{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,15]],"date-time":"2026-09-15T17:06:52Z","timestamp":1789492012393,"version":"build-2803163510"},"reference-count":47,"publisher":"Wiley","issue":"5","license":[{"start":{"date-parts":[[2026,5,21]],"date-time":"2026-05-21T00:00:00Z","timestamp":1779321600000},"content-version":"vor","delay-in-days":20,"URL":"http:\/\/onlinelibrary.wiley.com\/termsAndConditions#vor"},{"start":{"date-parts":[[2026,5,1]],"date-time":"2026-05-01T00:00:00Z","timestamp":1777593600000},"content-version":"tdm","delay-in-days":0,"URL":"http:\/\/doi.wiley.com\/10.1002\/tdm_license_1.1"}],"content-domain":{"domain":["aapm.onlinelibrary.wiley.com"],"crossmark-restriction":true},"short-container-title":["Medical Physics"],"published-print":{"date-parts":[[2026,5]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>\n                    <jats:bold>Background<\/jats:bold>\n                    : Radiomic imaging biomarkers are increasingly studied in oncology as a means to support disease prognosis and personalized treatment planning. While deep learning (DL) offers scalable alternatives to handcrafted radiomic features, DL\u2010derived biomarkers are sensitive to variations in image acquisition protocols and scanner hardware\u2014even within a single imaging modality. To ensure reliable and reproducible biomarker estimation, it is essential to (1) provide clinicians with quantitative uncertainty estimates associated with biomarker predictions, and (2) address acquisition\u2010induced variability through harmonization strategies that render consistent performance across diverse imaging conditions.\n                  <\/jats:p>\n                  <jats:p>\n                    <jats:bold>Purpose<\/jats:bold>\n                    : To evaluate uncertainty estimation as a reliability metric for biomarker prediction and to assess the role of image harmonization in improving cross\u2010scanner inference, we develop deep learning radiomics (DLR) models for joint estimation of biomarkers and associated uncertainties and apply linear harmonization filters to standardize imaging conditions.\n                  <\/jats:p>\n                  <jats:p>\n                    <jats:bold>Methods<\/jats:bold>\n                    : We constructed hybrid digital phantoms by embedding 20,000 virtual colorectal liver metastases into 20 clinical CT liver images to generate metastases\u2010laden simulated scans. The proposed DLR models jointly estimated a selected set of biomarkers and their associated aleatoric uncertainties from simulated images. Models were trained, validated, and tested using a 70:20:10 data split. Variability in CT image acquisition was modelled using two scanner types, two reconstruction kernels, and three x\u2010ray tube current settings. DLR performance was evaluated under direct inference (matched training and test scanners), cross\u2010scanner inference, and harmonized inference using linear harmonization filters. Quantitative evaluation was based on correlation coefficients and root mean squared errors (RMSEs).\n                  <\/jats:p>\n                  <jats:p>\n                    <jats:bold>Results<\/jats:bold>\n                    : The estimated uncertainties were consistently higher for biomarker predictions with larger deviations from ground truth. Excluding high\u2010uncertainty predictions improved concordance between predictions and ground truth biomarkers.\u00a0The proportion of uncertain predictions increased under cross\u2010scanner inference relative to direct inference, indicating reduced biomarker reliability under heterogeneous imaging conditions. Furthermore, application of harmonization filters reduced RMSEs by an average of 43% across biomarkers and experiments during cross\u2010scanner inference, demonstrating improved cross\u2010scanner consistency.