{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,14]],"date-time":"2026-09-14T13:02:44Z","timestamp":1789390964737,"version":"build-2803163510"},"reference-count":34,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,9,14]],"date-time":"2026-09-14T00:00:00Z","timestamp":1789344000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100007409","name":"Chengde Medical University","doi-asserted-by":"publisher","id":[{"id":"10.13039\/501100007409","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100003482","name":"Department of Education of Hebei Province","doi-asserted-by":"publisher","id":[{"id":"10.13039\/501100003482","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Oncol."],"abstract":"<jats:sec>\n                    <jats:title>Introduction<\/jats:title>\n                    <jats:p>Multicenter radiomic studies are fundamentally constrained by inter-center feature variability arising from differences in scanner hardware, acquisition protocols, and reconstruction parameters. Existing harmonization methods, including ComBat, treat these sources of variability as undifferentiated batch noise and primarily correct feature distribution location, leaving scale-level heterogeneity unaddressed. This study proposes Category-ComBat with Distribution Coefficient (Category-ComBat-\u03c9), a two-level harmonization framework for multicenter CT radiomics.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>A total of 1,789 pulmonary nodule CT images were retrospectively collected from three public datasets (LIDC-IDRI, NSCLC-Radiomics, and Lung-PET-CT-Dx), with 93 radiomic features extracted per case. Category-ComBat reconstructs the ComBat framework by explicitly separating imaging-related variability into systematic components, including scanner type, acquisition protocol, and reconstruction kernel, and random components. Individualized least-squares estimation was used to estimate systematic biases, together with category-stratified mean normalization. A distribution coefficient \u03c9 \u2208 (0, 1] was introduced as an orthogonal scale-adjustment mechanism to adaptively compress feature distribution spread without perturbing category-level means. Performance was evaluated using Wilcoxon tests, Q-Q plots, three-dimensional principal component analysis (PCA), probability density plots, coefficient of variation (COV), intraclass correlation coefficient (ICC), and downstream machine learning classification experiments.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>Category-ComBat effectively removed systematic inter-center variability while preserving benign-malignant discriminative separation. Introduction of \u03c9 progressively improved feature reproducibility and inter-center consistency as \u03c9 decreased. The best overall performance, evaluated using ICC, COV, and classification AUC, was observed for \u03c9 values between 0.5 and 0.9. Scale compression remained independent of location adjustment, confirming the orthogonality of the two-level design. Category-ComBat achieved the highest classification performance, with the mean AUC increasing from 0.84 for the Original features and 0.82 for ComBat to 0.89 after Category-ComBat harmonization.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Discussion<\/jats:title>\n                    <jats:p>Category-ComBat-\u03c9 provides a principled approach to multicenter radiomic harmonization by addressing both systematic imaging-related confounding and scale heterogeneity. Its tunable distribution coefficient enables flexible adjustment of feature distribution spread while preserving category-level means, providing an adaptable framework for different multicenter radiomic study requirements.