Federated learning enables big data for rare cancer boundary detection
Although machine learning (ML) has shown promise across disciplines, out-of-sample generalizability is concerning. This is currently addressed by sharing multi-site data, but such centralization is challenging/infeasible to scale due to various limitations. Federated ML (FL) provides an alternative...
| Huvudupphovsmän: | , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , |
|---|---|
| Materialtyp: | Journal article |
| Språk: | English |
| Publicerad: |
Springer Nature
2022
|
| _version_ | 1826313244341436416 |
|---|---|
| author | Pati, S Baid, U Edwards, B Sheller, M Wang, S-H Reina, GA Foley, P Gruzdev, A Karkada, D Davatzikos, C Sako, C Ghodasara, S Bilello, M Mohan, S Vollmuth, P Brugnara, G Preetha, CJ Sahm, F Maier-Hein, K Zenk, M Bendszus, M Wick, W Calabrese, E Rudie, J Villanueva-Meyer, J Cha, S Ingalhalikar, M Jadhav, M Pandey, U Saini, J Garrett, J Larson, M Jeraj, R Currie, S Frood, R Fatania, K Huang, RY Chang, K Balaña, C Capellades, J Puig, J Trenkler, J Pichler, J Necker, G Haunschmidt, A Meckel, S Shukla, G Liem, S Alexander, GS Lombardo, J Kamnitsas, K |
| author_facet | Pati, S Baid, U Edwards, B Sheller, M Wang, S-H Reina, GA Foley, P Gruzdev, A Karkada, D Davatzikos, C Sako, C Ghodasara, S Bilello, M Mohan, S Vollmuth, P Brugnara, G Preetha, CJ Sahm, F Maier-Hein, K Zenk, M Bendszus, M Wick, W Calabrese, E Rudie, J Villanueva-Meyer, J Cha, S Ingalhalikar, M Jadhav, M Pandey, U Saini, J Garrett, J Larson, M Jeraj, R Currie, S Frood, R Fatania, K Huang, RY Chang, K Balaña, C Capellades, J Puig, J Trenkler, J Pichler, J Necker, G Haunschmidt, A Meckel, S Shukla, G Liem, S Alexander, GS Lombardo, J Kamnitsas, K |
| author_sort | Pati, S |
| collection | OXFORD |
| description | Although machine learning (ML) has shown promise across disciplines, out-of-sample generalizability is concerning. This is currently addressed by sharing multi-site data, but such centralization is challenging/infeasible to scale due to various limitations. Federated ML (FL) provides an alternative paradigm for accurate and generalizable ML, by only sharing numerical model updates. Here we present the largest FL study to-date, involving data from 71 sites across 6 continents, to generate an automatic tumor boundary detector for the rare disease of glioblastoma, reporting the largest such dataset in the literature (n = 6, 314). We demonstrate a 33% delineation improvement for the surgically targetable tumor, and 23% for the complete tumor extent, over a publicly trained model. We anticipate our study to: 1) enable more healthcare studies informed by large diverse data, ensuring meaningful results for rare diseases and underrepresented populations, 2) facilitate further analyses for glioblastoma by releasing our consensus model, and 3) demonstrate the FL effectiveness at such scale and task-complexity as a paradigm shift for multi-site collaborations, alleviating the need for data-sharing. |
| first_indexed | 2024-09-25T04:08:27Z |
| format | Journal article |
| id | oxford-uuid:193f0b7d-e9d1-460d-80e2-dbc7a12bdb3c |
| institution | University of Oxford |
| language | English |
| last_indexed | 2024-09-25T04:08:27Z |
| publishDate | 2022 |
| publisher | Springer Nature |
| record_format | dspace |
| spelling | oxford-uuid:193f0b7d-e9d1-460d-80e2-dbc7a12bdb3c2024-06-14T12:28:32ZFederated learning enables big data for rare cancer boundary detectionJournal articlehttp://purl.org/coar/resource_type/c_dcae04bcuuid:193f0b7d-e9d1-460d-80e2-dbc7a12bdb3cEnglishSymplectic ElementsSpringer Nature2022Pati, SBaid, UEdwards, BSheller, MWang, S-HReina, GAFoley, PGruzdev, AKarkada, DDavatzikos, CSako, CGhodasara, SBilello, MMohan, SVollmuth, PBrugnara, GPreetha, CJSahm, FMaier-Hein, KZenk, MBendszus, MWick, WCalabrese, ERudie, JVillanueva-Meyer, JCha, SIngalhalikar, MJadhav, MPandey, USaini, JGarrett, JLarson, MJeraj, RCurrie, SFrood, RFatania, KHuang, RYChang, KBalaña, CCapellades, JPuig, JTrenkler, JPichler, JNecker, GHaunschmidt, AMeckel, SShukla, GLiem, SAlexander, GSLombardo, JKamnitsas, KAlthough machine learning (ML) has shown promise across disciplines, out-of-sample generalizability is concerning. This is currently addressed by sharing multi-site data, but such centralization is challenging/infeasible to scale due to various limitations. Federated ML (FL) provides an alternative paradigm for accurate and generalizable ML, by only sharing numerical model updates. Here we present the largest FL study to-date, involving data from 71 sites across 6 continents, to generate an automatic tumor boundary detector for the rare disease of glioblastoma, reporting the largest such dataset in the literature (n = 6, 314). We demonstrate a 33% delineation improvement for the surgically targetable tumor, and 23% for the complete tumor extent, over a publicly trained model. We anticipate our study to: 1) enable more healthcare studies informed by large diverse data, ensuring meaningful results for rare diseases and underrepresented populations, 2) facilitate further analyses for glioblastoma by releasing our consensus model, and 3) demonstrate the FL effectiveness at such scale and task-complexity as a paradigm shift for multi-site collaborations, alleviating the need for data-sharing. |
