Federated learning enables big data for rare cancer boundary detection
Data Sharing
Sample (material)
DOI:
10.1038/s41467-022-33407-5
Publication Date:
2022-12-05T11:02:37Z
AUTHORS (279)
ABSTRACT
Although machine learning (ML) has shown promise in numerous domains, there are concerns about generalizability to out-of-sample data. This is currently addressed by centrally sharing ample, and importantly diverse, data from multiple sites. However, such centralization challenging scale (or even not feasible) due various limitations. Federated ML (FL) provides an alternative train accurate generalizable models, only numerical model updates. Here we present findings the largest FL study to-date, involving 71 healthcare institutions across 6 continents, generate automatic tumor boundary detector for rare disease of glioblastoma, utilizing dataset patients ever used literature (25,256 MRI scans 6,314 patients). We demonstrate a 33% improvement over publicly trained delineate surgically targetable tumor, 23% tumor's entire extent. anticipate our to: 1) enable more studies informed large diverse data, ensuring meaningful results diseases underrepresented populations, 2) facilitate further quantitative analyses glioblastoma via performance optimization consensus eventual public release, 3) effectiveness at task complexity as paradigm shift multi-site collaborations, alleviating need sharing.
SUPPLEMENTAL MATERIAL
Coming soon ....
REFERENCES (124)
CITATIONS (192)
EXTERNAL LINKS
PlumX Metrics
RECOMMENDATIONS
FAIR ASSESSMENT
Coming soon ....
JUPYTER LAB
Coming soon ....