{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,24]],"date-time":"2026-10-24T07:29:26Z","timestamp":1792826966356,"version":"4.5.0"},"reference-count":27,"publisher":"Springer Science and Business Media LLC","issue":"1","license":[{"start":{"date-parts":[[2024,8,7]],"date-time":"2024-08-07T00:00:00Z","timestamp":1722988800000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2024,8,7]],"date-time":"2024-08-07T00:00:00Z","timestamp":1722988800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"funder":[{"name":"DKFZ Clinician Scientist Program, supported by the Dieter Morszeck Foundation"},{"name":"Agora 3.0 project"},{"name":"Universit\u00e4tsklinikum Freiburg"}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["Radiat Oncol"],"abstract":"<jats:title>Abstract<\/jats:title><jats:sec>\n                <jats:title>Purpose<\/jats:title>\n                <jats:p>Convolutional Neural Networks (CNNs) have emerged as transformative tools in the field of radiation oncology, significantly advancing the precision of contouring practices. However, the adaptability of these algorithms across diverse scanners, institutions, and imaging protocols remains a considerable obstacle. This study aims to investigate the effects of incorporating institution-specific datasets into the training regimen of CNNs to assess their generalization ability in real-world clinical environments. Focusing on a data-centric analysis, the influence of varying multi- and single center training approaches on algorithm performance is conducted.<\/jats:p>\n              <\/jats:sec><jats:sec>\n                <jats:title>Methods<\/jats:title>\n                <jats:p>nnU-Net is trained using a dataset comprising 161 <jats:sup>18<\/jats:sup>F-PSMA-1007 PET images collected from four distinct institutions (Freiburg: n\u2009=\u200996, Munich: n\u2009=\u200919, Cyprus: n\u2009=\u200932, Dresden: n\u2009=\u200914). The dataset is partitioned such that data from each center are systematically excluded from training and used solely for testing to assess the model's generalizability and adaptability to data from unfamiliar sources. Performance is compared through a 5-Fold Cross-Validation, providing a detailed comparison between models trained on datasets from single centers to those trained on aggregated multi-center datasets. Dice Similarity Score, Hausdorff distance and volumetric analysis are used as primary evaluation metrics.<\/jats:p>\n              <\/jats:sec><jats:sec>\n                <jats:title>Results<\/jats:title>\n                <jats:p>The mixed training approach yielded a median DSC of 0.76 (IQR: 0.64\u20130.84) in a five-fold cross-validation, showing no significant differences (p\u2009=\u20090.18) compared to models trained with data exclusion from each center, which performed with a median DSC of 0.74 (IQR: 0.56\u20130.86). Significant performance improvements regarding multi-center training were observed for the Dresden cohort (multi-center median DSC 0.71, IQR: 0.58\u20130.80 vs. single-center 0.68, IQR: 0.50\u20130.80, p\u2009&lt;\u20090.001) and Cyprus cohort (multi-center 0.74, IQR: 0.62\u20130.83 vs. single-center 0.72, IQR: 0.54\u20130.82, p\u2009&lt;\u20090.01). While Munich and Freiburg also showed performance improvements with multi-center training, results showed no statistical significance (Munich: multi-center DSC 0.74, IQR: 0.60\u20130.80 vs. single-center 0.72, IQR: 0.59\u20130.82, p\u2009&gt;\u20090.05; Freiburg: multi-center 0.78, IQR: 0.53\u20130.87 vs. single-center 0.71, IQR: 0.53\u20130.83, p\u2009=\u20090.23).