{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,18]],"date-time":"2026-08-18T18:24:48Z","timestamp":1787077488511,"version":"build-2736575974"},"reference-count":31,"publisher":"Public Library of Science (PLoS)","issue":"8","license":[{"start":{"date-parts":[[2026,8,18]],"date-time":"2026-08-18T00:00:00Z","timestamp":1787011200000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["www.plosone.org"],"crossmark-restriction":false},"short-container-title":["PLoS One"],"abstract":"<jats:sec id=\"sec001\">\n                    <jats:title>Objectives<\/jats:title>\n                    <jats:p>This research aims to comprehensively assess the impact of both conservatively and extensively delineated Regions of Interest (ROI) on subsequent quantitative in-vivo characterisation of paediatric brain tumours using diffusion-weighted MRI.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec id=\"sec002\">\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>Utilising a retrospective cohort of 106 paediatric brain tumour patients, ground truth (GT) ROIs delineating tumour boundaries were eroded or dilated, simulating conservative and extensive ROI drawing strategies respectively. Stability was evaluated for 19 first-order radiomic features extracted from Apparent Diffusion Coefficient (ADC) maps within each ROI. Further, these features were used to train a series of machine learning models to evaluate the impact of ROI boundaries on downstream diagnostic classification.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec id=\"sec003\">\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>\n                      For 18\/19 first-order features, ROI dilation introduced significantly (\n                      <jats:italic>p<\/jats:italic>\n                      \u2009&lt;\u20090.01) greater feature variability than erosion, with large effect size (d\u2009&gt;\u20090.8) for 11 features. This relationship was variable between diagnoses, and strongest amongst pilocytic astrocytomas. Diagnostic models trained using features from GT ROIs were negatively impacted with classification accuracy reduced by 3.8\u2009\u00b1\u20090.8% and 5.6\u2009\u00b1\u20090.9% for low-level erosion and dilation respectively. Inclusion of eroded\/dilated ROI features into the training dataset combined with stable-feature selection partially mitigated the impact of erosion\/dilation on model accuracy with 1.4\u2009\u00b1\u20090.7% and 2.9\u2009\u00b1\u20090.3% accuracy drop compared model accuracy on features extracted from GT ROIs.\n                    <\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec id=\"sec004\">\n                    <jats:title>Conclusion<\/jats:title>\n                    <jats:p>The consistently reduced impact of conservative boundaries over extensive ones suggests that, in terms of segmentation strategies, exclusion of ambiguous boundary regions may be preferable over their inclusion. Additionally, diagnostic models exhibited improved robustness to variable ROI drawing strategies through training augmentation and selection of stable features.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1371\/journal.pone.0356382","type":"journal-article","created":{"date-parts":[[2026,8,18]],"date-time":"2026-08-18T17:28:21Z","timestamp":1787074101000},"page":"e0356382","update-policy":"https:\/\/doi.org\/10.1371\/journal.pone.corrections_policy","source":"Crossref","is-referenced-by-count":0,"title":["Evaluating the impact of segmentation strategies on radiomic feature stability for paediatric brain tumour diagnosis using diffusion weighted imaging"],"prefix":"10.1371","volume":"21","author":[{"ORCID":"https:\/\/orcid.org\/0009-0000-1699-4415","authenticated-orcid":true,"given":"Timothy","family":"Mulvany","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Daniel","family":"Griffiths-King","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Lara","family":"Worthington","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0006-4405-7561","authenticated-orcid":true,"given":"Katherine","family":"Crombie","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-0346-1334","authenticated-orcid":true,"given":"Heather E. L.","family":"Rose","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Andrew","family":"Peet","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"John","family":"Apps","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-5173-3608","authenticated-orcid":true,"given":"Jan","family":"Novak","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"340","published-online":{"date-parts":[[2026,8,18]]},"reference":[{"issue":"11","key":"pone.0356382.ref001","article-title":"The correlation between apparent diffusion coefficient and tumor cellularity in patients: a meta-analysis","volume":"8","author":"L Chen","year":"2013","journal-title":"PLoS One"},{"issue":"4","key":"pone.0356382.ref002","doi-asserted-by":"crossref","first-page":"488","DOI":"10.2967\/jnumed.118.222893","article-title":"Introduction