000307499 001__ 307499
000307499 005__ 20260106120338.0
000307499 0247_ $$2doi$$a10.2967/jnumed.125.270260
000307499 0247_ $$2pmid$$apmid:41469162
000307499 0247_ $$2ISSN$$a0097-9058
000307499 0247_ $$2ISSN$$a0022-3123
000307499 0247_ $$2ISSN$$a0161-5505
000307499 0247_ $$2ISSN$$a1535-5667
000307499 0247_ $$2ISSN$$a2159-662X
000307499 037__ $$aDKFZ-2026-00004
000307499 041__ $$aEnglish
000307499 082__ $$a610
000307499 1001_ $$aDexl, Jakob$$b0
000307499 245__ $$aAutoPET Challenge on Fully Automated Lesion Segmentation in Oncologic PET/CT Imaging, Part 2: Domain Generalization.
000307499 260__ $$aNew York, NY$$bSoc.$$c2025
000307499 3367_ $$2DRIVER$$aarticle
000307499 3367_ $$2DataCite$$aOutput Types/Journal article
000307499 3367_ $$0PUB:(DE-HGF)16$$2PUB:(DE-HGF)$$aJournal Article$$bjournal$$mjournal$$s1767626643_2785347
000307499 3367_ $$2BibTeX$$aARTICLE
000307499 3367_ $$2ORCID$$aJOURNAL_ARTICLE
000307499 3367_ $$00$$2EndNote$$aJournal Article
000307499 500__ $$aepub
000307499 520__ $$aThis article reports the results of the second iteration of the autoPET challenge on automated lesion segmentation in whole-body PET/CT, held in conjunction with the 26th International Conference on Medical Image Computing and Computer Assisted Intervention in 2023. In contrast to the first autoPET challenge, which served as a proof of concept, this study investigates whether machine learning-based segmentation models trained on data from a single source can maintain performance across clinically relevant variations in PET/CT data, reflecting the demands of real-world deployment. Methods: A comprehensive biomedical segmentation challenge on PET/CT domain generalization was designed and conducted. Participants were tasked to train machine learning models on annotated whole-body 18F-FDG data (n = 1,014). These models were then evaluated on a test set of 200 samples from 5 clinically relevant domains, including variations in institutions, pathologies, and populations and a different tracer. Performance was measured in terms of average dice similarity coefficient, average false-positive volume, and average false-negative volume. The best-performing teams were awarded in 3 categories. Furthermore, a detailed analysis was conducted after the challenge, examining results across domains and unique instances, along with a ranking analysis. Results: Generalization from a single-source domain remains a significant challenge. Seventeen international teams successfully participated in the challenge. The best-performing team reached an average dice similarity coefficient of 0.5038, a mean false-positive volume of 87.8388 mL, and a mean false-negative volume of 8.4154 mL on the test set. nnU-Net was the most commonly used framework, with most participants using a 3-dimensional U-Net. Despite competitive in-domain results, out-of-domain performance deteriorated substantially, particularly on pediatric and prostate-specific membrane antigen data. Detailed error analysis revealed frequent false-positives due to physiologic uptake and decreased sensitivity in detecting small or low-uptake lesions. A majority-vote ensemble offered minimal performance gains, whereas an oracle ensemble indicates hypothetical gains. Ranking analysis showed no single team consistently outperformed all others across ranking schemes. Conclusion: The second autoPET challenge provides a comprehensive evaluation of the current state of automated PET/CT tumor segmentation, highlighting both progress and persistent challenges of single-source domain generalization and the need for diverse public datasets to enhance algorithm robustness.
000307499 536__ $$0G:(DE-HGF)POF4-315$$a315 - Bildgebung und Radioonkologie (POF4-315)$$cPOF4-315$$fPOF IV$$x0
000307499 588__ $$aDataset connected to CrossRef, PubMed, , Journals: inrepo02.dkfz.de
000307499 650_7 $$2Other$$aPET/CT
000307499 650_7 $$2Other$$abiomedical image analysis challenge
000307499 650_7 $$2Other$$adeep learning
000307499 650_7 $$2Other$$adomain generalization
000307499 650_7 $$2Other$$aoncology
000307499 650_7 $$2Other$$asegmentation
