{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,22]],"date-time":"2026-10-22T08:40:39Z","timestamp":1792658439896,"version":"4.5.0"},"reference-count":31,"publisher":"Oxford University Press (OUP)","issue":"1187","license":[{"start":{"date-parts":[[2026,8,7]],"date-time":"2026-08-07T00:00:00Z","timestamp":1786060800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/academic.oup.com\/pages\/standard-publication-reuse-rights"}],"funder":[{"name":"Natural Science Foundation of Chongqing, China","award":["CSTB2022NSCQ-MSX0116"],"award-info":[{"award-number":["CSTB2022NSCQ-MSX0116"]}]},{"name":"Chongqing medical scientific research project","award":["2025MSXM009"],"award-info":[{"award-number":["2025MSXM009"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2026,11,1]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:sec>\n                    <jats:title>Objectives<\/jats:title>\n                    <jats:p>To develop a multi-task learning (MTL) framework for the simultaneous segmentation of spontaneous intracerebral hemorrhage (sICH) and perihematomal edema (PHE), along with the prediction of hematoma expansion (HE).<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>This multicenter retrospective study collected baseline noncontrast CT (NCCT) images of 981 patients with sICH from 4 centers, with 429 cases for training, 183 cases for internal validation, and 369 cases for external testing. A multi-task sICH model (MTsICHM) was designed to simultaneously segment hematoma and PHE and predict HE, which employs a hard parameter-sharing architecture based on a 3D U-Net backbone. The model processes dual-window NCCT inputs and generates 3 outputs: a binary hematoma mask, a binary PHE mask, and a binary HE label. For comparative evaluation and ablation studies, 4 control models were constructed: a multi-task segmentation comparative model, a multi-task classification comparative model, a single-task segmentation comparative model, and a single-task classification comparative model.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>For hematoma segmentation, MTsICHM achieved mean Dice of 0.909 and Intersection over Union of 0.872; for PHE segmentation, it attained 0.686 and 0.644, respectively, consistently outperforming the comparative models. For HE prediction, it yielded an AUC of 0.888, with corresponding accuracy, sensitivity, and specificity of 0.765, 0.949, and 0.710.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusions<\/jats:title>\n                    <jats:p>The MTsICHM accurately segments hematoma and PHE while effectively predicting HE, demonstrating substantial potential for clinical application.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Advances in knowledge<\/jats:title>\n                    <jats:p>MTsICHM simultaneously quantifies hematoma and PHE and predicts HE from a single NCCT, with high consistency with manual segmentation and clinical utility for risk identification.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1093\/bjr\/tqag179","type":"journal-article","created":{"date-parts":[[2026,8,7]],"date-time":"2026-08-07T12:47:28Z","timestamp":1786106848000},"page":"2375-2386","source":"Crossref","is-referenced-by-count":0,"title":["A non-contrast CT-based multi-task learning framework for co-segmenting hematoma and edema and predicting hematoma expansion in spontaneous intracerebral hemorrhage"],"prefix":"10.1093","volume":"99","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-3980-6795","authenticated-orcid":false,"given":"Lu","family":"Zhou","sequence":"first","affiliation":[{"name":"Department of Radiology, The Third Affiliated Hospital of Chongqing Medical University , Chongqing 400016,","place":["China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jinhua","family":"Zhou","sequence":"additional","affiliation":[{"name":"Department of Radiology, The Second Affiliated Hospital of Chongqing Medical University , Chongqing 400010,","place":["China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Hongli","family":"Zhou","sequence":"additional","affiliation":[{"name":"Department of Radiology, The Nanchong Central Hospital , Sichuan 637000,","place":["China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Hong","family":"Xu","sequence":"additional","affiliation":[{"name":"The Second Clinical College of Chongqing Medical University , Chongqing 400010,","place":["China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-7028-7866","authenticated-orcid":false,"given":"Zhiming","family":"Zhou","sequence":"additional","affiliation":[{"name":"Department of Radiology, The Second Affiliated Hospital of Chongqing Medical University , Chongqing 400010,","place":["China"]},{"name":"Key Laboratory of Intelligent Processing and Applications for Medical Imaging Big Data , Chongqing 400010,","place":["China"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"286","published-online":{"date-parts":[[2026,8,7]]},"reference":[{"key":"2026102204075344500_tqag179-B1","doi-asserted-by":"publisher","first-page":"1295","DOI":"10.1001\/jama.2019.2413","article-title":"Cerebral intraparenchymal hemorrhage: a review","volume":"321","author":"Gross","year":"2019","journal-title":"JAMA"},{"key":"2026102204075344500_tqag179-B2","doi-asserted-by":"publisher","first-page":"1408288","DOI":"10.3389\/fnins.2024.1408288","article-title":"Neuro-imaging in intracerebral hemorrhage: updates and knowledge gaps","volume":"18","author":"Penckofer","year":"2024","journal-title":"Front Neurosci"},{"key":"2026102204075344500_tqag179-B3","doi-asserted-by":"publisher","first-page":"2776","DOI":"10.1161\/STROKEAHA.116.013779","article-title":"Development and validation of an automatic segmentation algorithm for quantification of intracerebral hemorrhage","volume":"47","author":"Scherer","year":"2016","journal-title":"Stroke"},{"key":"2026102204075344500_tqag179-B4","doi-asserted-by":"publisher","first-page":"379","DOI":"10.1016\/j.nicl.2017.02.007","article-title":"PItcHPERFeCT: primary intracranial hemorrhage probability estimation using random forests on CT","volume":"14","author":"Muschelli","year":"2017","journal-title":"Neuroimage Clin"},{"key":"2026102204075344500_tqag179-B5","doi-asserted-by":"publisher","first-page":"19389","DOI":"10.1038\/s41598-020-76459-7","article-title":"A fast and fully-automated deep-learning approach for accurate hemorrhage segmentation and volume quantification in non-contrast whole-head CT","volume":"10","author":"Arab","year":"2020","journal-title":"Sci Rep"},{"key":"2026102204075344500_tqag179-B6","doi-asserted-by":"publisher","first-page":"648","DOI":"10.1161\/STROKEAHA.119.027657","article-title":"Deep learning for automated measurement of hemorrhage and perihematomal edema in supratentorial intracerebral hemorrhage","volume":"51","author":"Dhar","year":"2020","journal-title":"Stroke"},{"key":"2026102204075344500_tqag179-B7","doi-asserted-by":"publisher","first-page":"20546","DOI":"10.1038\/s41598-020-77441-z","article-title":"Detection and classification of intracranial haemorrhage on CT images using a novel deep-learning algorithm","volume":"10","author":"Lee","year":"2020","journal-title":"Sci Rep"},{"key":"2026102204075344500_tqag179-B8","doi-asserted-by":"publisher","first-page":"1646","DOI":"10.1109\/JBHI.2020.3028243","article-title":"Deep learning for hemorrhagic lesion detection and segmentation on brain CT images","volume":"25","author":"Li","year":"2020","journal-title":"IEEE J Biomed Health Informat"},{"key":"2026102204075344500_tqag179-B9","doi-asserted-by":"publisher","first-page":"1120","DOI":"10.1161\/STROKEAHA.119.027451","article-title":"Redefining hematoma expansion with the inclusion of intraventricular hemorrhage growth","volume":"51","author":"Yogendrakumar","year":"2020","journal-title":"Stroke"},{"key":"2026102204075344500_tqag179-B10","doi-asserted-by":"publisher","first-page":"1305","DOI":"10.1161\/STROKEAHA.119.028574","article-title":"Evaluating hematoma expansion scores in acute spontaneous intracerebral hemorrhage: a systematic scoping review","volume":"51","author":"Yogendrakumar","year":"2020","journal-title":"Stroke"},{"key":"2026102204075344500_tqag179-B11","doi-asserted-by":"publisher","first-page":"973104","DOI":"10.3389\/fonc.2022.973104","article-title":"Efficacy of non-enhanced