to Radiomics","volume":"61","author":"ME Mayerhoefer","year":"2020","journal-title":"J Nucl Med"},{"issue":"12","key":"pone.0356382.ref003","doi-asserted-by":"crossref","first-page":"749","DOI":"10.1038\/nrclinonc.2017.141","article-title":"Radiomics: the bridge between medical imaging and personalized medicine","volume":"14","author":"P Lambin","year":"2017","journal-title":"Nat Rev Clin Oncol"},{"issue":"12","key":"pone.0356382.ref004","doi-asserted-by":"crossref","first-page":"2036","DOI":"10.1002\/pbc.24578","article-title":"Apparent diffusion coefficient of pediatric cerebellar tumors: A biomarker of tumor grade?","volume":"60","author":"A Poretti","year":"2013","journal-title":"Pediatr Blood Cancer"},{"issue":"9","key":"pone.0356382.ref005","doi-asserted-by":"crossref","first-page":"785","DOI":"10.4329\/wjr.v8.i9.785","article-title":"Diffusion weighted imaging: Technique and applications","volume":"8","author":"V Baliyan","year":"2016","journal-title":"World J Radiol"},{"key":"pone.0356382.ref006","doi-asserted-by":"crossref","first-page":"102172","DOI":"10.1016\/j.nicl.2020.102172","article-title":"Distinguishing between paediatric brain tumour types using multi-parametric magnetic resonance imaging and machine learning: A multi-site study","volume":"25","author":"JT Grist","year":"2020","journal-title":"Neuroimage Clin"},{"issue":"1","key":"pone.0356382.ref007","doi-asserted-by":"crossref","first-page":"2987","DOI":"10.1038\/s41598-021-82214-3","article-title":"Classification of paediatric brain tumours by diffusion weighted imaging and machine learning","volume":"11","author":"J Novak","year":"2021","journal-title":"Sci Rep"},{"key":"pone.0356382.ref008","doi-asserted-by":"crossref","first-page":"147","DOI":"10.1002\/jmri.28007","article-title":"Meta-analysis of apparent diffusion coefficient in pediatric medulloblastoma, ependymoma, and pilocytic astrocytoma","volume":"56","author":"RJ Dury","year":"2022","journal-title":"J Magn Resonance Imag"},{"issue":"3","key":"pone.0356382.ref009","doi-asserted-by":"crossref","first-page":"448","DOI":"10.1002\/1522-2594(200103)45:3<448::AID-MRM1059>3.0.CO;2-W","article-title":"Noise correction for the exact determination of apparent diffusion coefficients at low SNR","volume":"45","author":"O Dietrich","year":"2001","journal-title":"Magn Reson Med"},{"issue":"1","key":"pone.0356382.ref010","doi-asserted-by":"crossref","first-page":"12688","DOI":"10.1038\/s41598-020-69534-6","article-title":"Radiomics feature reproducibility under inter-rater variability in segmentations of CT images","volume":"10","author":"C Haarburger","year":"2020","journal-title":"Sci Rep"},{"issue":"1","key":"pone.0356382.ref011","doi-asserted-by":"crossref","first-page":"14163","DOI":"10.1038\/s41598-020-70940-z","article-title":"MRI-based radiomics in breast cancer: feature robustness with respect to inter-observer segmentation variability","volume":"10","author":"RWY Granzier","year":"2020","journal-title":"Sci Rep"},{"issue":"1","key":"pone.0356382.ref012","doi-asserted-by":"crossref","first-page":"64","DOI":"10.1007\/s00247-013-2752-8","article-title":"Pediatric neuroimaging using magnetic resonance imaging during non-sedated sleep","volume":"44","author":"3rd Dean DC","year":"2014","journal-title":"Pediatr Radiol"},{"key":"pone.0356382.ref013","doi-asserted-by":"crossref","first-page":"100886","DOI":"10.1016\/j.neo.2023.100886","article-title":"Current state of pediatric neuro-oncology imaging, challenges and future directions","volume":"37","author":"A Nabavizadeh","year":"2023","journal-title":"Neoplasia"},{"issue":"5","key":"pone.0356382.ref014","doi-asserted-by":"crossref","first-page":"518","DOI":"10.1007\/s11547-022-01468-7","article-title":"Diffusion-weighted MRI radiomics of spine bone tumors: feature stability and machine learning-based classification performance","volume":"127","author":"S Gitto","year":"2022","journal-title":"Radiol Med"},{"key":"pone.0356382.ref015","first-page":"612","article-title":"Stability assessment of first order statistics features computed on ADC maps in soft-tissue sarcoma","volume":"2017","author":"M Bologna","year":"2017","journal-title":"Annu Int Conf IEEE Eng Med Biol Soc"},{"issue":"1","key":"pone.0356382.ref016","doi-asserted-by":"crossref","first-page":"175","DOI":"10.1007\/s11060-014-1375-8","article-title":"Diffusion-weighted MRI derived apparent diffusion coefficient identifies prognostically distinct subgroups of pediatric diffuse intrinsic pontine glioma","volume":"117","author":"RM Lober","year":"2014","journal-title":"J Neurooncol"},{"issue":"1","key":"pone.0356382.ref017","doi-asserted-by":"crossref","first-page":"139","DOI":"10.1186\/s12885-019-5344-7","article-title":"Incidence of childhood CNS tumours in Britain and variation in rates by definition of malignant behaviour: population-based study","volume":"19","author":"CA Stiller","year":"2019","journal-title":"BMC Cancer"},{"issue":"4","key":"pone.0356382.ref018","doi-asserted-by":"crossref","first-page":"468","DOI":"10.1002\/nbm.3269","article-title":"Multi-centre reproducibility of diffusion MRI parameters for clinical sequences in the brain","volume":"28","author":"M Grech-Sollars","year":"2015","journal-title":"NMR Biomed"},{"key":"pone.0356382.ref019","doi-asserted-by":"crossref","unstructured":"Pieper S, Halle M, Kikinis R. 3D slicer (V5.2.1) software. 