<\/jats:p>\n              <\/jats:sec><jats:sec>\n                <jats:title>Conclusion<\/jats:title>\n                <jats:p>CNNs trained for auto contouring intraprostatic GTV in <jats:sup>18<\/jats:sup>F-PSMA-1007 PET on a diverse dataset from multiple centers mostly generalize well to unseen data from other centers. Training on a multicentric dataset can improve performance compared to training exclusively with a single-center dataset regarding intraprostatic <jats:sup>18<\/jats:sup>F-PSMA-1007 PET GTV segmentation. The segmentation performance of the same CNN can vary depending on the dataset employed for training and testing.<\/jats:p>\n              <\/jats:sec>","DOI":"10.1186\/s13014-024-02491-w","type":"journal-article","created":{"date-parts":[[2024,8,7]],"date-time":"2024-08-07T11:04:10Z","timestamp":1723028650000},"update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":15,"title":["The impact of multicentric datasets for the automated tumor delineation in primary prostate cancer using convolutional neural networks on 18F-PSMA-1007 PET"],"prefix":"10.1186","volume":"19","author":[{"given":"Julius C.","family":"Holzschuh","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Michael","family":"Mix","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Martin T.","family":"Freitag","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Tobias","family":"H\u00f6lscher","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Anja","family":"Braune","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"J\u00f6rg","family":"Kotzerke","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Alexis","family":"Vrachimis","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Paul","family":"Doolan","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Harun","family":"Ilhan","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ioana M.","family":"Marinescu","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Simon K. B.","family":"Spohn","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Tobias","family":"Fechter","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Dejan","family":"Kuhn","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Christian","family":"Gratzke","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Radu","family":"Grosu","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Anca-Ligia","family":"Grosu","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"C.","family":"Zamboglou","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"297","published-online":{"date-parts":[[2024,8,7]]},"reference":[{"issue":"6","key":"2491_CR1","doi-asserted-by":"publisher","first-page":"394","DOI":"10.3322\/caac.21492","volume":"68","author":"F Bray","year":"2018","unstructured":"Bray F, Ferlay J, Soerjomataram I, Siegel RL, Torre LA, Jemal A. \u201cGlobal cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin. 2018;68(6):394\u2013424. https:\/\/doi.org\/10.3322\/caac.21492.","journal-title":"CA Cancer J Clin"},{"issue":"1","key":"2491_CR2","doi-asserted-by":"publisher","first-page":"218","DOI":"10.1007\/s00259-022-05931-5","volume":"50","author":"SKB Spohn","year":"2022","unstructured":"Spohn SKB, et al. \u201cThe maximum standardized uptake value in patients with recurrent or persistent prostate cancer after radical prostatectomy and PSMA-PET-guided salvage radiotherapy-a multicenter retrospective analysis. Eur J Nucl Med Mol Imaging. 2022;50(1):218\u201327. https:\/\/doi.org\/10.1007\/s00259-022-05931-5.","journal-title":"Eur J Nucl Med Mol Imaging"},{"issue":"9","key":"2491_CR3","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1007\/s00066-021-01812-2","volume":"197","author":"C Lapa","year":"2021","unstructured":"Lapa C, et al. \u201cValue of PET imaging for radiation therapy. Strahlenther Onkol. 