文献速递医学人工智能与信息学基础与转化研究

生成模型提升医学分类器在分布偏移下的公平性

Generative models improve fairness of medical classifiers under distribution shifts.
PubMed 文献 · Nat Med · 2024-04-10

该研究探讨生成式人工智能能否通过合成样本缓解医学机器学习中代表性不足的问题。研究使用扩散模型从数据中自动学习现实增强,并在组织病理学、胸部X线和皮肤科图像三个医学影像场景中评估。结果显示,用合成样本补充真实样本可提升模型在三个任务中的稳健性,并通过提高代表性不足组的临床诊断准确率改善公平性,尤其在分布外条件下。

要点

  • 扩散模型可从数据中以标签高效的方式自动学习现实增强。
  • 合成样本补充真实样本提升了三种医学影像任务的模型稳健性。
  • 合成样本补充真实样本提高了代表性不足组的诊断准确率,尤其在分布外条件下。

核心结论

  • 生成式人工智能可通过合成样本补充真实数据,以可控方式缓解代表性不足问题,使模型在分布内和分布外更稳健且统计上更公平。

研究设计与样本

  • 研究在组织病理学、胸部X线和皮肤科图像三个医学影像场景中评估扩散模型学习的增强对模型稳健性和公平性的影响。

主要发现

  • 用合成样本补充真实样本在所有三个医学任务中均提升了模型稳健性。
  • 用合成样本补充真实样本通过提高代表性不足组的临床诊断准确率改善了公平性,尤其在分布外条件下。

关键数值与推荐

  • 扩散模型能以标签高效的方式从数据中自动学习现实增强。

临床怎么用

  • 原文未据此给出临床推荐。

局限性

  • 原文未报告局限性。

以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。

原文

Domain generalization is a ubiquitous challenge for machine learning in healthcare. Model performance in real-world conditions might be lower than expected because of discrepancies between the data encountered during deployment and development. Underrepresentation of some groups or conditions during model development is a common cause of this phenomenon. This challenge is often not readily addressed by targeted data acquisition and 'labeling' by expert clinicians, which can be prohibitively expensive or practically impossible because of the rarity of conditions or the available clinical expertise. We hypothesize that advances in generative artificial intelligence can help mitigate this unmet need in a steerable fashion, enriching our training dataset with synthetic examples that address shortfalls of underrepresented conditions or subgroups. We show that diffusion models can automatically learn realistic augmentations from data in a label-efficient manner. We demonstrate that learned augmentations make models more robust and statistically fair in-distribution and out of distribution. To evaluate the generality of our approach, we studied three distinct medical imaging contexts of varying difficulty: (1) histopathology, (2) chest X-ray and (3) dermatology images. Complementing real samples with synthetic ones improved the robustness of models in all three medical tasks and increased fairness by improving the accuracy of clinical diagnosis within underrepresented groups, especially out of distribution.

来源:Nat Med,查看原文。版权归原作者所有。

引用

  1. Ktena I, Wiles O, Albuquerque I, et al. Generative models improve fairness of medical classifiers under distribution shifts. Nat Med. 2024. PMID: 38600282. doi:10.1038/s41591-024-02838-6

供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。