\n                  <\/jats:p>\n                  <jats:p>\n                    <jats:bold>Conclusions<\/jats:bold>\n                    : In this controlled in\u2010silico study, uncertainty estimation provided a practical reliability metric for DL\u2010based radiomic biomarker prediction, while image harmonization improved reproducibility across heterogeneous acquisition conditions. These findings demonstrate methodological feasibility within a simulation\u2010based framework but have not yet been validated on clinical CT data, motivating future clinical validation and translation.\n                  <\/jats:p>","DOI":"10.1002\/mp.70500","type":"journal-article","created":{"date-parts":[[2026,5,22]],"date-time":"2026-05-22T06:50:36Z","timestamp":1779432636000},"update-policy":"https:\/\/doi.org\/10.1002\/crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["Towards quality control and harmonization of deep learning CT radiomics: An in\u2010silico feasibility study with virtual colorectal liver metastases"],"prefix":"10.1002","volume":"53","author":[{"given":"Mamatha","family":"Venugopal","sequence":"first","affiliation":[{"name":"Technology &amp; Innovation Center GE HealthCare, JFWTC  Bengaluru Karnataka India"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Sathish","family":"Ramani","sequence":"additional","affiliation":[{"name":"Technology &amp; Innovation Center GE HealthCare  Niskayuna New York USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jacob J.","family":"Peoples","sequence":"additional","affiliation":[{"name":"Department of Radiology Memorial Sloan Kettering Cancer Center  New York New York USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Richard K. G.","family":"Do","sequence":"additional","affiliation":[{"name":"Department of Radiology Memorial Sloan Kettering Cancer Center  New York New York USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Amber L.","family":"Simpson","sequence":"additional","affiliation":[{"name":"School of Computing\/Department of Biomedical and Molecular Sciences Queen's University Ontario Canada"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ge","family":"Wang","sequence":"additional","affiliation":[{"name":"Department of Biomedical Engineering Rensselaer Polytechnic Institute  Troy New York USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Bruno","family":"De Man","sequence":"additional","affiliation":[{"name":"Technology &amp; Innovation Center GE HealthCare  Niskayuna New York USA"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"311","published-online":{"date-parts":[[2026,5,21]]},"reference":[{"key":"e_1_2_8_2_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.ejca.2011.11.036"},{"key":"e_1_2_8_3_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.mri.2012.06.010"},{"key":"e_1_2_8_4_1","doi-asserted-by":"publisher","DOI":"10.1038\/nrclinonc.2017.141"},{"key":"e_1_2_8_5_1","doi-asserted-by":"publisher","DOI":"10.1038\/ncomms5006"},{"key":"e_1_2_8_6_1","doi-asserted-by":"publisher","DOI":"10.1038\/s41598\u2010022\u201018085\u2010z"},{"key":"e_1_2_8_7_1","doi-asserted-by":"publisher","DOI":"10.21037\/tcr.2018.05.02"},{"key":"e_1_2_8_8_1","doi-asserted-by":"publisher","DOI":"10.1148\/radiol.2015151169"},{"key":"e_1_2_8_9_1","doi-asserted-by":"publisher","DOI":"10.1002\/mp.13678"},{"key":"e_1_2_8_10_1","doi-asserted-by":"publisher","DOI":"10.1038\/s41598\u2010017\u201005848\u20102"},{"key":"e_1_2_8_11_1","doi-asserted-by":"publisher","DOI":"10.1002\/ctm2.570"},{"key":"e_1_2_8_12_1","doi-asserted-by":"publisher","DOI":"10.1002\/mp.14563"},{"key":"e_1_2_8_13_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.eclinm.2023.102271"},{"key":"e_1_2_8_14_1","doi-asserted-by":"publisher","DOI":"10.1038\/s41598\u2010024\u201068158\u20104"},{"key":"e_1_2_8_15_1","doi-asserted-by":"publisher","DOI":"10.1002\/mp.13385"},{"key":"e_1_2_8_16_1","doi-asserted-by":"publisher","DOI":"10.1002\/acm2.12666"},{"key":"e_1_2_8_17_1","doi-asserted-by":"publisher","DOI":"10.1002\/mp.15491"},{"key":"e_1_2_8_18_1","unstructured":"KendallA