000273976 001__ 273976
000273976 005__ 20250120103259.0
000273976 0247_ $$2doi$$a10.1016/j.ejrad.2024.111825
000273976 0247_ $$2pmid$$apmid:39657546
000273976 0247_ $$2ISSN$$a0720-048X
000273976 0247_ $$2ISSN$$a1872-7727
000273976 0247_ $$2altmetric$$aaltmetric:170943835
000273976 037__ $$aDZNE-2024-01425
000273976 041__ $$aEnglish
000273976 082__ $$a610
000273976 1001_ $$aStüber, Anna Theresa$$b0
000273976 245__ $$aReplication study of PD-L1 status prediction in NSCLC using PET/CT radiomics.
000273976 260__ $$aAmsterdam [u.a.]$$bElsevier Science$$c2025
000273976 3367_ $$2DRIVER$$aarticle
000273976 3367_ $$2DataCite$$aOutput Types/Journal article
000273976 3367_ $$0PUB:(DE-HGF)16$$2PUB:(DE-HGF)$$aJournal Article$$bjournal$$mjournal$$s1737365530_4997
000273976 3367_ $$2BibTeX$$aARTICLE
000273976 3367_ $$2ORCID$$aJOURNAL_ARTICLE
000273976 3367_ $$00$$2EndNote$$aJournal Article
000273976 520__ $$aThis study investigates the predictive capability of radiomics in determining programmed cell death ligand 1 (PD-L1) expression (>=1%) status in non-small cell lung cancer (NSCLC) patients using a newly collected [18F]FDG PET/CT dataset. We aimed to replicate and validate the radiomics-based machine learning (ML) model proposed by Zhao et al. [1] predicting PD-L1 status from PET/CT-imaging. An independent cohort of 254 NSCLC patients underwent [18F]FDG PET/CT imaging, with primary tumor segmentation conducted using lung tissue window (LTW) and more conservative soft tissue window (STW) methods. Radiomics models ('Rad-score' and 'complex model') and a clinical-stage model from Zhao et al. were evaluated via 10-fold cross-validation and AUC analysis, alongside a benchmark-study comparing different ML-model pipelines. Clinicopathological data were collected from medical records. On our data, the Rad-score model yielded mean AUCs of 0.593 (STW) and 0.573 (LTW), below Zhao et al.'s 0.761. The complex model achieved mean AUCs of 0.505 (STW) and 0.519 (LTW), lower than Zhao et al.'s 0.769. The clinical model showed a mean AUC of 0.555, below Zhao et al.'s 0.64. All models performed significantly lower than Zhao et al.'s findings. Our benchmark study on four ML pipelines revealed consistently low performance across all configurations. Our study failed to replicate original findings, suggesting poor model performance and questioning predictive value of radiomics features in classifying PD-L1 expression from PET/CT imaging. These results highlight challenges in replicating radiomics-based ML models and stress the need for rigorous validation.
