MedSegNet10: A Publicly Accessible Network Repository for Split Federated Medical Image Segmentation

2025-03-26

AI Summary

Medical image segmentation faces challenges including data privacy sensitivity, scarcity of expert annotations, and difficulties in cross-institutional collaboration. To address these, we propose MedSegNet10—the first open-source SplitFed framework specifically designed for medical image segmentation—uniquely integrating Split Learning with Federated Learning to enable end-to-end collaborative training without exposing raw data outside local domains. The framework supports heterogeneous multimodal data (e.g., embryonic microscopy, dermoscopy, endoscopy) and facilitates multi-center joint modeling while strictly preserving data sovereignty. Built on PyTorch and U-Net variants, it incorporates standardized multi-site preprocessing pipelines and adaptive federated aggregation strategies. MedSegNet10 provides over ten pre-trained models covering three clinical tasks. Under strict data isolation, it achieves Dice scores exceeding 92% of those attained by centralized training—demonstrating strong efficacy and significantly advancing privacy-preserving, collaborative AI in medicine.

Technology Category

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

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Federated Web and WoT systems, including distributed, federated and edge-based data processing
• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization
• Security and Privacy: Data transparency and provenance

Abstract

Machine Learning (ML) and Deep Learning (DL) have shown significant promise in healthcare, particularly in medical image segmentation, which is crucial for accurate disease diagnosis and treatment planning. Despite their potential, challenges such as data privacy concerns, limited annotated data, and inadequate training data persist. Decentralized learning approaches such as federated learning (FL), split learning (SL), and split federated learning (SplitFed/SFL) address these issues effectively. This paper introduces"MedSegNet10,"a publicly accessible repository designed for medical image segmentation using split-federated learning. MedSegNet10 provides a collection of pre-trained neural network architectures optimized for various medical image types, including microscopic images of human blastocysts, dermatoscopic images of skin lesions, and endoscopic images of lesions, polyps, and ulcers, with applications extending beyond these examples. By leveraging SplitFed's benefits, MedSegNet10 allows collaborative training on privately stored, horizontally split data, ensuring privacy and integrity. This repository supports researchers, practitioners, trainees, and data scientists, aiming to advance medical image segmentation while maintaining patient data privacy. The repository is available at: https://vault.sfu.ca/index.php/s/ryhf6t12O0sobuX (password upon request to the authors).

Problem

• Addresses privacy and data limitations in medical image segmentation
• Provides pre-trained models for diverse medical image types
• Enables collaborative training with decentralized data via SplitFed

Innovation

• Split-federated learning for medical image segmentation
• Public repository with pre-trained neural networks
• Collaborative training on private, horizontally split data