Real-world federated learning in radiology: hurdles to overcome and benefits to gain
OBJECTIVE: Federated Learning (FL) enables collaborative model training while keeping data locally. Currently, most FL studies in radiology are conducted in simulated environments due to numerous hurdles impeding its translation into practice. The few existing real-world FL initiatives rarely commun...
Gespeichert in:
| Hauptverfasser: | , , , , , , , , , , , , , , , , , , , |
|---|---|
| Dokumenttyp: | Article (Journal) |
| Sprache: | Englisch |
| Veröffentlicht: |
January 2025
|
| In: |
Journal of the American Medical Informatics Association
Year: 2025, Jahrgang: 32, Heft: 1, Pages: 193-205 |
| ISSN: | 1527-974X |
| DOI: | 10.1093/jamia/ocae259 |
| Online-Zugang: | Verlag, kostenfrei, Volltext: https://doi.org/10.1093/jamia/ocae259
|
| Verfasserangaben: | Markus Ralf Bujotzek, MSc, Ünal Akünal, MSc, Stefan Denner, MSc, Peter Neher, PhD, Maximilian Zenk, MSc, Eric Frodl, MSc, Astha Jaiswal, PhD, Moon Kim, MD, Nicolai R. Krekiehn, MSc, Manuel Nickel, MSc, Richard Ruppel, MSc, Marcus Both, MD, Felix Döllinger, MD, Marcel Opitz, MD, Thorsten Persigehl, MD, Jens Kleesiek, PhD, MD, Tobias Penzkofer, MD, Klaus Maier-Hein, PhD, Andreas Bucher, MD, Rickmer Braren, MD |
MARC
| LEADER | 00000caa a2200000 c 4500 | ||
|---|---|---|---|
| 001 | 1933111933 | ||
| 003 | DE-627 | ||
| 005 | 20260210111816.0 | ||
| 007 | cr uuu---uuuuu | ||
| 008 | 250813s2025 xx |||||o 00| ||eng c | ||
| 024 | 7 | |a 10.1093/jamia/ocae259 |2 doi | |
| 035 | |a (DE-627)1933111933 | ||
| 035 | |a (DE-599)KXP1933111933 | ||
| 035 | |a (OCoLC)1559703842 | ||
| 040 | |a DE-627 |b ger |c DE-627 |e rda | ||
| 041 | |a eng | ||
| 084 | |a 33 |2 sdnb | ||
| 100 | 1 | |a Bujotzek, Markus |e VerfasserIn |0 (DE-588)1364083175 |0 (DE-627)192389823X |4 aut | |
| 245 | 1 | 0 | |a Real-world federated learning in radiology |b hurdles to overcome and benefits to gain |c Markus Ralf Bujotzek, MSc, Ünal Akünal, MSc, Stefan Denner, MSc, Peter Neher, PhD, Maximilian Zenk, MSc, Eric Frodl, MSc, Astha Jaiswal, PhD, Moon Kim, MD, Nicolai R. Krekiehn, MSc, Manuel Nickel, MSc, Richard Ruppel, MSc, Marcus Both, MD, Felix Döllinger, MD, Marcel Opitz, MD, Thorsten Persigehl, MD, Jens Kleesiek, PhD, MD, Tobias Penzkofer, MD, Klaus Maier-Hein, PhD, Andreas Bucher, MD, Rickmer Braren, MD |
| 264 | 1 | |c January 2025 | |
| 300 | |b Illustrationen | ||
| 300 | |a 13 | ||
| 336 | |a Text |b txt |2 rdacontent | ||
| 337 | |a Computermedien |b c |2 rdamedia | ||
| 338 | |a Online-Ressource |b cr |2 rdacarrier | ||
| 500 | |a Online veröffentlicht: 25. Oktober 2024 | ||
| 500 | |a Gesehen am 13.08.2025 | ||
| 520 | |a OBJECTIVE: Federated Learning (FL) enables collaborative model training while keeping data locally. Currently, most FL studies in radiology are conducted in simulated environments due to numerous hurdles impeding its translation into practice. The few existing real-world FL initiatives rarely communicate specific measures taken to overcome these hurdles. To bridge this significant knowledge gap, we propose a comprehensive guide for real-world FL in radiology. Minding efforts to implement real-world FL, there is a lack of comprehensive assessments comparing FL to less complex alternatives in challenging real-world settings, which we address through extensive benchmarking. - MATERIALS AND METHODS: We developed our own FL infrastructure within the German Radiological Cooperative Network (RACOON) and demonstrated its functionality by training FL models on lung pathology segmentation tasks across six university hospitals. Insights gained while establishing our FL initiative and running the extensive benchmark experiments were compiled and categorized into the guide. - RESULTS: The proposed guide outlines essential steps, identified hurdles, and implemented solutions for