000273976 536__ $$0G:(DE-HGF)POF4-352$$a352 - Disease Mechanisms (POF4-352)$$cPOF4-352$$fPOF IV$$x0
000273976 588__ $$aDataset connected to CrossRef, PubMed, , Journals: pub.dzne.de
000273976 650_7 $$2Other$$aMachine learning benchmark
000273976 650_7 $$2Other$$aNSCLC
000273976 650_7 $$2Other$$aPD-L1
000273976 650_7 $$2Other$$aPET/CT imaging data
000273976 650_7 $$2Other$$aRadiomics
000273976 650_7 $$2Other$$aReplication study
000273976 7001_ $$aHeimer, Maurice M$$b1
000273976 7001_ $$aTa, Johanna$$b2
000273976 7001_ $$aFabritius, Matthias P$$b3
000273976 7001_ $$aHoppe, Boj F$$b4
000273976 7001_ $$aSheikh, Gabriel$$b5
000273976 7001_ $$0P:(DE-2719)9001539$$aBrendel, Matthias$$b6$$udzne
000273976 7001_ $$aUnterrainer, Lena$$b7
000273976 7001_ $$aJurmeister, Philip$$b8
000273976 7001_ $$aTufman, Amanda$$b9
000273976 7001_ $$aRicke, Jens$$b10
000273976 7001_ $$aCyran, Clemens C$$b11
000273976 7001_ $$aIngrisch, Michael$$b12
000273976 773__ $$0PERI:(DE-600)2005350-2$$a10.1016/j.ejrad.2024.111825$$gVol. 183, p. 111825 -$$p111825$$tEuropean journal of radiology$$v183$$x0720-048X$$y2025
000273976 8564_ $$uhttps://pub.dzne.de/record/273976/files/DZNE-2024-01425%20SUP.pdf
000273976 8564_ $$uhttps://pub.dzne.de/record/273976/files/DZNE-2024-01425_Restricted.pdf
000273976 8564_ $$uhttps://pub.dzne.de/record/273976/files/DZNE-2024-01425%20SUP.pdf?subformat=pdfa$$xpdfa
000273976 8564_ $$uhttps://pub.dzne.de/record/273976/files/DZNE-2024-01425_Restricted.pdf?subformat=pdfa$$xpdfa
000273976 909CO $$ooai:pub.dzne.de:273976$$pVDB
000273976 9101_ $$0I:(DE-588)1065079516$$6P:(DE-2719)9001539$$aDeutsches Zentrum für Neurodegenerative Erkrankungen$$b6$$kDZNE
000273976 9131_ $$0G:(DE-HGF)POF4-352$$1G:(DE-HGF)POF4-350$$2G:(DE-HGF)POF4-300$$3G:(DE-HGF)POF4$$4G:(DE-HGF)POF$$aDE-HGF$$bGesundheit$$lNeurodegenerative Diseases$$vDisease Mechanisms$$x0
000273976 9141_ $$y2025
000273976 915__ $$0StatID:(DE-HGF)0420$$2StatID$$aNationallizenz$$d2023-08-26$$wger
000273976 915__ $$0StatID:(DE-HGF)0100$$2StatID$$aJCR$$bEUR J RADIOL : 2022$$d2023-08-26
000273976 915__ $$0StatID:(DE-HGF)0200$$2StatID$$aDBCoverage$$bSCOPUS$$d2023-08-26
000273976 915__ $$0StatID:(DE-HGF)0300$$2StatID$$aDBCoverage$$bMedline$$d2023-08-26
000273976 915__ $$0StatID:(DE-HGF)0600$$2StatID$$aDBCoverage$$bEbsco Academic Search$$d2023-08-26
000273976 915__ $$0StatID:(DE-HGF)0030$$2StatID$$aPeer Review$$bASC$$d2023-08-26
000273976 915__ $$0StatID:(DE-HGF)0199$$2StatID$$aDBCoverage$$bClarivate Analytics Master Journal List$$d2023-08-26
000273976 915__ $$0StatID:(DE-HGF)0113$$2StatID$$aWoS$$bScience Citation Index Expanded$$d2023-08-26
000273976 915__ $$0StatID:(DE-HGF)0150$$2StatID$$aDBCoverage$$bWeb of Science Core Collection$$d2023-08-26
000273976 915__ $$0StatID:(DE-HGF)0160$$2StatID$$aDBCoverage$$bEssential Science Indicators$$d2023-08-26
000273976 915__ $$0StatID:(DE-HGF)1110$$2StatID$$aDBCoverage$$bCurrent Contents - Clinical Medicine$$d2023-08-26
000273976 915__ $$0StatID:(DE-HGF)9900$$2StatID$$aIF < 5$$d2023-08-26
000273976 9201_ $$0I:(DE-2719)1110007$$kAG Haass$$lMolecular Neurodegeneration$$x0
000273976 980__ $$ajournal
000273976 980__ $$aVDB
000273976 980__ $$aI:(DE-2719)1110007
000273976 980__ $$aUNRESTRICTED