computer tomography-based radiomics for predicting hematoma expansion: a meta-analysis","volume":"12","author":"Jiang","year":"2022","journal-title":"Front Oncol"},{"key":"2026102204075344500_tqag179-B12","doi-asserted-by":"publisher","first-page":"779","DOI":"10.3390\/jpm12050779","article-title":"Automatic and efficient prediction of hematoma expansion in patients with hypertensive intracerebral hemorrhage using deep learning based on CT images","volume":"12","author":"Ma","year":"2022","journal-title":"J Pers Med"},{"key":"2026102204075344500_tqag179-B13","doi-asserted-by":"publisher","first-page":"2908","DOI":"10.1007\/s00330-023-10410-y","article-title":"Prediction of early hematoma expansion of spontaneous intracerebral hemorrhage based on deep learning radiomics features of noncontrast computed tomography","volume":"34","author":"Feng","year":"2024","journal-title":"Eur Radiol"},{"key":"2026102204075344500_tqag179-B14","doi-asserted-by":"publisher","DOI":"10.1016\/j.jstrokecerebrovasdis.2024.107979","article-title":"Hybrid clinical-radiomics model based on fully automatic segmentation for predicting the early expansion of spontaneous intracerebral hemorrhage: a multi-center study","volume":"33","author":"Wang","year":"2024","journal-title":"J Stroke Cerebrovasc Dis"},{"key":"2026102204075344500_tqag179-B15","doi-asserted-by":"publisher","first-page":"285","DOI":"10.1186\/s12880-024-01455-6","article-title":"Spatial attention-based CSR-Unet framework for subdural and epidural hemorrhage segmentation and classification using CT images","volume":"24","author":"Nafees Ahmed","year":"2024","journal-title":"BMC Med Imaging"},{"key":"2026102204075344500_tqag179-B16","doi-asserted-by":"publisher","first-page":"173","DOI":"10.1038\/s41551-018-0324-9","article-title":"An explainable deep-learning algorithm for the detection of acute intracranial haemorrhage from small datasets","volume":"3","author":"Lee","year":"2019","journal-title":"Nat Biomed Eng"},{"key":"2026102204075344500_tqag179-B17","doi-asserted-by":"publisher","DOI":"10.3390\/app15010111","article-title":"Optimizing automated hematoma expansion classification from baseline and follow-up head computed tomography","volume":"15","author":"Tran","year":"2025","journal-title":"Appl Sci (Basel)"},{"key":"2026102204075344500_tqag179-B18","doi-asserted-by":"publisher","first-page":"102685","DOI":"10.1016\/j.media.2022.102685","article-title":"One model is all you need: Multi-task learning enables simultaneous histology image segmentation and classification","volume":"83","author":"Graham","year":"2023","journal-title":"Med Image Anal"},{"key":"2026102204075344500_tqag179-B19","doi-asserted-by":"publisher","first-page":"1538544","DOI":"10.3389\/fgene.2025.1538544","article-title":"Auto-branch multi-task learning for simultaneous prediction of multiple correlated traits associated with Alzheimer\u2019s disease","volume":"16","author":"Liang","year":"2025","journal-title":"Front Genet"},{"key":"2026102204075344500_tqag179-B20","doi-asserted-by":"publisher","first-page":"1851","DOI":"10.1038\/s41467-021-22188-y","article-title":"Predicting treatment response from longitudinal images using multi-task deep learning","volume":"12","author":"Jin","year":"2021","journal-title":"Nat Commun"},{"key":"2026102204075344500_tqag179-B21","doi-asserted-by":"publisher","first-page":"279","DOI":"10.1093\/neuonc\/noac166","article-title":"Combined molecular subtyping, grading, and segmentation of glioma using multi-task deep learning","volume":"25","author":"van der Voort","year":"2023","journal-title":"Neuro Oncol"},{"key":"2026102204075344500_tqag179-B22","doi-asserted-by":"publisher","first-page":"258","DOI":"10.1016\/j.gie.2021.08.022","article-title":"Deep learning model for diagnosing gastric mucosal lesions using endoscopic images: development, validation, and method comparison","volume":"95","author":"Nam","year":"2022","journal-title":"Gastrointest Endosc"},{"key":"2026102204075344500_tqag179-B23","doi-asserted-by":"publisher","first-page":"3614","DOI":"10.1109\/TPAMI.2021.3054719","article-title":"Multi-task learning for dense prediction