Papers
Topics
Authors
Recent
Search
2000 character limit reached

FairSkin: Fair Diffusion for Skin Disease Image Generation

Published 29 Oct 2024 in cs.CV | (2410.22551v2)

Abstract: Image generation is a prevailing technique for clinical data augmentation for advancing diagnostic accuracy and reducing healthcare disparities. Diffusion Model (DM) has become a leading method in generating synthetic medical images, but it suffers from a critical twofold bias: (1) The quality of images generated for Caucasian individuals is significantly higher, as measured by the Frechet Inception Distance (FID). (2) The ability of the downstream-task learner to learn critical features from disease images varies across different skin tones. These biases pose significant risks, particularly in skin disease detection, where underrepresentation of certain skin tones can lead to misdiagnosis or neglect of specific conditions. To address these challenges, we propose FairSkin, a novel DM framework that mitigates these biases through a three-level resampling mechanism, ensuring fairer representation across racial and disease categories. Our approach significantly improves the diversity and quality of generated images, contributing to more equitable skin disease detection in clinical settings.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (40)
  1. A reductions approach to fair classification. ArXiv, abs/1803.02453, 2018. URL https://api.semanticscholar.org/CorpusID:4725675.
  2. Diffusion-based data augmentation for skin disease classification: Impact across original medical datasets to fully synthetic images. In International Conference on Medical Image Computing and Computer-Assisted Intervention, pp.  99–109. Springer, 2023.
  3. Racial disparities in skin tone representation of dermatomyositis rashes: a systematic review. Rheumatology, 61(6):2255–2261, 2022.
  4. Inspecting the geographical representativeness of images from text-to-image models. In Proceedings of the IEEE/CVF International Conference on Computer Vision, pp.  5136–5147, 2023.
  5. Dm–ahr: A self-supervised conditional diffusion model for ai-generated hairless imaging for enhanced skin diagnosis applications. Cancers, 16(17):2947, 2024.
  6. Easily accessible text-to-image generation amplifies demographic stereotypes at large scale. In Proceedings of the 2023 ACM Conference on Fairness, Accountability, and Transparency, pp.  1493–1504, 2023.
  7. Generation of clinical skin images with pathology with scarce data. In AI for Health Equity and Fairness: Leveraging AI to Address Social Determinants of Health, pp.  47–64. Springer, 2024.
  8. The impact of generative conversational artificial intelligence on the lesbian, gay, bisexual, transgender, and queer community: scoping review. Journal of Medical Internet Research, 25:e52091, 2023.
  9. Attri-vae: Attribute-based interpretable representations of medical images with variational autoencoders. Computerized Medical Imaging and Graphics, 104:102158, 2023.
  10. Zirui Chen. Diffusion models-based data augmentation for the cell cycle phase classification. In Journal of Physics: Conference Series, pp.  012001. IOP Publishing, 2023.
  11. Conditional diffusion models for semantic 3d brain mri synthesis. IEEE Journal of Biomedical and Health Informatics, 2024.
  12. Synthetic data augmentation using gan for improved liver lesion classification. In 2018 IEEE 15th international symposium on biomedical imaging (ISBI 2018), pp.  289–293. IEEE, 2018.
  13. Evaluating deep neural networks trained on clinical images in dermatology with the fitzpatrick 17k dataset, 2021. URL https://arxiv.org/abs/2104.09957.
  14. Deep learning-aided decision support for diagnosis of skin disease across skin tones. Nature Medicine, 30(2):573–583, 2024.
  15. Debiasing text-to-image diffusion models. arXiv preprint arXiv:2402.14577, 2024.
  16. Gans trained by a two time-scale update rule converge to a local nash equilibrium. Advances in neural information processing systems, 30, 2017.
  17. Med-cdiff: Conditional medical image generation with diffusion models. Bioengineering, 10(11):1258, 2023.
  18. Universal debiased editing for fair medical image classification. arXiv preprint arXiv:2403.06104, 2024.
  19. Diffusion models in medical imaging: A comprehensive survey. Medical Image Analysis, 88:102846, 2023.
  20. Technology innovation to reduce health inequality in skin diagnosis and to improve patient outcomes for people of color: a thematic literature review and future research agenda. Frontiers in Artificial Intelligence, 7:1394386, 2024.
  21. Holistic evaluation of text-to-image models. Advances in Neural Information Processing Systems, 36, 2024.
  22. Fair text-to-image diffusion via fair mapping. arXiv preprint arXiv:2311.17695, 2023.
  23. Scoft: Self-contrastive fine-tuning for equitable image generation. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp.  10822–10832, 2024.
  24. Structure and illumination constrained gan for medical image enhancement. IEEE Transactions on Medical Imaging, 40(12):3955–3967, 2021.
  25. Towards ethical dermatology: Mitigating bias in skin condition classification. In 2024 International Joint Conference on Neural Networks (IJCNN), pp.  1–8. IEEE, 2024.
  26. Class-balancing diffusion models. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp.  18434–18443, 2023.
  27. Medical image generation techniques for data augmentation: Disc-vae versus gan. In 2024 6th International Conference on Pattern Analysis and Intelligent Systems (PAIS), pp.  1–8. IEEE, 2024.
  28. Zero-shot text-to-image generation. In International conference on machine learning, pp.  8821–8831. Pmlr, 2021.
  29. High-resolution image synthesis with latent diffusion models. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pp.  10684–10695, June 2022a.
  30. High-resolution image synthesis with latent diffusion models. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pp.  10684–10695, 2022b.
  31. Fairseg: A large-scale medical image segmentation dataset for fairness learning using segment anything model with fair error-bound scaling. In The Twelfth International Conference on Learning Representations, 2024. URL https://openreview.net/forum?id=qNrJJZAKI3.
  32. Modelling the distribution of 3d brain mri using a 2d slice vae. In Medical Image Computing and Computer Assisted Intervention–MICCAI 2020: 23rd International Conference, Lima, Peru, October 4–8, 2020, Proceedings, Part VII 23, pp.  657–666. Springer, 2020.
  33. Gpt-4: a new era of artificial intelligence in medicine. Irish Journal of Medical Science (1971-), 192(6):3197–3200, 2023.
  34. The male ceo and the female assistant: Probing gender biases in text-to-image models through paired stereotype test. arXiv preprint arXiv:2402.11089, 2024.
  35. T2iat: Measuring valence and stereotypical biases in text-to-image generation. arXiv preprint arXiv:2306.00905, 2023.
  36. Visual transformers: Token-based image representation and processing for computer vision, 2020.
  37. The limits of fair medical imaging ai in real-world generalization. Nature Medicine, pp.  1–11, 2024.
  38. Diverse diffusion: Enhancing image diversity in text-to-image generation. arXiv preprint arXiv:2310.12583, 2023.
  39. Adding conditional control to text-to-image diffusion models. In Proceedings of the IEEE/CVF International Conference on Computer Vision, pp.  3836–3847, 2023.
  40. Bias in generative ai. arXiv preprint arXiv:2403.02726, 2024.

Summary

No one has generated a summary of this paper yet.

Paper to Video (Beta)

No one has generated a video about this paper yet.

Whiteboard

No one has generated a whiteboard explanation for this paper yet.

Open Problems

We haven't generated a list of open problems mentioned in this paper yet.

Continue Learning

We haven't generated follow-up questions for this paper yet.

Collections

Sign up for free to add this paper to one or more collections.

Tweets

Sign up for free to view the 1 tweet with 0 likes about this paper.