Multi-site fMRI Analysis Using Privacy-preserving Federated Learning and Domain Adaptation: ABIDE Results

By Xiaoxiao Li, Yufeng Gu, Nicha C. Dvornek, Lawrence H. Staib, Pamela Ventola, James S. Duncan. 2020

471 citations recorded by the source corpus.

Deep learning models have shown their advantage in many different tasks, including neuroimage analysis. However, to effectively train a high-quality deep learning model, the aggregation of a significant amount of patient information is required. The time and cost for acquisition and annotation in assembling, for example, large fMRI datasets make it difficult to acquire large numbers at a single site. However, due to the need to protect the privacy of patient data, it is hard to assemble a central database from multiple institutions. Federated learning allows for population-level models to be trained without centralizing entities’ data by transmitting the global model to local entities, training the model locally, and then averaging the gradients or weights in the global model. However, some studies suggest that private information can be recovered from the model gradients or weights. In this work, we address the problem of multi-site fMRI classification with a privacy-preserving strategy. To solve the problem, we propose a federated learning approach, where a decentralized iterative optimization algorithm is implemented and shared local model weights are altered by a randomization mechanism. Considering the systemic differences of fMRI distributions from different sites, we further propose two domain adaptation methods in this federated learning formulation. We investigate various practical aspects of federated model optimization and compare federated learning with alternative training strategies. Overall, our results demonstrate that it is promising to utilize multi-site data without data sharing to boost neuroimage analysis performance and find reliable disease-related biomarkers. Our proposed pipeline can be generalized to other privacy-sensitive medical data analysis problems. Our code is publicly available at: https://github.com/xxlya/Fed_ABIDE/.

Subfields: Artificial Intelligence, Radiology, Nuclear Medicine and Imaging

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Radiology, Nuclear Medicine and Imaging

Citations in the corpus

The future of digital health with federated learning (2020, background): “Countermeasures, such as limiting the granularity of the shared model updates and to add specific noise to ensure differential privacy [34, 12, 57] may be needed and is still an active area of research [14].”
Federated Learning for Healthcare: Systematic Review and Architecture Proposal (2022, methodology): “[80] propose an FL architecture that combines differential privacy and the mixture of experts.”
A review of deep learning in medical imaging: Imaging traits, technology trends, case studies with progress highlights, and future promises (2020, background): “Recently, FL is applied, together with domain adaptation, to train a model with boosted analysis performance and reliable discovery of diseaserelated biomarkers [86].”
Learn from Others and Be Yourself in Heterogeneous Federated Learning (2022, background): “Latest researches have studied related problems of unsupervised domain adaptation for target domain [45, 65] and domain generalization on unseen domains [52].”
FedDG: Federated Domain Generalization on Medical Image Segmentation via Episodic Learning in Continuous Frequency Space (2021, background): “Latest literature has studied a related problem of unsupervised domain adaptation in FL paradigm [28, 41], whereas these methods typically require data from the target domain to adapt the model.”
Shifting machine learning for healthcare from development to deployment and from models to data (2022, background): “Cross-silo federated learning is an increasingly attractive solution to the shortcomings of centralized training 71 .”
Federated learning for medical image analysis: A survey (2023, background): “And several recent studies [42], [67], [87] propose to add Gaussian random noise to the computed gradients on the patients’ imaging data in each client/site, thus protecting privacy from the server and other clients.”
Federated learning for predicting clinical outcomes in patients with COVID-19 (2021, methodology): “Through investigation of a partial weight-sharing scheme 50–52 , we showed that models can reach a comparable performance even when only 25% of weight updates are shared (Extended Data Fig.”
Federated learning and differential privacy for medical image analysis (2021, methodology): “In histopathology, Lu et al. 13 reported DP guarantees for a neural network classifier trained with FL, following Li et al. 26 .”
On the Challenges and Perspectives of Foundation Models for Medical Image Analysis (2023, background): “The initial stage involves selecting the appropriate foundation model (including general, modality-specific, and organ/task-specific models), depending on the application and available resources.”