FedGIN: Federated Learning with Dynamic Global Intensity Non-linear Augmentation for Organ Segmentation using Multi-modal Images

2025-08-07

AI Summary

Addressing the challenges of cross-center and cross-modality generalization in multi-modal medical image (e.g., CT/MRI) organ segmentation—stemming from large inter-modal intensity discrepancies, scarce annotated data, and stringent privacy constraints—this paper proposes FedGIN, a novel federated learning framework. Its core innovation is a lightweight Global Intensity Nonlinear augmentation module (GIN), deployed locally to dynamically harmonize inter-modal intensity distributions without sharing raw data, thereby mitigating domain shift and data silos. Integrated with 3D Dice optimization, FedGIN significantly enhances robustness in cross-modality segmentation under federated settings: it achieves 12–18% Dice improvement on MRI test sets under data-limited conditions; with full data, it outperforms unimodal baselines by up to 30%, approaching centralized training performance.

Technology Category

• Computer Vision: Multi-modal Vision
• Machine Learning: Distributed Machine Learning & Federated Learning
• Natural Language Processing: Language Grounding & Multi-modal NLP

Application Category

• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization
• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Systems and Infrastructure for Web, Mobile and WoT: Federated Web and WoT systems, including distributed, federated and edge-based data processing

Abstract

Medical image segmentation plays a crucial role in AI-assisted diagnostics, surgical planning, and treatment monitoring. Accurate and robust segmentation models are essential for enabling reliable, data-driven clinical decision making across diverse imaging modalities. Given the inherent variability in image characteristics across modalities, developing a unified model capable of generalizing effectively to multiple modalities would be highly beneficial. This model could streamline clinical workflows and reduce the need for modality-specific training. However, real-world deployment faces major challenges, including data scarcity, domain shift between modalities (e.g., CT vs. MRI), and privacy restrictions that prevent data sharing. To address these issues, we propose FedGIN, a Federated Learning (FL) framework that enables multimodal organ segmentation without sharing raw patient data. Our method integrates a lightweight Global Intensity Non-linear (GIN) augmentation module that harmonizes modality-specific intensity distributions during local training. We evaluated FedGIN using two types of datasets: an imputed dataset and a complete dataset. In the limited dataset scenario, the model was initially trained using only MRI data, and CT data was added to assess its performance improvements. In the complete dataset scenario, both MRI and CT data were fully utilized for training on all clients. In the limited-data scenario, FedGIN achieved a 12 to 18% improvement in 3D Dice scores on MRI test cases compared to FL without GIN and consistently outperformed local baselines. In the complete dataset scenario, FedGIN demonstrated near-centralized performance, with a 30% Dice score improvement over the MRI-only baseline and a 10% improvement over the CT-only baseline, highlighting its strong cross-modality generalization under privacy constraints.

Problem

• Develops a unified model for multi-modal organ segmentation
• Addresses data scarcity and domain shift in medical imaging
• Ensures privacy by avoiding raw data sharing in federated learning

Innovation

• Federated Learning for multimodal organ segmentation
• Dynamic Global Intensity Non-linear augmentation module
• Improved cross-modality generalization under privacy constraints