000307499 7001_ $$aGatidis, Sergios$$b1
000307499 7001_ $$aFrüh, Marcel$$b2
000307499 7001_ $$aJeblick, Katharina$$b3
000307499 7001_ $$aMittermeier, Andreas$$b4
000307499 7001_ $$aStüber, Anna Theresa$$b5
000307499 7001_ $$aSchachtner, Balthasar$$b6
000307499 7001_ $$aTopalis, Johanna$$b7
000307499 7001_ $$aFabritius, Matthias P$$b8
000307499 7001_ $$aGu, Sijing$$b9
000307499 7001_ $$aMurugesan, Gowtham Krishnan$$b10
000307499 7001_ $$aVanOss, Jeff$$b11
000307499 7001_ $$aYe, Jin$$b12
000307499 7001_ $$aHe, Junjun$$b13
000307499 7001_ $$aAlloula, Anissa$$b14
000307499 7001_ $$aPapież, Bartłomiej W$$b15
000307499 7001_ $$aMesbah, Zacharia$$b16
000307499 7001_ $$aModzelewski, Romain$$b17
000307499 7001_ $$aHadlich, Matthias$$b18
000307499 7001_ $$aMarinov, Zdravko$$b19
000307499 7001_ $$aStiefelhagen, Rainer$$b20
000307499 7001_ $$0P:(DE-He78)7ea9af59d03ec7deb982a0e0562358fa$$aIsensee, Fabian$$b21$$udkfz
000307499 7001_ $$0P:(DE-He78)33c74005e1ce56f7025c4f6be15321b3$$aMaier-Hein, Klaus H$$b22$$udkfz
000307499 7001_ $$aGaldran, Adrian$$b23
000307499 7001_ $$aNikolaou, Konstantin$$b24
000307499 7001_ $$0P:(DE-HGF)0$$ala Fougère, Christian$$b25
000307499 7001_ $$aKim, Moon$$b26
000307499 7001_ $$aKallenberg, Nico$$b27
000307499 7001_ $$0P:(DE-He78)ec13544e7fd4c62ac008490a4547e990$$aKleesiek, Jens$$b28$$udkfz
000307499 7001_ $$aHerrmann, Ken$$b29
000307499 7001_ $$aWerner, Rudolf$$b30
000307499 7001_ $$aIngrisch, Michael$$b31
000307499 7001_ $$aCyran, Clemens C$$b32
000307499 7001_ $$aKüstner, Thomas$$b33
000307499 773__ $$0PERI:(DE-600)2040222-3$$a10.2967/jnumed.125.270260$$gp. jnumed.125.270260 -$$pnn$$tJournal of nuclear medicine$$vnn$$x0097-9058$$y2025
000307499 909CO $$ooai:inrepo02.dkfz.de:307499$$pVDB
000307499 915__ $$0StatID:(DE-HGF)0100$$2StatID$$aJCR$$bJ NUCL MED : 2022$$d2025-01-01
000307499 915__ $$0StatID:(DE-HGF)0200$$2StatID$$aDBCoverage$$bSCOPUS$$d2025-01-01
000307499 915__ $$0StatID:(DE-HGF)0300$$2StatID$$aDBCoverage$$bMedline$$d2025-01-01
000307499 915__ $$0StatID:(DE-HGF)0199$$2StatID$$aDBCoverage$$bClarivate Analytics Master Journal List$$d2025-01-01
000307499 915__ $$0StatID:(DE-HGF)1050$$2StatID$$aDBCoverage$$bBIOSIS Previews$$d2025-01-01
000307499 915__ $$0StatID:(DE-HGF)0160$$2StatID$$aDBCoverage$$bEssential Science Indicators$$d2025-01-01
000307499 915__ $$0StatID:(DE-HGF)1030$$2StatID$$aDBCoverage$$bCurrent Contents - Life Sciences$$d2025-01-01
000307499 915__ $$0StatID:(DE-HGF)1190$$2StatID$$aDBCoverage$$bBiological Abstracts$$d2025-01-01
000307499 915__ $$0StatID:(DE-HGF)1110$$2StatID$$aDBCoverage$$bCurrent Contents - Clinical Medicine$$d2025-01-01
000307499 915__ $$0StatID:(DE-HGF)0113$$2StatID$$aWoS$$bScience Citation Index Expanded$$d2025-01-01
000307499 915__ $$0StatID:(DE-HGF)0150$$2StatID$$aDBCoverage$$bWeb of Science Core Collection$$d2025-01-01
000307499 915__ $$0StatID:(DE-HGF)9905$$2StatID$$aIF >= 5$$bJ NUCL MED : 2022$$d2025-01-01
000307499 9141_ $$y2025
000307499 9101_ $$0I:(DE-588b)2036810-0$$6P:(DE-He78)7ea9af59d03ec7deb982a0e0562358fa$$aDeutsches Krebsforschungszentrum$$b21$$kDKFZ
000307499 9101_ $$0I:(DE-588b)2036810-0$$6P:(DE-He78)33c74005e1ce56f7025c4f6be15321b3$$aDeutsches Krebsforschungszentrum$$b22$$kDKFZ
000307499 9101_ $$0I:(DE-588b)2036810-0$$6P:(DE-HGF)0$$aDeutsches Krebsforschungszentrum$$b25$$kDKFZ
000307499 9101_ $$0I:(DE-588b)2036810-0$$6P:(DE-He78)ec13544e7fd4c62ac008490a4547e990$$aDeutsches Krebsforschungszentrum$$b28$$kDKFZ
000307499 9131_ $$0G:(DE-HGF)POF4-315$$1G:(DE-HGF)POF4-310$$2G:(DE-HGF)POF4-300$$3G:(DE-HGF)POF4$$4G:(DE-HGF)POF$$aDE-HGF$$bGesundheit$$lKrebsforschung$$vBildgebung und Radioonkologie$$x0
000307499 9201_ $$0I:(DE-He78)E230-20160331$$kE230$$lE230 Medizinische Bildverarbeitung$$x0
000307499 9201_ $$0I:(DE-He78)TU01-20160331$$kTU01$$lDKTK Koordinierungsstelle Tübingen$$x1
000307499 9201_ $$0I:(DE-He78)ED01-20160331$$kED01$$lDKTK Koordinierungsstelle Essen/Düsseldorf$$x2
000307499 980__ $$ajournal
000307499 980__ $$aVDB
000307499 980__ $$aI:(DE-He78)E230-20160331
000307499 980__ $$aI:(DE-He78)TU01-20160331
000307499 980__ $$aI:(DE-He78)ED01-20160331
000307499 980__ $$aUNRESTRICTED