Multi-site fMRI analysis using privacy-preserving federated learning and domain adaptation: ABIDE results

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

doi:10.1016/j.media.2020.101765

Cited by 272 publications

(192 citation statements)

References 37 publications

(48 reference statements)

Supporting

Mentioning

189

Contrasting

Unclassified

Order By: Relevance

“…2, a FL workflow can be realised with different topologies and compute plans. The two most common ones for healthcare applications are via an aggregation server [16][17][18] and peer to peer approaches 15,39 . In all cases, FL implicitly offers a certain degree of privacy, as FL participants never directly access data from other institutions and only receive model parameters that are aggregated over several participants.…”

Section: The Promise Of Federated Effortsmentioning

confidence: 99%

“…The applicability and advantages of FL have also been demonstrated in the field of medical imaging, for whole-brain segmentation in MRI 15 , as well as brain tumour segmentation 16,17 . Recently, the technique has been employed for fMRI classification to find reliable disease-related biomarkers 18 and suggested as a promising approach in the context of COVID-19 48 .…”

Section: The Promise Of Federated Effortsmentioning

confidence: 99%

“…Recent results, however, indicate that FL training is still feasible 59 , even if medical data is not uniformly distributed across the institutions 16,17 or includes a local bias 51 . Research addressing this problem includes, for example, FedProx 57 , part-data-sharing strategy 58 and FL with domainadaptation 18 . Another challenge is that data heterogeneity may lead to a situation in which the global optimal solution may not be optimal for an individual local participant.…”

Section: Technical Considerationsmentioning

confidence: 99%

“…Federated learning (FL) 9-11 is a learning paradigm seeking to address the problem of data governance and privacy by training algorithms collaboratively without exchanging the data itself. Originally developed for different domains, such as mobile and edge device use cases 12 , it recently gained traction for healthcare applications [13][14][15][16][17][18][19][20] . FL enables gaining insights collaboratively, e.g., in the form of a consensus model, without moving patient data beyond the firewalls of the institutions in which they reside.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

The future of digital health with federated learning

et al. 2020

View full text Add to dashboard Cite

Data-driven machine learning (ML) has emerged as a promising approach for building accurate and robust statistical models from medical data, which is collected in huge volumes by modern healthcare systems. Existing medical data is not fully exploited by ML primarily because it sits in data silos and privacy concerns restrict access to this data. However, without access to sufficient data, ML will be prevented from reaching its full potential and, ultimately, from making the transition from research to clinical practice. This paper considers key factors contributing to this issue, explores how federated learning (FL) may provide a solution for the future of digital health and highlights the challenges and considerations that need to be addressed.

show abstract

Section: The Promise Of Federated Effortsmentioning

confidence: 99%

Section: The Promise Of Federated Effortsmentioning

confidence: 99%

Section: Technical Considerationsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

The future of digital health with federated learning

et al. 2020

View full text Add to dashboard Cite

show abstract

“…However, authoritative views suggest that fully anonymized clinical imaging data can be used for research purposes with appropriate safeguards (The Royal College of Radiologists, 2017), and such use of data may be further facilitated by federated analyses of data, where the research pipelines are brought to the data (rather than vice versa), both for structural and functional imaging (X. Li et al, 2020;Silva et al, 2019).…”

Section: Leveraging Enigma To Address Challenges In Ams-tbi Researchmentioning

confidence: 99%

Toward a Global and Reproducible Science for Brain Imaging in Neurotrauma: The ENIGMA Adult Moderate/Severe Traumatic Brain Injury Working Group

Olsen¹,

Babikian²,

Bigler³

et al. 2019

Preprint

View full text Add to dashboard Cite

The global burden of mortality and morbidity caused by traumatic brain injury (TBI) is significant and the heterogeneity of TBI patients and the relatively small sample sizes of most current neuroimaging studies is a major challenge for scientific advances and clinical translation. The ENIGMA (Enhancing NeuroImaging Genetics through Meta-Analysis) Adult moderate/severe TBI (AMS-TBI) working group aims to be a driving force for new discoveries in AMS-TBI by providing researchers world-wide with an effective framework and platform for large-scale cross-border collaboration and data sharing. Based on the principles of transparency, rigor, reproducibility and collaboration, we will facilitate the development and dissemination of multiscale and big data analysis pipelines for harmonized analyses in AMS-TBI using structural and functional neuroimaging in combination with nonimaging biomarkers, genetics, as well as clinical and behavioral measures. Ultimately, we will offer investigators an unprecedented opportunity to test important hypotheses about recovery and morbidity in AMS-TBI by taking advantage of our robust methods for largescale neuroimaging data analysis. In this consensus statement we outline the working group’s short-term, intermediate, and long-term goals.

show abstract

Unsupervised contrastive graph learning for resting‐state functional MRI analysis and brain disorder detection

Wang,

Chu,

Wang

et al. 2023

Human Brain Mapping

View full text Add to dashboard Cite

Resting‐state functional magnetic resonance imaging (rs‐fMRI) helps characterize regional interactions that occur in the human brain at a resting state. Existing research often attempts to explore fMRI biomarkers that best predict brain disease progression using machine/deep learning techniques. Previous fMRI studies have shown that learning‐based methods usually require a large amount of labeled training data, limiting their utility in clinical practice where annotating data is often time‐consuming and labor‐intensive. To this end, we propose an unsupervised contrastive graph learning (UCGL) framework for fMRI‐based brain disease analysis, in which a pretext model is designed to generate informative fMRI representations using unlabeled training data, followed by model fine‐tuning to perform downstream disease identification tasks. Specifically, in the pretext model, we first design a bi‐level fMRI augmentation strategy to increase the sample size by augmenting blood‐oxygen‐level‐dependent (BOLD) signals, and then employ two parallel graph convolutional networks for fMRI feature extraction in an unsupervised contrastive learning manner. This pretext model can be optimized on large‐scale fMRI datasets, without requiring labeled training data. This model is further fine‐tuned on to‐be‐analyzed fMRI data for downstream disease detection in a task‐oriented learning manner. We evaluate the proposed method on three rs‐fMRI datasets for cross‐site and cross‐dataset learning tasks. Experimental results suggest that the UCGL outperforms several state‐of‐the‐art approaches in automated diagnosis of three brain diseases (i.e., major depressive disorder, autism spectrum disorder, and Alzheimer's disease) with rs‐fMRI data.

show abstract

Multi-site fMRI analysis using privacy-preserving federated learning and domain adaptation: ABIDE results

Cited by 272 publications

References 37 publications

The future of digital health with federated learning

The future of digital health with federated learning

Toward a Global and Reproducible Science for Brain Imaging in Neurotrauma: The ENIGMA Adult Moderate/Severe Traumatic Brain Injury Working Group

Unsupervised contrastive graph learning for resting‐state functional MRI analysis and brain disorder detection

Contact Info

Product

Resources

About