{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,10,31]],"date-time":"2025-10-31T02:48:15Z","timestamp":1761878895691,"version":"build-2065373602"},"reference-count":26,"publisher":"World Scientific Pub Co Pte Ltd","issue":"03","funder":[{"name":"CASSANDRA - Clinical ASSist AND aleRt Algorithms","award":["01VSF20015"],"award-info":[{"award-number":["01VSF20015"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Int. J. Semantic Computing"],"published-print":{"date-parts":[[2025,9]]},"abstract":"<jats:p>Federated learning (FL) has emerged as a promising solution to deal with the privacy concerns which often limit access to data for training machine learning (ML) models. It encompasses the exchange of model parameters between data owners and a central model aggregation instance without the need to share sensitive data directly. Although FL does not require sharing sensitive information, it may not guarantee complete privacy, as model parameters can potentially reveal sensitive details about the underlying training data. To enhance privacy, FL approaches can additionally integrate differential privacy (DP), which entails the careful addition of noise to the model.<\/jats:p>\n                  <jats:p>In this paper, we investigate the relationship between FL, DP and highly non-independent and identically distributed (non-IID) data. We present a specific real-life healthcare application of predicting patient mortality and revision surgery after visceral operations. Our contributions encompass assessing the performance of FL with and without DP in the presence of highly non-IID data, analyzing the fairness for individual training participants and investigating the efficacy of personalization to alleviate the privacy-utility trade-off introduced by DP.<\/jats:p>\n                  <jats:p>Our findings indicate that the addition of DP into FL given non-IID datasets disparately favors participants with larger datasets, while the model has limited utility for those with less data. The addition of a subsequent model fine-tuning step can remedy this disparity and enables all participants to reach a model performance comparable to the centralized training scenario.<\/jats:p>","DOI":"10.1142\/s1793351x25430019","type":"journal-article","created":{"date-parts":[[2025,9,12]],"date-time":"2025-09-12T08:28:43Z","timestamp":1757665723000},"page":"343-367","source":"Crossref","is-referenced-by-count":0,"title":["Differentially-Private Federated Learning with Non-IID Data For Surgical Risk Prediction"],"prefix":"10.1142","volume":"19","author":[{"ORCID":"https:\/\/orcid.org\/0000-0001-7824-8872","authenticated-orcid":false,"given":"Bjarne","family":"Pfitzner","sequence":"first","affiliation":[{"name":"Digital Health \u2013 Connected Healthcare, Hasso Plattner Institute, Rudolf-Breitscheid-Str. 187, 14482 Potsdam, Germany"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-0640-499X","authenticated-orcid":false,"given":"Max M.","family":"Maurer","sequence":"additional","affiliation":[{"name":"Department of Surgery, Experimental Surgery, Campus Charit\u00e9 Mitte Campus Virchow Klinikum, Charit\u00e9 \u2013 Universit\u00e4tsmedizin Berlin, Freie Universit\u00e4t Berlin, Humboldt-Universit\u00e4t zu Berlin, Augustenburger Platz 1, 13353 Berlin, Germany"},{"name":"Berlin Institute of Health at Charit\u00e9 \u2013 Universit\u00e4tsmedizin Berlin, BIH Biomedical Innovation Academy, BIH Charit\u00e9 Digital Clinician Scientist Program, Charit\u00e9platz 1, 10117 Berlin, Germany"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-7435-4993","authenticated-orcid":false,"given":"Axel","family":"Winter","sequence":"additional","affiliation":[{"name":"Department of Surgery, Experimental Surgery, Campus Charit\u00e9 Mitte Campus Virchow Klinikum, Charit\u00e9 \u2013 Universit\u00e4tsmedizin Berlin, Freie Universit\u00e4t Berlin, Humboldt-Universit\u00e4t zu Berlin, Augustenburger Platz 1, 13353 Berlin, Germany"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0009-0003-3763-404X","authenticated-orcid":false,"given":"Christoph","family":"Riepe","sequence":"additional","affiliation":[{"name":"Department of Surgery, Experimental Surgery, Campus Charit\u00e9 Mitte Campus Virchow Klinikum, Charit\u00e9 \u2013 Universit\u00e4tsmedizin Berlin, Freie Universit\u00e4t Berlin, Humboldt-Universit\u00e4t zu Berlin, Augustenburger Platz 1, 13353 Berlin, Germany"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-9355-937X","authenticated-orcid":false,"given":"Igor M.","family":"Sauer","sequence":"additional","affiliation":[{"name":"Department of Surgery, Experimental Surgery, Campus Charit\u00e9 Mitte Campus Virchow Klinikum, Charit\u00e9 \u2013 Universit\u00e4tsmedizin Berlin, Freie Universit\u00e4t Berlin, Humboldt-Universit\u00e4t zu Berlin, Augustenburger Platz 1, 13353 Berlin, Germany"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-2895-4872","authenticated-orcid":false,"given":"Robin","family":"van de Water","sequence":"additional","affiliation":[{"name":"Digital Health \u2013 Connected Healthcare, Hasso Plattner Institute, Rudolf-Breitscheid-Str. 187, 14482 