{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"institution":[{"name":"medRxiv"}],"indexed":{"date-parts":[[2026,9,25]],"date-time":"2026-09-25T06:00:42Z","timestamp":1790316042373,"version":"4.1.0"},"posted":{"date-parts":[[2025,11,19]]},"group-title":"Radiology and Imaging","reference-count":28,"publisher":"openRxiv","license":[{"start":{"date-parts":[[2025,11,19]],"date-time":"2025-11-19T00:00:00Z","timestamp":1763510400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/www.medrxiv.org\/about\/FAQ#license"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"accepted":{"date-parts":[[2025,11,19]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                <jats:sec>\n                  <jats:title>Background<\/jats:title>\n                  <jats:p>Quantification of ischemic brain tissue on non-contrast CT (NCCT) in acute ischemic stroke is challenging in the acute setting.<\/jats:p>\n                <\/jats:sec>\n                <jats:sec>\n                  <jats:title>Purpose<\/jats:title>\n                  <jats:p>To compare the spatial overlap and imaging marker agreement of acute ischemic regions of interest (ROIs) using deep-learning NCCT (DLNCCT) versus manual NCCT, CTP, and DWI-based ischemic segmentations.<\/jats:p>\n                <\/jats:sec>\n                <jats:sec>\n                  <jats:title>Methods<\/jats:title>\n                  <jats:p>We trained a deep learning model to segment ischemic ROIs using manual lesion annotations on admission NCCTs (DLNCCT). DLNCCT ischemic ROIs were compared with manual NCCT delineation, CTP (rCBF&lt;30%\/38%), and DWI within 5 hours after the NCCT or after recanalization in four external test sets. Spatial overlap was measured using the Dice Similarity Coefficient (DSC; mean\u00b1SD). For each ROI, we derived: average density (HU); modified net water uptake (mNWU in %); total volume (mL); and hypodense (&lt;26HU) volume (mL), and assessed agreement via Bland\u2013Altman (mean difference [95%CI]) and concordance correlation coefficient (CCC) analysis.<\/jats:p>\n                <\/jats:sec>\n                <jats:sec>\n                  <jats:title>Results<\/jats:title>\n                  <jats:p>218 training (n=104\/89\/25 male\/female\/unknown, mean age 68\u00b114 years) and 762 test cases (n=243\/206\/313 male\/female\/unknown, mean age 70\u00b115 years) were used. Spatial overlap was 0.30\u00b10.30 between DLNCCT and manual segmentation, 0.22\u00b10.25 between DLNCCT and DWI, 0.10\u00b10.19\/0.14\u00b10.21 between DLNCCT and CTP (rCBF&lt;30%\/&lt;38%), and 0.15\u00b10.22\/0.21\u00b10.24 between CTP (rCBF&lt;30%\/&lt;38%) and DWI. DLNCCT vs. DWI mean differences of ischemic ROI derived imaging markers were -1HU (95%CI:-7;6) for average density (CCC:0.71), 4.9% (95%CI:-7.0;16.8) for mNWU (CCC:0.35), -16mL (95%CI:-108;76) for total volume (CCC:0.57), and -4mL (95%CI:-31;23) for hypodense lesion volume (CCC: 0.75).<\/jats:p>\n                <\/jats:sec>\n                <jats:sec>\n                  <jats:title>Conclusion<\/jats:title>\n                  <jats:p>Spatial overlap and agreement of imaging markers between DLNCCT and DWI ischemic ROIs were comparable to CTP and DWI.<\/jats:p>\n                <\/jats:sec>\n                <jats:sec>\n                  <jats:title>Summary Statement<\/jats:title>\n                  <jats:p>Ischemic injury on NCCT is identified and quantified by a deep-learning model with accuracy similar to CTP and DWI in stroke patients with a large vessel occlusion.<\/jats:p>\n                <\/jats:sec>\n                <jats:sec>\n                  <jats:title>Key results<\/jats:title>\n                  <jats:list list-type=\"simple\">\n                    <jats:list-item>\n                      <jats:label>-<\/jats:label>\n                      <jats:p>Deep-learning models can segment ischemic brain tissue on NCCT.<\/jats:p>\n                    <\/jats:list-item>\n                    <jats:list-item>\n                      <jats:label>-<\/jats:label>\n                      <jats:p>Ischemic regions identified by our model demonstrate comparable overlap with ischemic core segmentation on CTP (Dice: 0.21\u00b10.24) and DWI (Dice: 0.22\u00b10.25).