Multi-task Federated Learning for Heterogeneous Pancreas Segmentation

TL;DR

This paper explores federated learning for pancreas segmentation across multiple institutions, addressing challenges of heterogeneous data with different labels, and proposes optimization methods to improve segmentation accuracy in multi-task settings.

Contribution

It introduces heterogeneous optimization techniques tailored for multi-task federated learning in medical image segmentation, enhancing model performance across diverse datasets.

Findings

Improved segmentation accuracy for pancreas and tumors.

Heterogeneous optimization outperforms standard federated averaging.

Effective handling of multi-task data heterogeneity.

Abstract

Federated learning (FL) for medical image segmentation becomes more challenging in multi-task settings where clients might have different categories of labels represented in their data. For example, one client might have patient data with "healthy'' pancreases only while datasets from other clients may contain cases with pancreatic tumors. The vanilla federated averaging algorithm makes it possible to obtain more generalizable deep learning-based segmentation models representing the training data from multiple institutions without centralizing datasets. However, it might be sub-optimal for the aforementioned multi-task scenarios. In this paper, we investigate heterogeneous optimization methods that show improvements for the automated segmentation of pancreas and pancreatic tumors in abdominal CT images with FL settings.