2004.","DOI":"10.1109\/ISBI.2004.1398617"},{"key":"pone.0356382.ref020","first-page":"120","article-title":"The OpenCV library","volume":"25","author":"G Bradski","year":"2000","journal-title":"DR Dobbs J"},{"issue":"21","key":"pone.0356382.ref021","doi-asserted-by":"crossref","DOI":"10.1158\/0008-5472.CAN-17-0339","article-title":"Computational Radiomics System to Decode the Radiographic Phenotype","volume":"77","author":"JJM van Griethuysen","year":"2017","journal-title":"Cancer Res"},{"issue":"3","key":"pone.0356382.ref022","doi-asserted-by":"crossref","first-page":"279","DOI":"10.4300\/JGME-D-12-00156.1","article-title":"Using Effect Size-or Why the P Value Is Not Enough","volume":"4","author":"GM Sullivan","year":"2012","journal-title":"J Grad Med Educ"},{"issue":"2","key":"pone.0356382.ref023","doi-asserted-by":"crossref","first-page":"155","DOI":"10.1016\/j.jcm.2016.02.012","article-title":"A Guideline of Selecting and Reporting Intraclass Correlation Coefficients for Reliability Research","volume":"15","author":"TK Koo","year":"2016","journal-title":"J Chiropr Med"},{"issue":"1","key":"pone.0356382.ref024","article-title":"Diagnostic accuracy of apparent diffusion coefficient ratio in distinguishing common pediatric CNS posterior fossa tumors","volume":"51","author":"MMM Esa","year":"2020","journal-title":"Egypt J Radiol Nucl Med"},{"issue":"8","key":"pone.0356382.ref025","doi-asserted-by":"crossref","first-page":"989","DOI":"10.1007\/s11547-023-01657-y","article-title":"MRI radiomics-based machine learning for classification of deep-seated lipoma and atypical lipomatous tumor of the extremities","volume":"128","author":"S Gitto","year":"2023","journal-title":"Radiol Med"},{"issue":"7","key":"pone.0356382.ref026","doi-asserted-by":"crossref","first-page":"1172","DOI":"10.3390\/jpm13071172","article-title":"Stability of Multi-Parametric Prostate MRI Radiomic Features to Variations in Segmentation","volume":"13","author":"S Thulasi Seetha","year":"2023","journal-title":"J Pers Med"},{"issue":"1","key":"pone.0356382.ref027","doi-asserted-by":"crossref","first-page":"6","DOI":"10.1186\/s41747-019-0087-4","article-title":"Quantitative DWI predicts event-free survival in children with neuroblastic tumours: preliminary findings from a retrospective cohort study","volume":"3","author":"A-L Peschmann","year":"2019","journal-title":"Eur Radiol Exp"},{"key":"pone.0356382.ref028","doi-asserted-by":"crossref","first-page":"49","DOI":"10.1016\/j.ejro.2018.12.004","article-title":"Can apparent diffusion coefficient values help distinguish between different types of pediatric brain tumors?","volume":"6","author":"AM Al-Sharydah","year":"2019","journal-title":"Eur J Radiol Open"},{"issue":"2","key":"pone.0356382.ref029","doi-asserted-by":"crossref","first-page":"238","DOI":"10.3390\/tomography7020022","article-title":"Stability of Radiomic Features across Different Region of Interest Sizes-A CT and MR Phantom Study","volume":"7","author":"LJ Jensen","year":"2021","journal-title":"Tomography"},{"issue":"6","key":"pone.0356382.ref030","doi-asserted-by":"crossref","first-page":"1487","DOI":"10.2214\/ajr.171.6.9843275","article-title":"Brain abscess and necrotic or cystic brain tumor: discrimination with signal intensity on diffusion-weighted MR imaging","volume":"171","author":"YJ Kim","year":"1998","journal-title":"AJR Am J Roentgenol"},{"issue":"6","key":"pone.0356382.ref031","doi-asserted-by":"crossref","first-page":"130","DOI":"10.3390\/jcm7060130","article-title":"The Role of Diffusion-Weighted Magnetic Resonance Imaging in the Differentiation of Head and Neck Masses","volume":"7","author":"L Kanmaz","year":"2018","journal-title":"J Clin Med"}],"container-title":["PLOS One"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0356382","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,8,18]],"date-time":"2026-08-18T17:28:27Z","timestamp":1787074107000},"score":1,"resource":{"primary":{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0356382"}},"subtitle":[],"editor":[{"given":"Lorenzo","family":"Faggioni","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"editor"}]}],"short-title":[],"issued":{"date-parts":[[2026,8,18]]},"references-count":31,"journal-issue":{"issue":"8","published-online":{"date-parts":[[2026,8,18]]}},"URL":"https:\/\/doi.org\/10.1371\/journal.pone.0356382","relation":{},"ISSN":["1932-6203"],"issn-type":[{"value":"1932-6203","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,8,18]]}}}