GalY.What uncertainties do we need in Bayesian deep learning for computer vision?In:Proceedings of the 31st International Conference on Neural Information Processing Systems (NIPS\u201917).Curran Associates Inc.;2017:5580\u20105590."},{"key":"e_1_2_8_19_1","doi-asserted-by":"publisher","DOI":"10.1016\/B978-0-12-824349-7.00033-5"},{"key":"e_1_2_8_20_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.inffus.2021.05.008"},{"key":"e_1_2_8_21_1","unstructured":"YarinG GhahramaniZ.Dropout as a Bayesian approximation: representing model uncertainty in deep learning. In:Proceedings of The 33rd International Conference on Machine Learning. Vol48.PMLR;2016:1050\u20101059.https:\/\/proceedings.mlr.press\/v48\/gal16.html"},{"key":"e_1_2_8_22_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.media.2019.101557"},{"key":"e_1_2_8_23_1","unstructured":"DenkerJS LeCunY.Transforming neural\u2010net output levels to probability distributions. In:Proceedings of the 4th International Conference on Neural Information Processing Systems (NIPS\u201990).Morgan Kaufmann Publishers Inc.;1990:853\u2010859."},{"key":"e_1_2_8_24_1","unstructured":"LiY TurnerRE.R\u00e9nyi divergence variational inference. In:Proceedings of the 30th International Conference on Neural Information Processing Systems (NIPS'16).Curran Associates Inc.;2016:1081\u20101089.10.5555\/3157096.3157217"},{"key":"e_1_2_8_25_1","unstructured":"GalY GhahramaniZ.Bayesian Convolutional Neural Networks with Bernoulli Approximate Variational Inference.arXiv. Preprint posted online January 18 2016. doi:10.48550\/arXiv.1506.02158"},{"key":"e_1_2_8_26_1","unstructured":"LakshminarayananB PritzelA BlundellC.Simple and scalable predictive uncertainty estimation using deep ensembles. InProceedings of the 31st International Conference on Neural Information Processing Systems (NIPS 2017).Curran Associates Inc.;2017:6405\u20106416.https:\/\/dl.acm.org\/doi\/10.5555\/3295222.3295387"},{"key":"e_1_2_8_27_1","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v34i04.5849"},{"key":"e_1_2_8_28_1","doi-asserted-by":"crossref","unstructured":"WangG LiW OurselinS VercauterenT.Automatic brain tumor segmentation using convolutional neural networks with test\u2010time augmentation. In:CrimiA BakasS KuijfH KeyvanF ReyesM Van WalsumT eds.Brainlesion: Glioma Multiple Sclerosis Stroke and Traumatic Brain Injuries. Vol11384.Springer International Publishing;2019:61\u201072. Lecture Notes in Computer Science. doi:10.1007\/978\u20103\u2010030\u201011726\u20109_6","DOI":"10.1007\/978-3-030-11726-9_6"},{"key":"e_1_2_8_29_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.media.2019.101619"},{"key":"e_1_2_8_30_1","doi-asserted-by":"crossref","unstructured":"HuS WorrallD KnegtS VeelingB HuismanH WellingM.Supervised uncertainty quantification for segmentation with multiple annotations. In:ShenD LiuT PetersTM et\u00a0al. eds.Medical Image Computing and Computer Assisted Intervention \u2013 MICCAI 2019. Vol11765.Springer International Publishing;2019:137\u2010145. Lecture Notes in Computer