Naumova Klavdiia, Devos Arnout, Karimireddy Sai Praneeth, Jaggi Martin, Hartley Mary-Anne
Laboratory for Intelligent Global Health and Humanitarian Response Technologies (LiGHT), Machine Learning and Optimization Laboratory, Swiss Federal Institute of Technology Lausanne (EPFL), Lausanne, Switzerland.
ETH AI Center, Swiss Federal Institute of Technology Zurich (ETH Zurich), Zurich, Switzerland.
NPJ Digit Med. 2024 Sep 7;7(1):238. doi: 10.1038/s41746-024-01226-1.
Distributed collaborative learning is a promising approach for building predictive models for privacy-sensitive biomedical images. Here, several data owners (clients) train a joint model without sharing their original data. However, concealed systematic biases can compromise model performance and fairness. This study presents MyThisYourThat (MyTH) approach, which adapts an interpretable prototypical part learning network to a distributed setting, enabling each client to visualize feature differences learned by others on their own image: comparing one client's 'This' with others' 'That'. Our setting demonstrates four clients collaboratively training two diagnostic classifiers on a benchmark X-ray dataset. Without data bias, the global model reaches 74.14% balanced accuracy for cardiomegaly and 74.08% for pleural effusion. We show that with systematic visual bias in one client, the performance of global models drops to near-random. We demonstrate how differences between local and global prototypes reveal biases and allow their visualization on each client's data without compromising privacy.
分布式协作学习是一种很有前途的方法，用于为隐私敏感的生物医学图像构建预测模型。在这里，几个数据所有者（客户端）在不共享其原始数据的情况下训练一个联合模型。然而，隐藏的系统偏差可能会损害模型性能和公平性。本研究提出了MyThisYourThat（MyTH）方法，该方法将一个可解释的原型部分学习网络应用于分布式设置，使每个客户端能够在自己的图像上可视化其他客户端学到的特征差异：将一个客户端的“这个”与其他客户端的“那个”进行比较。我们的设置展示了四个客户端在一个基准X射线数据集上协作训练两个诊断分类器。在没有数据偏差的情况下，全局模型对心脏肥大的平衡准确率达到74.14%，对胸腔积液的平衡准确率达到74.08%。我们表明，在一个客户端存在系统视觉偏差的情况下，全局模型的性能会降至接近随机水平。我们展示了局部和全局原型之间的差异如何揭示偏差，并允许在不损害隐私的情况下在每个客户端的数据上进行可视化。
