Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
156 tokens/sec
GPT-4o
7 tokens/sec
Gemini 2.5 Pro Pro
45 tokens/sec
o3 Pro
4 tokens/sec
GPT-4.1 Pro
38 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

Advances in Kidney Biopsy Lesion Assessment through Dense Instance Segmentation (2309.17166v3)

Published 29 Sep 2023 in cs.CV and cs.AI

Abstract: Renal biopsies are the gold standard for the diagnosis of kidney diseases. Lesion scores made by renal pathologists are semi-quantitative and exhibit high inter-observer variability. Automating lesion classification within segmented anatomical structures can provide decision support in quantification analysis, thereby reducing inter-observer variability. Nevertheless, classifying lesions in regions-of-interest (ROIs) is clinically challenging due to (a) a large amount of densely packed anatomical objects, (b) class imbalance across different compartments (at least 3), (c) significant variation in size and shape of anatomical objects and (d) the presence of multi-label lesions per anatomical structure. Existing models cannot address these complexities in an efficient and generic manner. This paper presents an analysis for a \textbf{generalized solution} to datasets from various sources (pathology departments) with different types of lesions. Our approach utilizes two sub-networks: dense instance segmentation and lesion classification. We introduce \textbf{DiffRegFormer}, an end-to-end dense instance segmentation sub-network designed for multi-class, multi-scale objects within ROIs. Combining diffusion models, transformers, and RCNNs, DiffRegFormer {is a computational-friendly framework that can efficiently recognize over 500 objects across three anatomical classes, i.e., glomeruli, tubuli, and arteries, within ROIs.} In a dataset of 303 ROIs from 148 Jones' silver-stained renal Whole Slide Images (WSIs), our approach outperforms previous methods, achieving an Average Precision of 52.1\% (detection) and 46.8\% (segmentation). Moreover, our lesion classification sub-network achieves 89.2\% precision and 64.6\% recall on 21889 object patches out of the 303 ROIs. Lastly, our model demonstrates direct domain transfer to PAS-stained renal WSIs without fine-tuning.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (50)
  1. Recent advances in medical image processing for the evaluation of chronic kidney disease. Medical Image Analysis 69, 101960.
  2. Deep learning–based segmentation and quantification in experimental kidney histopathology. Journal of the American Society of Nephrology: JASN 32, 52.
  3. Renal biopsy practice: What is the gold standard? World journal of nephrology 3, 287.
  4. Cascade r-cnn: High quality object detection and instance segmentation. IEEE transactions on pattern analysis and machine intelligence 43, 1483–1498.
  5. End-to-end object detection with transformers, in: European conference on computer vision, Springer. pp. 213–229.
  6. Regionvit: Regional-to-local attention for vision transformers. arXiv preprint arXiv:2106.02689 .
  7. Mmdetection: Open mmlab detection toolbox and benchmark. arXiv preprint arXiv:1906.07155 .
  8. Diffusiondet: Diffusion model for object detection. arXiv preprint arXiv:2211.09788 .
  9. A generalist framework for panoptic segmentation of images and videos. arXiv preprint arXiv:2210.06366 .
  10. Analog bits: Generating discrete data using diffusion models with self-conditioning. arXiv preprint arXiv:2208.04202 .
  11. Dynamic convolution: Attention over convolution kernels, in: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pp. 11030–11039.
  12. Masked-attention mask transformer for universal image segmentation, in: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pp. 1290–1299.
  13. Per-pixel classification is not all you need for semantic segmentation. Advances in Neural Information Processing Systems 34, 17864–17875.
  14. Sparse instance activation for real-time instance segmentation, in: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 4433–4442.
  15. Omni-seg: A scale-aware dynamic network for renal pathological image segmentation. IEEE Transactions on Biomedical Engineering .
  16. Diffusion models beat gans on image synthesis. Advances in neural information processing systems 34, 8780–8794.
