Distributed learning: a reliable privacy-preserving strategy to change multicenter collaborations using AI
| Autor: | Luca Mainardi, Pier Luca Lanzi, Francesco Amigoni, Arturo Chiti, Martina Sollini, Noemi Gozzi, Daniele Loiacono, Margarita Kirienko, Gaia Ninatti, Edoardo Giacomello |
|---|---|
| Jazyk: | angličtina |
| Rok vydání: | 2021 |
| Předmět: |
Databases
Factual Computer science MEDLINE Federated learning Review Article Field (computer science) 030218 nuclear medicine & medical imaging Machine Learning 03 medical and health sciences 0302 clinical medicine Side effect (computer science) Humans Multicenter Studies as Topic Radiology Nuclear Medicine and imaging Distributed learning Ethics Information retrieval General Medicine Term (time) Privacy preserving Clinical trial Privacy Research Design 030220 oncology & carcinogenesis Classifier (UML) Algorithms |
| Zdroj: | European Journal of Nuclear Medicine and Molecular Imaging |
| Popis: | 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. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|