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.3389\/fonc.2026.1896994","type":"journal-article","created":{"date-parts":[[2026,9,14]],"date-time":"2026-09-14T12:09:15Z","timestamp":1789387755000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["Category-ComBat with distribution coefficient: a scale-adaptive harmonization algorithm for multicenter radiomic feature standardization"],"prefix":"10.3389","volume":"16","author":[{"given":"Xiaolei","family":"Zhang","sequence":"first","affiliation":[{"name":"Department of Biomedical Engineering, Chengde Medical University","place":["Chengde, China"]},{"name":"Faculty of Engineering, Universiti Putra Malaysia","place":["Serdang, Malaysia"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xiaona","family":"Wang","sequence":"additional","affiliation":[{"name":"Department of Radiology, Dalian Women and Children\u2019s Medical Group","place":["Dalian, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Wei","family":"Wang","sequence":"additional","affiliation":[{"name":"The Affiliated Hospital of Chengde Medical University","place":["Chengde, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Lijun","family":"Lu","sequence":"additional","affiliation":[{"name":"School of Biomedical Engineering and Guangdong Provincial Key Laboratory of Medical Image Processing, Southern Medical University","place":["Guangzhou, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Mohammad Hamiruce","family":"Marhaban","sequence":"additional","affiliation":[{"name":"Faculty of Engineering, Universiti Putra Malaysia","place":["Serdang, Malaysia"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xianling","family":"Dong","sequence":"additional","affiliation":[{"name":"Department of Biomedical Engineering, Chengde Medical University","place":["Chengde, China"]},{"name":"Faculty of Engineering, Universiti Putra Malaysia","place":["Serdang, Malaysia"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yanli","family":"Liu","sequence":"additional","affiliation":[{"name":"Department of Biomedical Engineering, Chengde Medical University","place":["Chengde, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2026,9,14]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","first-page":"61","DOI":"10.21203\/rs.3.rs-4113664\/v1","article-title":"Preoperative CT-based radiomic prognostic index to predict the benefit of postoperative radiotherapy in patients with non-small cell lung cancer: a multicenter study","volume":"24","author":"Ma","year":"2024","journal-title":"Cancer Imaging"},{"key":"B2","doi-asserted-by":"publisher","first-page":"16","DOI":"10.1186\/s40644-024-00659-x","article-title":"MR radiomics predicts pathological complete response of esophageal squamous cell carcinoma after neoadjuvant chemoradiotherapy: a multicenter study","volume":"24","author":"Liu","year":"2024","journal-title":"Cancer Imaging"},{"key":"B3","doi-asserted-by":"publisher","DOI":"10.1038\/nrclinonc.2017.141","article-title":"Radiomics: the bridge between medical imaging and personalized medicine","volume":"14","author":"Lambin","year":"2017","journal-title":"Nat Rev Clin Oncol"},{"key":"B4","doi-asserted-by":"publisher","DOI":"10.1148\/radiol.2015151169","article-title":"Radiomics: images are more than pictures, they are data","volume":"278","author":"Gillies","year":"2016"},{"key":"B5","doi-asserted-by":"publisher","first-page":"3408","DOI":"10.3390\/cancers17213408","article-title":"Beyond imaging: integrating radiomics, genomics, and multi-omics for precision breast cancer management","volume":"17","author":"Wu","year":"2025"},{"key":"B6","doi-asserted-by":"publisher","DOI":"10.1016\/j.acra.2024.12.039","article-title":"Role of radiomics-based multiomics panel in the microenvironment and prognosis of hepatocellular carcinoma","volume":"32","author":"Wu","year":"2025","journal-title":"Acad Radiol"},{"key":"B7","doi-asserted-by":"publisher","first-page":"1088","DOI":"10.1186\/s12967-024-05896-z","article-title":"Association of radiomic features with genomic signatures in thyroid cancer: a systematic review","volume":"22","author":"Luciano","year":"2024","journal-title":"J Transl Med"},{"key":"B8","doi-asserted-by":"publisher","first-page":"2668","DOI":"10.3390\/cancers16152668","article-title":"Impact of preprocessing parameters in medical imaging-based radiomic studies: a systematic