Potsdam, Germany"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0009-0006-4717-6742","authenticated-orcid":false,"given":"Christian","family":"Denecke","sequence":"additional","affiliation":[{"name":"Department of Surgery, Experimental Surgery, Campus Charit\u00e9 Mitte Campus Virchow Klinikum, Charit\u00e9 \u2013 Universit\u00e4tsmedizin Berlin, Freie Universit\u00e4t Berlin, Humboldt-Universit\u00e4t zu Berlin, Augustenburger Platz 1, 13353 Berlin, Germany"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-9839-1369","authenticated-orcid":false,"given":"Johann","family":"Pratschke","sequence":"additional","affiliation":[{"name":"Department of Surgery, Experimental Surgery, Campus Charit\u00e9 Mitte Campus Virchow Klinikum, Charit\u00e9 \u2013 Universit\u00e4tsmedizin Berlin, Freie Universit\u00e4t Berlin, Humboldt-Universit\u00e4t zu Berlin, Augustenburger Platz 1, 13353 Berlin, Germany"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-8380-7667","authenticated-orcid":false,"given":"Bert","family":"Arnrich","sequence":"additional","affiliation":[{"name":"Digital Health \u2013 Connected Healthcare, Hasso Plattner Institute, Rudolf-Breitscheid-Str. 187, 14482 Potsdam, Germany"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"219","published-online":{"date-parts":[[2025,10,14]]},"reference":[{"key":"S1793351X25430019BIB001","first-page":"1273","volume-title":"Proc. 20th Int. Conf. Artificial Intelligence and Statistics","author":"McMahan B.","year":"2017"},{"key":"S1793351X25430019BIB002","first-page":"14774","volume-title":"Proc. 33rd Conf. Advances in Neural Information Processing Systems","author":"Zhu L.","year":"2019"},{"key":"S1793351X25430019BIB003","first-page":"16937","volume-title":"Proc. 34th Conf. Advances in Neural Information Processing Systems","author":"Geiping J.","year":"2020"},{"key":"S1793351X25430019BIB004","doi-asserted-by":"publisher","DOI":"10.3390\/s22145195"},{"key":"S1793351X25430019BIB005","doi-asserted-by":"publisher","DOI":"10.1001\/jamanetworkopen.2021.7737"},{"key":"S1793351X25430019BIB006","doi-asserted-by":"publisher","DOI":"10.1097\/SLA.0000000000002956"},{"key":"S1793351X25430019BIB007","doi-asserted-by":"publisher","DOI":"10.1109\/EMBC46164.2021.9630897"},{"issue":"44","key":"S1793351X25430019BIB008","first-page":"739","volume":"116","author":"Baum P.","year":"2019","journal-title":"Dtsch Arztebl Int."},{"volume-title":"Proc. Int. Conf. Learning Representations","year":"2021","author":"Reddi S. J.","key":"S1793351X25430019BIB009"},{"key":"S1793351X25430019BIB012","doi-asserted-by":"publisher","DOI":"10.1561\/0400000042"},{"key":"S1793351X25430019BIB013","first-page":"263","volume-title":"Proc. IEEE 30th Computer Security Foundations Symp.","author":"Mironov I.","year":"2017"},{"key":"S1793351X25430019BIB014","doi-asserted-by":"publisher","DOI":"10.1145\/2810103.2813677"},{"key":"S1793351X25430019BIB015","doi-asserted-by":"publisher","DOI":"10.1145\/2976749.2978318"},{"key":"S1793351X25430019BIB016","first-page":"1926","volume-title":"Proc. Int. Conf. Learning Representations","author":"McMahan H. B.","year":"2018"},{"key":"S1793351X25430019BIB018","doi-asserted-by":"publisher","DOI":"10.1016\/j.future.2022.05.003"},{"key":"S1793351X25430019BIB019","first-page":"4427","volume-title":"Proc. 31st Conf. Advances in Neural Information Processing Systems","author":"Smith V.","year":"2017"},{"key":"S1793351X25430019BIB021","doi-asserted-by":"publisher","DOI":"10.1109\/TII.2021.3073925"},{"key":"S1793351X25430019BIB022","doi-asserted-by":"publisher","DOI":"10.3390\/math11194123"},{"key":"S1793351X25430019BIB023","doi-asserted-by":"publisher","DOI":"10.1016\/j.spinee.2018.11.009"},{"key":"S1793351X25430019BIB024","doi-asserted-by":"publisher","DOI":"10.1007\/s00423-021-02348-w"},{"key":"S1793351X25430019BIB025","doi-asserted-by":"publisher","DOI":"10.1371\/journal.pdig.0000033"},{"key":"S1793351X25430019BIB026","doi-asserted-by":"publisher","DOI":"10.1109\/TMI.2022.3222126"},{"key":"S1793351X25430019BIB027","doi-asserted-by":"publisher","DOI":"10.1016\/0895-4356(92)90133-8"},{"key":"S1793351X25430019BIB028","doi-asserted-by":"publisher","DOI":"10.1613\/jair.1.14649"},{"key":"S1793351X25430019BIB029","doi-asserted-by":"publisher","DOI":"10.1038\/s41592-019-0470-3"},{"key":"S1793351X25430019BIB030","doi-asserted-by":"publisher","DOI":"10.1109\/AIMHC59811.2024.00030"}],"container-title":["International Journal of Semantic Computing"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.worldscientific.com\/doi\/pdf\/10.1142\/S1793351X25430019","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,31]],"date-time":"2025-10-31T02:10:06Z","timestamp":1761876606000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.worldscientific.com\/doi\/10.1142\/S1793351X25430019"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,9]]},"references-count":26,"journal-issue":{"issue":"03","published-print":{"date-parts":[[2025,9]]}},"alternative-id":["10.1142\/S1793351X25430019"],"URL":"https:\/\/doi.org\/10.1142\/s1793351x25430019","relation":{},"ISSN":["1793-351X","1793-7108"],"issn-type":[{"type":"print","value":"1793-351X"},{"type":"electronic","value":"1793-7108"}],"subject":[],"published":{"date-parts":[[2025,9]]}}}