| spellingShingle | Pati, S Baid, U Edwards, B Sheller, M Wang, S-H Reina, GA Foley, P Gruzdev, A Karkada, D Davatzikos, C Sako, C Ghodasara, S Bilello, M Mohan, S Vollmuth, P Brugnara, G Preetha, CJ Sahm, F Maier-Hein, K Zenk, M Bendszus, M Wick, W Calabrese, E Rudie, J Villanueva-Meyer, J Cha, S Ingalhalikar, M Jadhav, M Pandey, U Saini, J Garrett, J Larson, M Jeraj, R Currie, S Frood, R Fatania, K Huang, RY Chang, K Balaña, C Capellades, J Puig, J Trenkler, J Pichler, J Necker, G Haunschmidt, A Meckel, S Shukla, G Liem, S Alexander, GS Lombardo, J Kamnitsas, K Federated learning enables big data for rare cancer boundary detection |
| title | Federated learning enables big data for rare cancer boundary detection |
| title_full | Federated learning enables big data for rare cancer boundary detection |
| title_fullStr | Federated learning enables big data for rare cancer boundary detection |
| title_full_unstemmed | Federated learning enables big data for rare cancer boundary detection |
| title_short | Federated learning enables big data for rare cancer boundary detection |
| title_sort | federated learning enables big data for rare cancer boundary detection |
| work_keys_str_mv | AT patis federatedlearningenablesbigdataforrarecancerboundarydetection AT baidu federatedlearningenablesbigdataforrarecancerboundarydetection AT edwardsb federatedlearningenablesbigdataforrarecancerboundarydetection AT shellerm federatedlearningenablesbigdataforrarecancerboundarydetection AT wangsh federatedlearningenablesbigdataforrarecancerboundarydetection AT reinaga federatedlearningenablesbigdataforrarecancerboundarydetection AT foleyp federatedlearningenablesbigdataforrarecancerboundarydetection AT gruzdeva federatedlearningenablesbigdataforrarecancerboundarydetection AT karkadad federatedlearningenablesbigdataforrarecancerboundarydetection AT davatzikosc federatedlearningenablesbigdataforrarecancerboundarydetection AT sakoc federatedlearningenablesbigdataforrarecancerboundarydetection AT ghodasaras federatedlearningenablesbigdataforrarecancerboundarydetection AT bilellom federatedlearningenablesbigdataforrarecancerboundarydetection AT mohans federatedlearningenablesbigdataforrarecancerboundarydetection AT vollmuthp federatedlearningenablesbigdataforrarecancerboundarydetection AT brugnarag federatedlearningenablesbigdataforrarecancerboundarydetection AT preethacj federatedlearningenablesbigdataforrarecancerboundarydetection AT sahmf federatedlearningenablesbigdataforrarecancerboundarydetection AT maierheink federatedlearningenablesbigdataforrarecancerboundarydetection AT zenkm federatedlearningenablesbigdataforrarecancerboundarydetection AT bendszusm federatedlearningenablesbigdataforrarecancerboundarydetection AT wickw federatedlearningenablesbigdataforrarecancerboundarydetection AT calabresee federatedlearningenablesbigdataforrarecancerboundarydetection AT rudiej federatedlearningenablesbigdataforrarecancerboundarydetection AT villanuevameyerj federatedlearningenablesbigdataforrarecancerboundarydetection AT chas federatedlearningenablesbigdataforrarecancerboundarydetection AT ingalhalikarm federatedlearningenablesbigdataforrarecancerboundarydetection AT jadhavm federatedlearningenablesbigdataforrarecancerboundarydetection AT pandeyu federatedlearningenablesbigdataforrarecancerboundarydetection AT sainij federatedlearningenablesbigdataforrarecancerboundarydetection AT garrettj federatedlearningenablesbigdataforrarecancerboundarydetection AT larsonm federatedlearningenablesbigdataforrarecancerboundarydetection AT jerajr federatedlearningenablesbigdataforrarecancerboundarydetection AT curries federatedlearningenablesbigdataforrarecancerboundarydetection AT froodr federatedlearningenablesbigdataforrarecancerboundarydetection AT fataniak federatedlearningenablesbigdataforrarecancerboundarydetection AT huangry federatedlearningenablesbigdataforrarecancerboundarydetection AT changk federatedlearningenablesbigdataforrarecancerboundarydetection AT balanac federatedlearningenablesbigdataforrarecancerboundarydetection AT capelladesj federatedlearningenablesbigdataforrarecancerboundarydetection AT puigj federatedlearningenablesbigdataforrarecancerboundarydetection AT trenklerj federatedlearningenablesbigdataforrarecancerboundarydetection AT pichlerj federatedlearningenablesbigdataforrarecancerboundarydetection AT neckerg federatedlearningenablesbigdataforrarecancerboundarydetection AT haunschmidta federatedlearningenablesbigdataforrarecancerboundarydetection AT meckels federatedlearningenablesbigdataforrarecancerboundarydetection AT shuklag federatedlearningenablesbigdataforrarecancerboundarydetection AT liems federatedlearningenablesbigdataforrarecancerboundarydetection AT alexandergs federatedlearningenablesbigdataforrarecancerboundarydetection AT lombardoj federatedlearningenablesbigdataforrarecancerboundarydetection AT kamnitsask federatedlearningenablesbigdataforrarecancerboundarydetection |