establishing successful FL initiatives conducting real-world experiments. Our experimental results prove the practical relevance of our guide and show that FL outperforms less complex alternatives in all evaluation scenarios. - DISCUSSION AND CONCLUSION: Our findings justify the efforts required to translate FL into real-world applications by demonstrating advantageous performance over alternative approaches. Additionally, they emphasize the importance of strategic organization, robust management of distributed data and infrastructure in real-world settings. With the proposed guide, we are aiming to aid future FL researchers in circumventing pitfalls and accelerating translation of FL into radiological applications. | ||
| 650 | 4 | |a artificial intelligence | |
| 650 | 4 | |a Benchmarking | |
| 650 | 4 | |a distributed systems | |
| 650 | 4 | |a federated learning | |
| 650 | 4 | |a Germany | |
| 650 | 4 | |a healthcare infrastructure | |
| 650 | 4 | |a Humans | |
| 650 | 4 | |a radiology | |
| 650 | 4 | |a Radiology | |
| 700 | 1 | |a Akünal, Ünal |e VerfasserIn |4 aut | |
| 700 | 1 | |a Denner, Stefan |e VerfasserIn |4 aut | |
| 700 | 1 | |a Neher, Peter |d 1984- |e VerfasserIn |0 (DE-588)1052246052 |0 (DE-627)788077287 |0 (DE-576)408004746 |4 aut | |
| 700 | 1 | |a Zenk, Maximilian |d 1991- |e VerfasserIn |0 (DE-588)1383402647 |0 (DE-627)1944422668 |4 aut | |
| 700 | 1 | |a Frodl, Eric |e VerfasserIn |0 (DE-588)1344376479 |0 (DE-627)190520342X |4 aut | |
| 700 | 1 | |a Jaiswal, Astha |e VerfasserIn |4 aut | |
| 700 | 1 | |a Kim, Moon |e VerfasserIn |4 aut | |
| 700 | 1 | |a Krekiehn, Nicolai R. |e VerfasserIn |4 aut | |
| 700 | 1 | |a Nickel, Manuel |e VerfasserIn |4 aut | |
| 700 | 1 | |a Ruppel, Richard |e VerfasserIn |4 aut | |
| 700 | 1 | |a Both, Marcus |e VerfasserIn |4 aut | |
| 700 | 1 | |a Döllinger, Felix |e VerfasserIn |4 aut | |
| 700 | 1 | |a Opitz, Marcel |e VerfasserIn |4 aut | |
| 700 | 1 | |a Persigehl, Thorsten |e VerfasserIn |4 aut | |
| 700 | 1 | |a Kleesiek, Jens Philipp |d 1977- |e VerfasserIn |0 (DE-588)132998076 |0 (DE-627)530080745 |0 (DE-576)299554465 |4 aut | |
| 700 | 1 | |a Penzkofer, Tobias |d 1978- |e VerfasserIn |0 (DE-588)13699007X |0 (DE-627)694809519 |0 (DE-576)301378681 |4 aut | |
| 700 | 1 | |a Maier-Hein, Klaus H. |d 1980- |e VerfasserIn |0 (DE-588)1100551875 |0 (DE-627)85946461X |0 (DE-576)333771222 |4 aut | |
| 700 | 1 | |a Bucher, Andreas |e VerfasserIn |0 (DE-588)1257808184 |0 (DE-627)1802495649 |4 aut | |
| 700 | 1 | |a Braren, Rickmer |d 1971- |e VerfasserIn |0 (DE-588)124221777 |0 (DE-627)706576306 |0 (DE-576)294075321 |4 aut | |
| 773 | 0 | 8 | |i Enthalten in |a American Medical Informatics Association |t Journal of the American Medical Informatics Association |d Oxford : Oxford Univ. Press, 1994 |g 32(2025), 1 vom: Jan., Seite 193-205 |h Online-Ressource |w (DE-627)316003891 |w (DE-600)2018371-9 |w (DE-576)09408064X |x 1527-974X |7 nnas |
| 773 | 1 | 8 | |g volume:32 |g year:2025 |g number:1 |g month:01 |g pages:193-205 |g extent:13 |a Real-world federated learning in radiology hurdles to overcome and benefits to gain |
| 856 | 4 | 0 | |u https://doi.org/10.1093/jamia/ocae259 |x Verlag |x Resolving-System |z kostenfrei |3 Volltext |7 0 |
| 951 | |a AR | ||
| 992 | |a 20250813 | ||
| 993 | |a Article | ||
| 994 | |a 2025 | ||
| 998 | |g 1100551875 |a Maier-Hein, Klaus H. |m 1100551875:Maier-Hein, Klaus H. |d 910000 |d 911400 |e 910000PM1100551875 |e 911400PM1100551875 |k 0/910000/ |k 1/910000/911400/ | ||
| 998 | |g 132998076 |a Kleesiek, Jens Philipp |m 132998076:Kleesiek, Jens Philipp |d 50000 |e 50000PK132998076 |k 0/50000/ |p 16 | ||
| 998 | |g 1383402647 |a Zenk, Maximilian |m 1383402647:Zenk, Maximilian |d 50000 |e 50000PZ1383402647 |k 0/50000/ |p 5 | ||
| 998 | |g 1052246052 |a Neher, Peter |m 1052246052:Neher, Peter |d 910000 |d 911400 |e 910000PN1052246052 |e 911400PN1052246052 |k 0/910000/ |k 1/910000/911400/ |p 4 | ||
