Whole-brain radiomics for clustered federated personalization in brain tumor segmentation

TL;DR

This paper introduces a novel federated personalization method for brain tumor segmentation that uses radiomic features and clustering to address inter and intra-institutional feature shifts caused by different scanners and acquisition parameters.

Contribution

The paper presents the first algorithm to account for both inter and intra-institution feature shifts using radiomic features and clustering in federated learning for brain tumor segmentation.

Findings

Improved segmentation accuracy on FeTS2022 dataset.

Effective handling of scanner and acquisition variability.

Enhanced convergence speed in federated training.

Abstract

Federated learning and its application to medical image segmentation have recently become a popular research topic. This training paradigm suffers from statistical heterogeneity between participating institutions' local datasets, incurring convergence slowdown as well as potential accuracy loss compared to classical training. To mitigate this effect, federated personalization emerged as the federated optimization of one model per institution. We propose a novel personalization algorithm tailored to the feature shift induced by the usage of different scanners and acquisition parameters by different institutions. This method is the first to account for both inter and intra-institution feature shift (multiple scanners used in a single institution). It is based on the computation, within each centre, of a series of radiomic features capturing the global texture of each 3D image volume, followed by a clustering analysis pooling all feature vectors transferred from the local institutions to the central server. Each computed clustered decentralized dataset (potentially including data from different institutions) then serves to finetune a global model obtained through classical federated learning. We validate our approach on the Federated Brain Tumor Segmentation 2022 Challenge dataset (FeTS2022). Our code is available at (https://github.com/MatthisManthe/radiomics_CFFL).