2021;197(9):1\u201323. https:\/\/doi.org\/10.1007\/s00066-021-01812-2.","journal-title":"Strahlenther Onkol"},{"issue":"5","key":"2491_CR4","doi-asserted-by":"publisher","first-page":"829","DOI":"10.1016\/j.eururo.2015.12.053","volume":"70","author":"M Eiber","year":"2016","unstructured":"Eiber M, et al. Simultaneous 68Ga-PSMA HBED-CC PET\/MRI Improves the Localization of Primary Prostate Cancer. Eur Urol. 2016;70(5):829\u201336. https:\/\/doi.org\/10.1016\/j.eururo.2015.12.053.","journal-title":"Eur Urol"},{"key":"2491_CR5","doi-asserted-by":"publisher","DOI":"10.3389\/fonc.2020.600690","volume":"10","author":"SKB Spohn","year":"2020","unstructured":"Spohn SKB, et al. Comparison of manual and semi-automatic [18F]PSMA-1007 PET based contouring techniques for intraprostatic tumor delineation in patients with primary prostate cancer and validation with histopathology as standard of reference. Front Oncol. 2020;10: 600690. https:\/\/doi.org\/10.3389\/fonc.2020.600690.","journal-title":"Front Oncol"},{"issue":"4","key":"2491_CR6","doi-asserted-by":"publisher","first-page":"219","DOI":"10.1016\/j.clon.2023.01.014","volume":"35","author":"S Hindocha","year":"2023","unstructured":"Hindocha S, et al. Artificial intelligence for radiotherapy auto-contouring: current use, perceptions of and barriers to implementation. Clin Oncol. 2023;35(4):219\u201326. https:\/\/doi.org\/10.1016\/j.clon.2023.01.014.","journal-title":"Clin Oncol"},{"issue":"3","key":"2491_CR7","doi-asserted-by":"publisher","first-page":"e105","DOI":"10.1016\/j.ijrobp.2022.07.905","volume":"114","author":"E Tillman-Schwartz","year":"2022","unstructured":"Tillman-Schwartz E, Tillman G, Tansky JY, Paly JJ, Efstathiou JA, Wang Y. Prospective analysis of a deep learning auto-contouring model for definitive radiation of localized prostate cancer. Int J Radiat Oncol Biol Phys. 2022;114(3):e105\u20136. https:\/\/doi.org\/10.1016\/j.ijrobp.2022.07.905.","journal-title":"Int J Radiat Oncol Biol Phys"},{"issue":"4","key":"2491_CR8","doi-asserted-by":"publisher","first-page":"577","DOI":"10.1016\/j.cpet.2021.06.001","volume":"16","author":"F Yousefirizi","year":"2021","unstructured":"Yousefirizi F, Jha AK, Brosch-Lenz J, Saboury B, Rahmim A. Toward high-throughput artificial intelligence-based segmentation in oncological PET imaging. PET Clinics. 2021;16(4):577\u201396. https:\/\/doi.org\/10.1016\/j.cpet.2021.06.001.","journal-title":"PET Clinics"},{"key":"2491_CR9","doi-asserted-by":"publisher","DOI":"10.3390\/cancers15245735","author":"SMH Hoque","year":"2023","unstructured":"Hoque SMH, et al. Clinical use of a commercial artificial intelligence-based software for autocontouring in radiation therapy: geometric performance and dosimetric impact. Cancers. 2023. https:\/\/doi.org\/10.3390\/cancers15245735.","journal-title":"Cancers"},{"key":"2491_CR10","doi-asserted-by":"publisher","DOI":"10.3389\/fonc.2023.1213068","author":"PJ Doolan","year":"2023","unstructured":"Doolan PJ, et al. A clinical evaluation of the performance of five commercial artificial intelligence contouring systems for radiotherapy. Front Oncol. 2023. https:\/\/doi.org\/10.3389\/fonc.2023.1213068.","journal-title":"Front Oncol"},{"key":"2491_CR11","doi-asserted-by":"publisher","DOI":"10.1016\/j.radonc.2023.109774","volume":"188","author":"JC Holzschuh","year":"2023","unstructured":"Holzschuh JC, et al. \u201cDeep learning based automated delineation of the intraprostatic gross tumour volume in PSMA-PET for patients with primary prostate cancer. Radiother Oncol. 