| 999 | |a KXP-PPN1933111933 |e 4756986315 | ||
| BIB | |a Y | ||
| SER | |a journal | ||
| JSO | |a {"id":{"eki":["1933111933"],"doi":["10.1093/jamia/ocae259"]},"physDesc":[{"noteIll":"Illustrationen","extent":"13 S."}],"origin":[{"dateIssuedDisp":"January 2025","dateIssuedKey":"2025"}],"type":{"media":"Online-Ressource","bibl":"article-journal"},"person":[{"role":"aut","display":"Bujotzek, Markus","family":"Bujotzek","given":"Markus"},{"role":"aut","display":"Akünal, Ünal","family":"Akünal","given":"Ünal"},{"role":"aut","display":"Denner, Stefan","family":"Denner","given":"Stefan"},{"role":"aut","display":"Neher, Peter","family":"Neher","given":"Peter"},{"family":"Zenk","given":"Maximilian","role":"aut","display":"Zenk, Maximilian"},{"role":"aut","display":"Frodl, Eric","family":"Frodl","given":"Eric"},{"family":"Jaiswal","given":"Astha","role":"aut","display":"Jaiswal, Astha"},{"role":"aut","display":"Kim, Moon","family":"Kim","given":"Moon"},{"given":"Nicolai R.","family":"Krekiehn","display":"Krekiehn, Nicolai R.","role":"aut"},{"family":"Nickel","given":"Manuel","role":"aut","display":"Nickel, Manuel"},{"role":"aut","display":"Ruppel, Richard","family":"Ruppel","given":"Richard"},{"display":"Both, Marcus","role":"aut","given":"Marcus","family":"Both"},{"family":"Döllinger","given":"Felix","role":"aut","display":"Döllinger, Felix"},{"role":"aut","display":"Opitz, Marcel","family":"Opitz","given":"Marcel"},{"given":"Thorsten","family":"Persigehl","display":"Persigehl, Thorsten","role":"aut"},{"given":"Jens Philipp","family":"Kleesiek","display":"Kleesiek, Jens Philipp","role":"aut"},{"given":"Tobias","family":"Penzkofer","display":"Penzkofer, Tobias","role":"aut"},{"given":"Klaus H.","family":"Maier-Hein","display":"Maier-Hein, Klaus H.","role":"aut"},{"given":"Andreas","family":"Bucher","display":"Bucher, Andreas","role":"aut"},{"role":"aut","display":"Braren, Rickmer","family":"Braren","given":"Rickmer"}],"relHost":[{"disp":"American Medical Informatics AssociationJournal of the American Medical Informatics Association","physDesc":[{"extent":"Online-Ressource"}],"id":{"issn":["1527-974X"],"eki":["316003891"],"zdb":["2018371-9"]},"note":["Gesehen am 16.12.14","Fortsetzung der Druck-Ausgabe","Ungezählte Beil.: Suppl"],"type":{"media":"Online-Ressource","bibl":"periodical"},"origin":[{"dateIssuedKey":"1994","publisher":"Oxford Univ. Press ; Hanley & Belfus ; Elsevier ; BMJ Publishing Group","dateIssuedDisp":"1994-","publisherPlace":"Oxford ; Philadelphia, Pa. ; New York, NY [u.a.] ; London"}],"recId":"316003891","title":[{"title":"Journal of the American Medical Informatics Association","subtitle":"JAMIA","title_sort":"Journal of the American Medical Informatics Association"}],"part":{"text":"32(2025), 1 vom: Jan., Seite 193-205","issue":"1","volume":"32","extent":"13","pages":"193-205","year":"2025"},"pubHistory":["1.1994 -"],"language":["eng"],"titleAlt":[{"title":"JAMIA"}],"corporate":[{"display":"American Medical Informatics Association","role":"aut"}]}],"note":["Online veröffentlicht: 25. Oktober 2024","Gesehen am 13.08.2025"],"recId":"1933111933","name":{"displayForm":["Markus Ralf Bujotzek, MSc, Ünal Akünal, MSc, Stefan Denner, MSc, Peter Neher, PhD, Maximilian Zenk, MSc, Eric Frodl, MSc, Astha Jaiswal, PhD, Moon Kim, MD, Nicolai R. Krekiehn, MSc, Manuel Nickel, MSc, Richard Ruppel, MSc, Marcus Both, MD, Felix Döllinger, MD, Marcel Opitz, MD, Thorsten Persigehl, MD, Jens Kleesiek, PhD, MD, Tobias Penzkofer, MD, Klaus Maier-Hein, PhD, Andreas Bucher, MD, Rickmer Braren, MD"]},"language":["eng"],"title":[{"title_sort":"Real-world federated learning in radiology","title":"Real-world federated learning in radiology","subtitle":"hurdles to overcome and benefits to gain"}]} | ||
| SRT | |a BUJOTZEKMAREALWORLDF2025 | ||