  17. Instances as queries. arXiv:2105.01928.
  18. Fast r-cnn. arXiv:1504.08083.
  19. Digital Image Processing (3rd Edition). Prentice-Hall, Inc., USA.
  20. Diffusioninst: Diffusion model for instance segmentation. arXiv preprint arXiv:2212.02773 .
  21. Mask r-cnn, in: Proceedings of the IEEE international conference on computer vision, pp. 2961--2969.
  22. Deep residual learning for image recognition. arXiv:1512.03385.
  23. Deep learning--based histopathologic assessment of kidney tissue. Journal of the American Society of Nephrology: JASN 30, 1968.
  24. Denoising diffusion probabilistic models. Advances in neural information processing systems 33, 6840--6851.
  25. Stochastic geometric mechanics in nonequilibrium thermodynamics: Schrödinger meets onsager. Journal of Physics A: Mathematical and Theoretical 56, 134003.
  26. Mask scoring r-cnn. arXiv:1903.00241.
  27. Instance segmentation for whole slide imaging: end-to-end or detect-then-segment. Journal of Medical Imaging 8, 014001--014001.
  28. A deep learning-based approach for glomeruli instance segmentation from multistained renal biopsy pathologic images. The American Journal of Pathology 191, 1431--1441.
  29. Dn-detr: Accelerate detr training by introducing query denoising, in: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 13619--13627.
  30. Mask dino: Towards a unified transformer-based framework for object detection and segmentation, in: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 3041--3050.
  31. Diffusion-lm improves controllable text generation. Advances in Neural Information Processing Systems 35, 4328--4343.
  32. Feature pyramid networks for object detection. arXiv:1612.03144.
  33. Microsoft coco: Common objects in context, in: Computer Vision--ECCV 2014: 13th European Conference, Zurich, Switzerland, September 6-12, 2014, Proceedings, Part V 13, Springer. pp. 740--755.
  34. Diagnosis of diabetic kidney disease in whole slide images via ai-driven quantification of pathological indicators. Computers in Biology and Medicine 166, 107470.
  35. Learning in implicit generative models. arXiv preprint arXiv:1610.03483 .
  36. Improved denoising diffusion probabilistic models, in: International Conference on Machine Learning, PMLR. pp. 8162--8171.
  37. Grad-tts: A diffusion probabilistic model for text-to-speech. arXiv:2105.06337.
  38. Faster r-cnn: Towards real-time object detection with region proposal networks. arXiv:1506.01497.
  39. High-resolution image synthesis with latent diffusion models, in: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pp. 10684--10695.
  40. Automated assessment of glomerulosclerosis and tubular atrophy using deep learning. Computerized Medical Imaging and Graphics 90, 101930.
  41. Spatially aware transformer networks for contextual prediction of diabetic nephropathy progression from whole slide images, in: Medical Imaging 2023: Digital and Computational Pathology, SPIE. pp. 129--140.
  42. Deep unsupervised learning using nonequilibrium thermodynamics, in: International conference on machine learning, PMLR. pp. 2256--2265.
  43. Denoising diffusion implicit models. arXiv preprint arXiv:2010.02502 .
  44. Deep neural network models for computational histopathology: A survey. Medical image analysis 67, 101813.
  45. What makes for end-to-end object detection?, in: International Conference on Machine Learning, PMLR. pp. 9934--9944.
  46. Attention is all you need. Advances in neural information processing systems 30.
  47. Deep learning of feature representation with multiple instance learning for medical image analysis, in: 2014 IEEE international conference on acoustics, speech and signal processing (ICASSP), IEEE. pp. 1626--1630.
  48. Devil is in the queries: advancing mask transformers for real-world medical image segmentation and out-of-distribution localization, in: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 23879--23889.
  49. Dino: Detr with improved denoising anchor boxes for end-to-end object detection. arXiv preprint arXiv:2203.03605 .
  50. Deformable detr: Deformable transformers for end-to-end object detection. arXiv preprint arXiv:2010.04159 .

Summary

We haven't generated a summary for this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Summarize Now