2023;188: 109774. https:\/\/doi.org\/10.1016\/j.radonc.2023.109774.","journal-title":"Radiother Oncol"},{"issue":"6","key":"2491_CR12","doi-asserted-by":"publisher","first-page":"823","DOI":"10.2967\/jnumed.120.254623","volume":"62","author":"D Kostyszyn","year":"2021","unstructured":"Kostyszyn D, et al. Intraprostatic tumor segmentation on PSMA PET images in patients with primary prostate cancer with a convolutional neural network. J Nucl Med. 2021;62(6):823\u20138. https:\/\/doi.org\/10.2967\/jnumed.120.254623.","journal-title":"J Nucl Med"},{"key":"2491_CR13","doi-asserted-by":"publisher","unstructured":"S. Ghezzo et al., \u201cExternal validation of a convolutional neural network for the automatic segmentation of intraprostatic tumor lesions on 68Ga-PSMA PET images. Frontiers in Medicine, vol. 10, 2023, Accessed 13 Jan 2024. https:\/\/www.frontiersin.org\/articles\/https:\/\/doi.org\/10.3389\/fmed.2023.1133269","DOI":"10.3389\/fmed.2023.1133269"},{"key":"2491_CR14","doi-asserted-by":"publisher","DOI":"10.1038\/s41746-022-00592-y","author":"G Varoquaux","year":"2022","unstructured":"Varoquaux G, Cheplygina V. Machine learning for medical imaging: methodological failures and recommendations for the future. NPJ Digit Med. 2022. https:\/\/doi.org\/10.1038\/s41746-022-00592-y.","journal-title":"NPJ Digit. Med."},{"key":"2491_CR15","doi-asserted-by":"publisher","first-page":"245","DOI":"10.1016\/j.jclinepi.2015.04.005","volume":"69","author":"EW Steyerberg","year":"2016","unstructured":"Steyerberg EW, Harrell FE. Prediction models need appropriate internal, internal\u2013external, and external validation. J Clin Epidemiol. 2016;69:245\u20137. https:\/\/doi.org\/10.1016\/j.jclinepi.2015.04.005.","journal-title":"J Clin Epidemiol"},{"key":"2491_CR16","doi-asserted-by":"publisher","DOI":"10.1038\/s41592-020-01008-z","author":"F Isensee","year":"2021","unstructured":"Isensee F, Jaeger PF, Kohl SAA, Petersen J, Maier-Hein KH. nnU-Net: a self-configuring method for deep learning-based biomedical image segmentation. Nat Methods. 2021. https:\/\/doi.org\/10.1038\/s41592-020-01008-z.","journal-title":"Nat Methods"},{"issue":"3","key":"2491_CR17","doi-asserted-by":"publisher","first-page":"297","DOI":"10.2307\/1932409","volume":"26","author":"LR Dice","year":"1945","unstructured":"Dice LR. Measures of the amount of ecologic association between species. Ecology. 1945;26(3):297\u2013302. https:\/\/doi.org\/10.2307\/1932409.","journal-title":"Ecology"},{"key":"2491_CR18","doi-asserted-by":"publisher","DOI":"10.1038\/s41592-023-02151-z","author":"L Maier-Hein","year":"2024","unstructured":"Maier-Hein L, et al. Metrics reloaded: recommendations for image analysis validation. Nat Methods. 2024. https:\/\/doi.org\/10.1038\/s41592-023-02151-z.","journal-title":"Nat Methods"},{"issue":"9","key":"2491_CR19","doi-asserted-by":"publisher","first-page":"850","DOI":"10.1109\/34.232073","volume":"15","author":"DP Huttenlocher","year":"1993","unstructured":"Huttenlocher DP, Klanderman GA, Rucklidge WJ. Comparing images using the Hausdorff distance. IEEE Trans Pattern Anal Mach Intell. 1993;15(9):850\u201363. https:\/\/doi.org\/10.1109\/34.232073.","journal-title":"IEEE Trans Pattern Anal Mach Intell"},{"issue":"1150","key":"2491_CR20","doi-asserted-by":"publisher","first-page":"20230023","DOI":"10.1259\/bjr.20230023","volume":"96","author":"JW Gichoya","year":"2023","unstructured":"Gichoya JW, et al. AI pitfalls and what not to do: mitigating bias in AI. Br J Radiol. 2023;96(1150):20230023. https:\/\/doi.org\/10.1259\/bjr.20230023.","journal-title":"Br J Radiol"},{"key":"2491_CR21","doi-asserted-by":"publisher","DOI":"10.1038\/s42256-020-00257-z","author":"R Geirhos","year":"2020","unstructured":"Geirhos R, et al. Shortcut learning in deep neural networks. Nat Mach Intell. 2020. https:\/\/doi.org\/10.1038\/s42256-020-00257-z.","journal-title":"Nat Mach Intell"},{"issue":"5","key":"2491_CR22","doi-asserted-by":"publisher","DOI":"10.1148\/rg.220105","volume":"43","author":"Y Choi","year":"2023","unstructured":"Choi Y, et al. Translating AI to clinical practice: overcoming data shift with explainability. Radiographics. 