Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
162 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

FourierLoss: Shape-Aware Loss Function with Fourier Descriptors (2309.12106v1)

Published 21 Sep 2023 in cs.CV

Abstract: Encoder-decoder networks become a popular choice for various medical image segmentation tasks. When they are trained with a standard loss function, these networks are not explicitly enforced to preserve the shape integrity of an object in an image. However, this ability of the network is important to obtain more accurate results, especially when there is a low-contrast difference between the object and its surroundings. In response to this issue, this work introduces a new shape-aware loss function, which we name FourierLoss. This loss function relies on quantifying the shape dissimilarity between the ground truth and the predicted segmentation maps through the Fourier descriptors calculated on their objects, and penalizing this dissimilarity in network training. Different than the previous studies, FourierLoss offers an adaptive loss function with trainable hyperparameters that control the importance of the level of the shape details that the network is enforced to learn in the training process. This control is achieved by the proposed adaptive loss update mechanism, which end-to-end learns the hyperparameters simultaneously with the network weights by backpropagation. As a result of using this mechanism, the network can dynamically change its attention from learning the general outline of an object to learning the details of its contour points, or vice versa, in different training epochs. Working on 2879 computed tomography images of 93 subjects, our experiments revealed that the proposed adaptive shape-aware loss function led to statistically significantly better results for liver segmentation, compared to its counterparts.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (30)
  1. Shape-aware deep convolutional neural network for vertebrae segmentation, in: Int. Workshop Comput. Methods and Clin. Appl. Musculoskelet. Imaging, Springer. pp. 12–24.
  2. Liver segmentation from computed tomography images using cascade deep learning. Comput. Biol. Med. 140, 105095.
  3. FourierNet: Shape-preserving network for Henle’s fiber Layer segmentation in Optical Coherence Tomography images. IEEE J. Biomed. Health. Inf. 27, 1036–1047.
  4. Bezierseg: Parametric shape representation for fast object segmentation in medical images. Life 13, 743.
  5. Transunet: Transformers make strong encoders for medical image segmentation. arXiv preprint arXiv:2102.04306 .
  6. Learning active contour models for medical image segmentation, in: CVPR, pp. 11632–11640.
  7. Channel-unet: a spatial channel-wise convolutional neural network for liver and tumors segmentation. Front. Genet. 10, 1110.
  8. Automatic liver and tumor segmentation of ct and mri volumes using cascaded fully convolutional neural networks. arXiv preprint arXiv:1702.05970 .
  9. Imagenet: A large-scale hierarchical image database, in: CVPR, Ieee. pp. 248–255.
  10. Deep residual learning for image recognition, in: CVPR, pp. 770–778.
  11. Fcsn: Global context aware segmentation by learning the fourier coefficients of objects in medical images. IEEE Journal of Biomedical and Health Informatics , 1–11.
  12. Reducing the hausdorff distance in medical image segmentation with convolutional neural networks. IEEE Trans. Med. Imaging 39, 499–513.
  13. Fully automated liver segmentation for low-and high-contrast ct volumes based on probabilistic atlases, in: IEEE Int. Conf. Image Process., IEEE. pp. 1733–1736.
  14. H-denseunet: hybrid densely connected unet for liver and tumor segmentation from ct volumes. IEEE Trans. Med. Imaging 37, 2663–2674.
  15. Abcnet: Real-time scene text spotting with adaptive bezier-curve network, in: CVPR, pp. 9809–9818.
  16. Automatic liver and tumour segmentation from ct images using deep learning algorithm. Results Control Optim. 6, 100087.
  17. Psi-net: Shape and boundary aware joint multi-task deep network for medical image segmentation, in: Proc. 41st Annu. Int. Conf. IEEE Eng. Med. Biol. Soc. (EMBC), IEEE. pp. 7223–7226.
  18. Fouriernet: Compact mask representation for instance segmentation using differentiable shape decoders, in: ICPR, IEEE. pp. 7833–7840.
  19. U-net: Convolutional networks for biomedical image segmentation, in: MICCAI, Springer. pp. 234–241.
  20. Segmentation of liver tumor in ct scan using resu-net. Appl. Sci. 12, 8650.
  21. Multiscale centerline detection by learning a scale-space distance transform, in: CVPR, pp. 2697–2704.
  22. Dsl: Automatic liver segmentation with faster r-cnn and deeplab, in: ICANN, Springer. pp. 137–147.
  23. Prior information guided regularized deep learning for cell nucleus detection. IEEE Trans. Med. Imaging 38, 2047–2058.
  24. Deep distance transform for tubular structure segmentation in ct scans, in: CVPR, pp. 3833–3842.
  25. Polarmask: Single shot instance segmentation with polar representation, in: CVPR, pp. 12193–12202.
  26. Explicit shape encoding for real-time instance segmentation, in: CVPR, pp. 5168–5177.
  27. Efficient multi-atlas abdominal segmentation on clinically acquired ct with simple context learning. Med. Image Anal. 24, 18–27.
  28. Shape-aware organ segmentation by predicting signed distance maps, in: Proc. AAAI Conf. Artif. Intell., pp. 12565–12572.
  29. Shapepu: A new pu learning framework regularized by global consistency for scribble supervised cardiac segmentation, in: MICCAI, Springer Nature Switzerland, Cham. pp. 162–172.
  30. Convolutional neural network with shape prior applied to cardiac mri segmentation. IEEE J. Biomed. Health. Inf. 23, 1119–1128.

Summary

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

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