Distributed learning: a reliable privacy-preserving strategy to change multicenter collaborations using AI

Abstract

Purpose The present scoping review aims to assess the non-inferiority of distributed learning over centrally and locally trained machine learning (ML) models in medical applications. Methods We performed a literature search using the term “distributed learning” OR “federated learning” in the PubMed/MEDLINE and EMBASE databases. No start date limit was used, and the search was extended until July 21, 2020. We excluded articles outside the field of interest; guidelines or expert opinion, review articles and meta-analyses, editorials, letters or commentaries, and conference abstracts; articles not in the English language; and studies not using medical data. Selected studies were classified and analysed according to their aim(s). Results We included 26 papers aimed at predicting one or more outcomes: namely risk, diagnosis, prognosis, and treatment side effect/adverse drug reaction. Distributed learning was compared to centralized or localized training in 21/26 and 14/26 selected papers, respectively. Regardless of the aim, the type of input, the method, and the classifier, distributed learning performed close to centralized training, but two experiments focused on diagnosis. In all but 2 cases, distributed learning outperformed locally trained models. Conclusion Distributed learning resulted in a reliable strategy for model development; indeed, it performed equally to models trained on centralized datasets. Sensitive data can get preserved since they are not shared for model development. Distributed learning constitutes a promising solution for ML-based research and practice since large, diverse datasets are crucial for success.

Decentralized Distributed Multi-institutional PET Image Segmentation Using a Federated Deep Learning Framework

“…The FL framework can address these challenges by providing decentralized training procedures for DL models. This approach preserves privacy and paves the way to train DL models collaboratively on large multicentric data sets without sharing data sets between centers 23–25 …”

“…Overall, statistical analysis showed no significant differences ( P > 0.05) between these 2 strategies for different quantitative metrics. Collaborative DL model training without sharing data sets between different hospitals and centers to preserve patients’ privacy using FL has been reported in few studies 23–25 . Dayan et al 35 developed an FL-based model for oxygen requirements in COVID-19 patients using vital, laboratory, and chest x-ray images.…”

“…This approach preserves privacy and paves the way to train DL models collaboratively on large multicentric data sets without sharing data sets between centers. [23][24][25] In the current study, we compared the performance of centralized and FL models for the segmentation of HNC PET images. Overall, a high consistency was observed between centralized and FL approaches in terms of quantitative image segmentation metrics, including Dice coefficient (0.84 ± 0.06 vs 0.84 ± 0.05) and Jaccard coefficient (0.73 ± 0.08 vs 0.73 ± 0.07).…”

“…In a wide range of medical situations, this becomes a critical challenge, because sharing medical data between different centers is limited because of concerns over patients’ privacy and ethical and legal issues 23–25 . To overcome this limitation, the concept of federated learning (FL) is being increasingly explored in the context of medical data and more recently in medical imaging.…”

Ensemble learning using individual neonatal data for seizure detection

“…the Privacy Rule of the U.S. Health Insurance Portability and Accountability Act (HIPAA), or the European General Data Protection Regulation (GDPR)) making data sharing between institutions difficult, if not impossible [10,54]. Challenges in sharing data have triggered growing interest in distributed approaches to statistical learning [20].…”