2023;43(5): e220105. https:\/\/doi.org\/10.1148\/rg.220105.","journal-title":"Radiographics"},{"issue":"supplement 1","key":"2491_CR23","first-page":"399","volume":"60","author":"K Leung","year":"2019","unstructured":"Leung K, et al. A fully automated deep-learning based method for lesion segmentation in 18F-DCFPyL PSMA PET images of patients with prostate cancer. J Nucl Med. 2019;60(supplement 1):399\u2013399.","journal-title":"J Nucl Med"},{"issue":"1","key":"2491_CR24","doi-asserted-by":"publisher","first-page":"67","DOI":"10.1007\/s00259-022-05927-1","volume":"50","author":"J Kendrick","year":"2022","unstructured":"Kendrick J, Francis RJ, Hassan GM, Rowshanfarzad P, Ong JSL, Ebert MA. Fully automatic prognostic biomarker extraction from metastatic prostate lesion segmentations in whole-body [68Ga]Ga-PSMA-11 PET\/CT images. Eur J Nucl Med Mol Imaging. 2022;50(1):67\u201379. https:\/\/doi.org\/10.1007\/s00259-022-05927-1.","journal-title":"Eur J Nucl Med Mol Imaging"},{"key":"2491_CR25","doi-asserted-by":"publisher","DOI":"10.1007\/s00259-023-06551-3","author":"B Huang","year":"2023","unstructured":"Huang B, et al. Deep learning\u2013based whole-body characterization of prostate cancer lesions on [68Ga]Ga-PSMA-11 PET\/CT in patients with post-prostatectomy recurrence. Eur J Nucl Med Mol Imaging. 2023. https:\/\/doi.org\/10.1007\/s00259-023-06551-3.","journal-title":"Eur J Nucl Med Mol Imaging"},{"key":"2491_CR26","doi-asserted-by":"publisher","DOI":"10.1007\/s00259-023-06555-z","author":"E Jafari","year":"2023","unstructured":"Jafari E, et al. A convolutional neural network\u2013based system for fully automatic segmentation of whole-body [68Ga]Ga-PSMA PET images in prostate cancer. Eur J Nucl Med Mol Imaging. 2023. https:\/\/doi.org\/10.1007\/s00259-023-06555-z.","journal-title":"Eur J Nucl Med Mol Imaging"},{"key":"2491_CR27","doi-asserted-by":"publisher","DOI":"10.1038\/s41592-023-02150-0","author":"A Reinke","year":"2024","unstructured":"Reinke A, et al. Understanding metric-related pitfalls in image analysis validation. Nat Methods. 2024. https:\/\/doi.org\/10.1038\/s41592-023-02150-0.","journal-title":"Nat Methods"}],"container-title":["Radiation Oncology"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1186\/s13014-024-02491-w.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/article\/10.1186\/s13014-024-02491-w\/fulltext.html","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1186\/s13014-024-02491-w.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2024,8,7]],"date-time":"2024-08-07T11:11:32Z","timestamp":1723029092000},"score":1,"resource":{"primary":{"URL":"https:\/\/ro-journal.biomedcentral.com\/articles\/10.1186\/s13014-024-02491-w"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2024,8,7]]},"references-count":27,"journal-issue":{"issue":"1","published-online":{"date-parts":[[2024,12]]}},"alternative-id":["2491"],"URL":"https:\/\/doi.org\/10.1186\/s13014-024-02491-w","relation":{},"ISSN":["1748-717X"],"issn-type":[{"value":"1748-717X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2024,8,7]]},"assertion":[{"value":"4 April 2024","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"17 July 2024","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"7 August 2024","order":3,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"order":1,"name":"Ethics","group":{"name":"EthicsHeading","label":"Declarations"}},{"value":"Local ethics committees from all participating centers granted approval or exemption for the study.","order":2,"name":"Ethics","group":{"name":"EthicsHeading","label":"Ethics"}},{"value":"Not Applicable.","order":3,"name":"Ethics","group":{"name":"EthicsHeading","label":"Consent for publication"}},{"value":"The authors declare no competing interests.","order":4,"name":"Ethics","group":{"name":"EthicsHeading","label":"Conflict of Interest"}}],"article-number":"106"}}