review","volume":"16","author":"Trojani","year":"2024","journal-title":"Cancers"},{"key":"B9","doi-asserted-by":"publisher","DOI":"10.1007\/s00330-025-11374-x","article-title":"Robustness of radiomics within photon-counting detector CT: impact of acquisition and reconstruction factors","volume":"35","author":"Zhang","year":"2025","journal-title":"Eur Radiol"},{"key":"B10","doi-asserted-by":"publisher","DOI":"10.1002\/jmri.29442","article-title":"Reproducible radiomics features from multi-MRI-scanner test-retest-study: influence on performance and generalizability of models","volume":"61","author":"Wennmann","year":"2025","journal-title":"J Magn Reson Imaging"},{"key":"B11","doi-asserted-by":"publisher","first-page":"1039","DOI":"10.3390\/s26031039","article-title":"Image quality standardization in radiomics: a systematic review of artifacts, variability, and feature stability","volume":"26","author":"Felicetti","year":"2026","journal-title":"Sensors"},{"key":"B12","doi-asserted-by":"publisher","first-page":"24TR02","DOI":"10.1088\/1361-6560\/aba798","article-title":"Harmonization strategies for multicenter radiomics investigations","volume":"65","author":"Da-Ano","year":"2020","journal-title":"Phys Med Biol"},{"key":"B13","doi-asserted-by":"publisher","DOI":"10.1002\/med.21846","article-title":"A review in radiomics: making personalized medicine a reality via routine imaging","volume":"42","author":"Guiot","year":"2022","journal-title":"Med Res Rev"},{"key":"B14","doi-asserted-by":"publisher","DOI":"10.1038\/s41598-020-69298-z","article-title":"Standardization of brain MR images across machines and protocols: bridging the gap for MRI-based radiomics","volume":"10","author":"Carr\u00e9","year":"2020","journal-title":"Sci Rep"},{"key":"B15","doi-asserted-by":"publisher","DOI":"10.1093\/biostatistics\/kxj037","article-title":"Adjusting batch effects in microarray expression data using empirical Bayes methods","volume":"8","author":"Johnson","year":"2007","journal-title":"Biostatistics"},{"key":"B16","doi-asserted-by":"publisher","DOI":"10.1007\/s10278-024-01253-0","article-title":"Investigation of ComBat harmonization on radiomic and deep features from multi-center abdominal MRI data","volume":"38","author":"Jia","year":"2025","journal-title":"J Digit Imaging"},{"key":"B17","doi-asserted-by":"publisher","first-page":"270","DOI":"10.3390\/jimaging10110270","article-title":"Mitigating interobserver variability in radiomics with ComBat: a feasibility study","volume":"10","author":"D'Anna","year":"2024","journal-title":"J Imaging"},{"key":"B18","doi-asserted-by":"publisher","first-page":"1212","DOI":"10.3390\/bioengineering11121212","article-title":"Enhancing radiomics reproducibility: deep learning-based harmonization of abdominal computed tomography (CT) images","volume":"11","author":"Lee","year":"2024","journal-title":"Bioengineering"},{"key":"B19","doi-asserted-by":"publisher","first-page":"111891","DOI":"10.1016\/j.ejrad.2024.111891","article-title":"Addressing inter-reconstruction variability in multi-energy myocardial CT radiomics: the benefits of ComBat harmonization","volume":"183","author":"Priya","year":"2025","journal-title":"Eur J Radiol"},{"key":"B20","doi-asserted-by":"crossref","DOI":"10.1097\/RLI.0000000000000970","article-title":"ComBat harmonization for MRI radiomics: impact on nonbinary tissue classification by machine learning","volume":"58","author":"Leithner","year":"2023","journal-title":"Invest Radiol"},{"key":"B21","doi-asserted-by":"publisher","first-page":"41508","DOI":"10.1038\/s41598-025-25400-x","article-title":"Challenges and best practices when using ComBat to harmonize diffusion MRI data","volume":"15","author":"Jodoin","year":"2025","journal-title":"Sci Rep"},{"key":"B22","doi-asserted-by":"publisher","first-page":"19009","DOI":"10.1038\/s41598-022-23328-0","article-title":"Improved generalized ComBat methods for harmonization of radiomic features","volume":"12","author":"Horng","year":"2022","journal-title":"Sci Rep"},{"key":"B23","doi-asserted-by":"publisher","DOI":"10.1101\/2023.04.24.537396","article-title":"DeepComBat: a statistically motivated, hyperparameter-robust, deep learning approach to harmonization of neuroimaging data","volume":"45","author":"Hu","year":"2024","journal-title":"Hum Brain Mapp"},{"key":"B24","doi-asserted-by":"publisher","DOI":"10.1371\/journal.pone.0322365","article-title":"Harmonizing CT scanner acquisition variability in an anthropomorphic phantom: a comparative study of image-level and feature-level harmonization using GAN, ComBat, and their combination","volume":"20","author":"Mali","year":"2025","journal-title":"PLoS One"},{"key":"B25","doi-asserted-by":"publisher","DOI":"10.1148\/radiol.2020191145","article-title":"The image biomarker standardization initiative: standardized quantitative radiomics for high-throughput image-based phenotyping","volume":"295","author":"Zwanenburg","year":"2020","journal-title":"Radiology"},{"key":"B26","doi-asserted-by":"publisher","DOI":"10.18502\/fbt.v8i4.7754","article-title":"The impact of preprocessing on the PET-CT radiomics features in non-small cell lung cancer","volume":"8","author":"Hosseini","year":"2021","journal-title":"Front Biomed Technol"},{"key":"B27","doi-asserted-by":"publisher","DOI":"10.2967\/jnmt.124.268252","article-title":"Robustness of (18)F-FDG PET radiomic features in lung cancer: impact of advanced reconstruction algorithm","volume":"53","author":"Dwivedi","year":"2025","journal-title":"J Nucl Med Technol"},{"key":"B28","doi-asserted-by":"publisher","DOI":"10.1007\/s13246-025-01625-y","article-title":"PET-based radiomic analysis in multicentre lung cancer study and impact of feature domain harmonization","volume":"48","author":"Dwivedi","year":"2025","journal-title":"Phys Eng Sci Med"},{"key":"B29","doi-asserted-by":"publisher","first-page":"842","DOI":"10.3390\/jpm11090842","article-title":"Making radiomics more reproducible across scanner and imaging protocol variations: a review of harmonization methods","volume":"11","author":"Mali","year":"2021","journal-title":"J Pers Med"},{"key":"B30","doi-asserted-by":"publisher","first-page":"10248","DOI":"10.1038\/s41598-020-66110-w","article-title":"Performance comparison of modified ComBat for harmonization of radiomic features for multicenter studies","volume":"10","author":"Da-Ano","year":"2020","journal-title":"Sci Rep"},{"key":"B31","doi-asserted-by":"publisher","DOI":"10.1186\/s12859-015-0478-3","article-title":"Removing batch effects from purified plasma cell gene expression microarrays with modified ComBat","volume":"16","author":"Stein","year":"2015","journal-title":"BMC Bioinformatics"},{"key":"B32","doi-asserted-by":"publisher","DOI":"10.1093\/nargab\/lqaa078","article-title":"ComBat-seq: batch effect adjustment for RNA-seq count data","volume":"2","author":"Zhang","year":"2020","journal-title":"NAR Genom Bioinform"},{"key":"B33","doi-asserted-by":"publisher","DOI":"10.3390\/cancers14194871","article-title":"Oncologic imaging and radiomics: a walkthrough review of methodological challenges","volume":"14","author":"Stanzione","year":"2022","journal-title":"Cancers"},{"key":"B34","doi-asserted-by":"publisher","first-page":"137","DOI":"10.1186\/s12880-024-01306-4","article-title":"The impact of the ComBat method on radiomics feature compensation and analysis of scanners from different manufacturers","volume":"24","author":"Zhang","year":"2024","journal-title":"BMC Med Imaging"}],"container-title":["Frontiers in Oncology"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fonc.2026.1896994\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,14]],"date-time":"2026-09-14T12:09:17Z","timestamp":1789387757000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fonc.2026.1896994\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,9,14]]},"references-count":34,"alternative-id":["10.3389\/fonc.2026.1896994"],"URL":"https:\/\/doi.org\/10.3389\/fonc.2026.1896994","relation":{},"ISSN":["2234-943X"],"issn-type":[{"value":"2234-943X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,9,14]]},"article-number":"1896994"}}