Science. doi:10.1007\/978\u20103\u2010030\u201032245\u20108_16","DOI":"10.1007\/978-3-030-32245-8_16"},{"key":"e_1_2_8_31_1","doi-asserted-by":"publisher","DOI":"10.3390\/jimaging8110303"},{"key":"e_1_2_8_32_1","doi-asserted-by":"publisher","DOI":"10.1038\/s41598\u2010022\u201008412\u20109"},{"key":"e_1_2_8_33_1","doi-asserted-by":"publisher","DOI":"10.1200\/CCI.22.00023"},{"key":"e_1_2_8_34_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.ejmp.2021.03.009"},{"key":"e_1_2_8_35_1","doi-asserted-by":"publisher","DOI":"10.1167\/16.12.326"},{"key":"e_1_2_8_36_1","doi-asserted-by":"publisher","DOI":"10.2967\/jnumed.119.228106"},{"key":"e_1_2_8_37_1","doi-asserted-by":"publisher","DOI":"10.1097\/SLA.0b013e31815aa2c2"},{"key":"e_1_2_8_38_1","doi-asserted-by":"publisher","DOI":"10.1007\/s11547\u2010017\u20100838\u20103"},{"key":"e_1_2_8_39_1","doi-asserted-by":"publisher","DOI":"10.1007\/s00330\u2010021\u201008303\u2010z"},{"key":"e_1_2_8_40_1","doi-asserted-by":"publisher","DOI":"10.1158\/0008\u20105472.CAN\u201017\u20100339"},{"key":"e_1_2_8_41_1","doi-asserted-by":"publisher","DOI":"10.1148\/radiol.2020191145"},{"key":"e_1_2_8_42_1","doi-asserted-by":"publisher","DOI":"10.1186\/s42492\u2010024\u201000161\u2010y"},{"key":"e_1_2_8_43_1","unstructured":"MorganMA.Liver Imaging Reporting and Data System (LI\u2010RADS). Reference article Radiopaedia.org(Accessed on 15 May2026). doi:10.53347\/rID\u201032465"},{"key":"e_1_2_8_44_1","doi-asserted-by":"crossref","unstructured":"De ManB BasuS ChandraN et\u00a0al.CatSim: a new computer assisted tomography simulation environment. In:HsiehJ FlynnMJ eds.Medical Imaging 2007: Physics of Medical Imaging.SPIE;2007:65102G.Proceedings of SPIE; vol6510. doi:10.1117\/12.710713","DOI":"10.1117\/12.710713"},{"key":"e_1_2_8_45_1","unstructured":"COLORECTAL\u2010LIVER\u2010METASTASES.The Cancer Imaging Archive (TCIA). Accessed January 23 2026.https:\/\/www.cancerimagingarchive.net\/collection\/colorectal\u2010liver\u2010metastases\/"},{"key":"e_1_2_8_46_1","doi-asserted-by":"crossref","unstructured":"HeK ZhangX RenS SunJ.Deep residual learning for image recognition. In:2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).IEEE;2016:770\u2010778. doi:10.1109\/CVPR.2016.90","DOI":"10.1109\/CVPR.2016.90"},{"key":"e_1_2_8_47_1","unstructured":"ChungY CharI GuoH SchneiderJ NeiswangerW.Uncertainty toolbox: an open\u2010source library for assessing visualizing and improving uncertainty quantification.arXiv. Preprint posted online September 21 2021"},{"key":"e_1_2_8_48_1","doi-asserted-by":"publisher","DOI":"10.1007\/s10994\u2010021\u201005946\u20103"}],"container-title":["Medical Physics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/aapm.onlinelibrary.wiley.com\/doi\/pdf\/10.1002\/mp.70500","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/aapm.onlinelibrary.wiley.com\/doi\/full-xml\/10.1002\/mp.70500","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/aapm.onlinelibrary.wiley.com\/doi\/pdf\/10.1002\/mp.70500","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,15]],"date-time":"2026-09-15T16:41:51Z","timestamp":1789490511000},"score":1,"resource":{"primary":{"URL":"https:\/\/aapm.onlinelibrary.wiley.com\/doi\/10.1002\/mp.70500"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,5]]},"references-count":47,"journal-issue":{"issue":"5","published-print":{"date-parts":[[2026,5]]}},"alternative-id":["10.1002\/mp.70500"],"URL":"https:\/\/doi.org\/10.1002\/mp.70500","archive":["Portico"],"relation":{},"ISSN":["0094-2405","2473-4209"],"issn-type":[{"value":"0094-2405","type":"print"},{"value":"2473-4209","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,5]]},"assertion":[{"value":"2025-11-06","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2026-04-28","order":2,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2026-05-21","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}],"article-number":"e70500"}}