tasks: a survey","volume":"44","author":"Vandenhende","year":"2022","journal-title":"IEEE Trans Pattern Anal Mach Intell"},{"key":"2026102204075344500_tqag179-B24","doi-asserted-by":"publisher","first-page":"e220096","DOI":"10.1148\/ryai.220096","article-title":"Semantic segmentation of spontaneous intracerebral hemorrhage, intraventricular hemorrhage, and associated edema on CT images using deep learning","volume":"4","author":"Kok","year":"2022","journal-title":"Radiol Artif Intell"},{"key":"2026102204075344500_tqag179-B25","doi-asserted-by":"publisher","first-page":"umaf012","DOI":"10.1093\/radadv\/umaf012","article-title":"Cross-institutional automated multilabel segmentation for acute intracerebral hemorrhage, intraventricular hemorrhage, and perihematomal edema on CT","volume":"2","author":"Nawabi","year":"2025","journal-title":"Radiol Adv"},{"key":"2026102204075344500_tqag179-B26","doi-asserted-by":"publisher","first-page":"299","DOI":"10.4274\/dir.2024.222088","article-title":"Application of deep learning and radiomics in the prediction of hematoma expansion in intracerebral hemorrhage: a fully automated hybrid approach","volume":"30","author":"Lu","year":"2024","journal-title":"Diagn Interv Radiol"},{"key":"2026102204075344500_tqag179-B27","doi-asserted-by":"publisher","first-page":"577","DOI":"10.1007\/s00234-024-03298-y","article-title":"Predicting hematoma expansion in acute spontaneous intracerebral hemorrhage: integrating clinical factors with a multitask deep learning model for non-contrast head CT","volume":"66","author":"Lee","year":"2024","journal-title":"Neuroradiology"},{"key":"2026102204075344500_tqag179-B28","doi-asserted-by":"publisher","first-page":"1323","DOI":"10.1111\/j.1468-1331.2011.03395.x","article-title":"Semi-automatic volumetric assessment of perihemorrhagic edema with computed tomography","volume":"18","author":"Volbers","year":"2011","journal-title":"Eur J Neurol"},{"key":"2026102204075344500_tqag179-B29","doi-asserted-by":"publisher","first-page":"126","DOI":"10.1016\/j.compbiomed.2019.01.022","article-title":"Automated segmentation of haematoma and perihaematomal oedema in MRI of acute spontaneous intracerebral haemorrhage","volume":"106","author":"Pszczolkowski","year":"2019","journal-title":"Comput Biol Med"},{"key":"2026102204075344500_tqag179-B30","doi-asserted-by":"publisher","first-page":"488","DOI":"10.1136\/jnnp-2012-303160","article-title":"Volume-dependent effect of perihaematomal oedema on outcome for spontaneous intracerebral haemorrhages","volume":"84","author":"Appelboom","year":"2013","journal-title":"J Neurol Neurosurg Psychiatry"},{"key":"2026102204075344500_tqag179-B31","doi-asserted-by":"publisher","first-page":"815","DOI":"10.1161\/STROKEAHA.119.026764","article-title":"Fully automated segmentation algorithm for perihematomal edema volumetry after spontaneous intracerebral hemorrhage","volume":"51","author":"Ironside","year":"2020","journal-title":"Stroke"}],"container-title":["British Journal of Radiology"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/academic.oup.com\/bjr\/advance-article-pdf\/doi\/10.1093\/bjr\/tqag179\/70529698\/tqag179.pdf","content-type":"application\/pdf","content-version":"am","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/bjr\/article-pdf\/99\/1187\/2375\/70529698\/tqag179.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/bjr\/article-pdf\/99\/1187\/2375\/70529698\/tqag179.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,10,22]],"date-time":"2026-10-22T08:07:59Z","timestamp":1792656479000},"score":1,"resource":{"primary":{"URL":"https:\/\/academic.oup.com\/bjr\/article\/99\/1187\/2375\/8753983"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,8,7]]},"references-count":31,"journal-issue":{"issue":"1187","published-online":{"date-parts":[[2026,8,7]]},"published-print":{"date-parts":[[2026,11,1]]}},"URL":"https:\/\/doi.org\/10.1093\/bjr\/tqag179","relation":{},"ISSN":["0007-1285","1748-880X"],"issn-type":[{"value":"0007-1285","type":"print"},{"value":"1748-880X","type":"electronic"}],"subject":[],"published-other":{"date-parts":[[2026,11]]},"published":{"date-parts":[[2026,8,7]]}}}