accurate (DSC of 0.77 vs. 0.73) than non-CMEDL methods with an informative teacher for CT lung tumor, with a weak teacher (DSC of 0.84 vs. 0.81) for MRI lung tumor, and with equal teacher (DSC of 0.90 vs. 0.88) for MRI multi-organ segmentation. CMEDL also reduced inter-rater lung tumor segmentation variabilities..

Key facts. Cited by 2 papers in the map (data through 7 Oct 2026). Builds on 34 of its 60 references that are in the map. First posted on arXiv 16 Jul 2021.

Builds on

34 of 60 references.

References · 34

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2015U-Net: Convolutional Networks for Biomedical Image Segmentationcited by 19,604
  3. 2014Very Deep Convolutional Networks for Large-Scale Image Recognitioncited by 15,267
  4. 2015Distilling the Knowledge in a Neural Networkcited by 10,076
  5. 2016Densely Connected Convolutional Networkscited by 6,691
  6. 2017Squeeze-and-Excitation Networkscited by 4,763
  7. 2014FitNets: Hints for Thin Deep Netscited by 1,619
  8. 2018UNet++: A Nested U-Net Architecture for Medical Image Segmentationcited by 1,518
  9. 2018Attention U-Net: Learning Where to Look for the Pancreascited by 1,420
  10. 2017Deep Mutual Learningcited by 621
  11. 2018Born Again Neural Networkscited by 539
  12. 2016RefineNet: Multi-Path Refinement Networks for High-Resolution Semantic Segmentationcited by 463
22 more
  1. 2018Diverse Image-to-Image Translation via Disentangled Representationscited by 296
  2. 2015Cross Modal Distillation for Supervision Transfercited by 173
  3. 2016Dual Learning for Machine Translationcited by 146
  4. 2018The Contextual Loss for Image Transformation with Non-Aligned Datacited by 135
  5. 2016The One Hundred Layers Tiramisu: Fully Convolutional DenseNets for Semantic Segmentationcited by 122
  6. 2018Snapshot Distillation: Teacher-Student Optimization in One Generationcited by 99
  7. 2018Translating and Segmenting Multimodal Medical Volumes with Cycle- and Shape-Consistency Generative Adversarial Networkcited by 97
  8. 2020Unpaired Multi-modal Segmentation via Knowledge Distillationcited by 81
  9. 2018Deep learning to achieve clinically applicable segmentation of head and neck anatomy for radiotherapycited by 76
  10. 2018Collaborative Learning for Deep Neural Networkscited by 69
  11. 2018SynSeg-Net: Synthetic Segmentation Without Target Modality Ground Truthcited by 68
  12. 2018HyperDense-Net: A hyper-densely connected CNN for multi-modal image segmentationcited by 66
  13. 2020Towards Cross-modality Medical Image Segmentation with Online Mutual Knowledge Distillationcited by 38
  14. 2017Automatic 3D Cardiovascular MR Segmentation with Densely-Connected Volumetric ConvNetscited by 32
  15. 2020Unsupervised Wasserstein Distance Guided Domain Adaptation for 3D Multi-Domain Liver Segmentationcited by 28
  16. 2020Dual-Teacher: Integrating Intra-domain and Inter-domain Teachers for Annotation-efficient Cardiac Segmentationcited by 24
  17. 2019Integrating cross-modality hallucinated MRI with CT to aid mediastinal lung tumor segmentationcited by 14
  18. 2020PSIGAN: Joint probabilistic segmentation and image distribution matching for unpaired cross-modality adaptation based MRI segmentationcited by 14
  19. 2019Soft labeling by Distilling Anatomical knowledge for Improved MS Lesion Segmentationcited by 11
  20. 2020KD-MRI: A knowledge distillation framework for image reconstruction and image restoration in MRI workflowcited by 9
  21. 2018CT Image Enhancement Using Stacked Generative Adversarial Networks and Transfer Learning for Lesion Segmentation Improvementcited by 5
  22. 2021Cross-Modality Brain Tumor Segmentation via Bidirectional Global-to-Local Unsupervised Domain Adaptationcited by 3

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2026: 2.

2 papers.

  1. 2026Clinical Graph-Mediated Distillation for Unpaired MRI-to-CFI Hypertension Predictioncited by 0
  2. 2026UnDA: Unpaired Domain Alignment for Cross-Modal Knowledge Transfer in Medical Imagingcited by 0