Improving multi-organ CT scans with shared teaching in federated learning
FedMust: Semi-supervised Multi-task Student-Teacher Federated Learning for Multi-organ CT Segmentation
Computer Vision and Pattern RecognitionArtificial Intelligence
Summary
Medical centers often want to use artificial intelligence to outline organs in CT scans but have limited labeled data and cannot share patient data due to privacy. The authors propose a learning method that lets hospitals teach each other’s AI models without sharing raw data, combining knowledge from labeled and unlabeled scans. Their system uses two steps: first, hospitals with similar tasks jointly create teacher models, then all hospitals build a student model that learns from the teachers and handles multiple organ tasks together. This approach improved organ identification accuracy across different hospitals by about 13 percent.
What this means in practice
- •For medical imaging teams: Create multi-organ segmentation models using data from multiple hospitals while keeping patient data private.
- •For clinical ai developers: Reduce the amount of manually labeled CT scans needed to train accurate multi-organ segmentation models.
- •For industrial ai builders: Train federated multi-task models that handle incomplete labels by combining teacher and student networks across sites.$Commercial implications: Enables development of privacy-preserving AI products that improve efficiency and accuracy in medical image analysis.
Authors
Ashkan Moradi, Bendik Skarre Abrahamsen, Mattijs Elschot
Abstract
Multi-organ segmentation using deep learning requires large amounts of annotated patient data; however, institutions often lack sufficiently large and diverse annotated datasets. Privacy constraints further prevent institutions from sharing patient data to overcome this limitation. Moreover, due to the labor-intensive nature of annotation and the scarcity of diverse expertise, institutions typically have labels for only a small portion of their local data, leaving the larger unlabeled portion unused. In this work, we propose a flexible semi-supervised federated multi-task student-teacher framework that leverages federated learning (FL) to improve multi-organ segmentation using both labeled and unlabeled data across participating sites. At each communication round, the proposed framework initiates local training, where clients with labels for the same task form a federation to produce an aggregated teacher model. The resulting teachers generate task-specific features for all data at each client. Subsequently, all clients form a second federation to train a multi-task student model with a shared encoder and task-specific decoders that replicate the teacher-generated features across all segmentation tasks. The aggregated student model is then used to update the local teachers and initiate the next training round. Extensive experiments demonstrated the effectiveness of the proposed method compared with local and federated single-organ models, yielding an average performance gain of 13 percent across clients. The experiments also demonstrated the impact of multi-task learning and unlabeled data and the applicability of the framework in relaxing labeled-data requirements for client participation. The code is available at https://github.com/AshknMrd/FedMust.