<\/jats:p>\n                    <\/jats:list-item>\n                    <jats:list-item>\n                      <jats:label>-<\/jats:label>\n                      <jats:p>Deep learning NCCT showed high agreement with follow-up DWI in determining the hypodense (&lt;26 HU) lesion volume (mean difference -4mL [95%CI:-31;23], CCC: 0.75).<\/jats:p>\n                    <\/jats:list-item>\n                  <\/jats:list>\n                <\/jats:sec>","DOI":"10.1101\/2025.11.18.25340504","type":"posted-content","created":{"date-parts":[[2025,11,20]],"date-time":"2025-11-20T01:45:36Z","timestamp":1763603136000},"source":"Crossref","is-referenced-by-count":1,"title":["Deep learning based ischemic lesion markers on non-contrast head CT compared to CTP and DWI"],"prefix":"10.64898","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-2647-3557","authenticated-orcid":false,"given":"Henk","family":"van Voorst","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Bin","family":"Jiang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Praneeta","family":"Konduri","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Adrien ter","family":"Schiphorst","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Aroosa","family":"Zamarud","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Seena","family":"Dehkharghani","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Lieselotte","family":"vandeWalle","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ewout","family":"Heylen","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yongkai","family":"Liu","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Michael","family":"Mlynash","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Soren","family":"Christensen","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Nicole","family":"Yuen","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Benjamin FJ","family":"Verhaaren","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Abdelkader","family":"Mahammedi","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Patrik","family":"Michel","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Max","family":"Wintermark","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Gregory W","family":"Albers","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Greg","family":"Zaharchuk","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Maarten G","family":"Lansberg","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jeremy J","family":"Heit","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"54368","reference":[{"key":"2025112104510726000_2025.11.18.25340504v1.1","doi-asserted-by":"publisher","DOI":"10.1007\/s00330-022-08658-x"},{"issue":"8","key":"2025112104510726000_2025.11.18.25340504v1.2","doi-asserted-by":"crossref","first-page":"5308","DOI":"10.1007\/s00330-024-10599-6","article-title":"Net water uptake as a predictive neuroimaging marker for acute ischemic stroke outcomes: a meta-analysis","volume":"34","year":"2024","journal-title":"Eur Radiol"},{"key":"2025112104510726000_2025.11.18.25340504v1.3","doi-asserted-by":"publisher","DOI":"10.1038\/s41598-020-59121-0"},{"issue":"6","key":"2025112104510726000_2025.11.18.25340504v1.4","doi-asserted-by":"crossref","first-page":"1366","DOI":"10.1161\/STROKEAHA.124.048899","article-title":"Association of Ischemic Core Hypodensity With Thrombectomy Treatment Effect in Large Core Stroke: A Secondary Analysis of the SELECT2 Randomized Controlled Trial","volume":"56","year":"2025","journal-title":"Stroke"},{"key":"2025112104510726000_2025.11.18.25340504v1.5","doi-asserted-by":"crossref","unstructured":"Broocks G , Meyer L , Hanning U , Faizy TD , Bechstein M , Kniep H , et al. Haemorrhage after thrombectomy with adjuvant thrombolysis in unknown onset stroke depends on high early lesion water uptake. Stroke Vasc Neurol. 2023;","DOI":"10.1136\/svn-2022-002264"},{"issue":"2","key":"2025112104510726000_2025.11.18.25340504v1.6","doi-asserted-by":"crossref","first-page":"e250708","DOI":"10.1148\/radiol.250708","article-title":"Net Water Uptake at CT Predicts the Treatment Effect of Thrombectomy for Low ASPECTS Stroke","volume":"317","year":"2025","journal-title":"Radiology"},{"key":"2025112104510726000_2025.11.18.25340504v1.7","doi-asserted-by":"crossref","unstructured":"Kimberly WT , Saver JL , Campbell BC , Albers GW , Molyneaux BJ , Hinson H , et al. Intravenous Glyburide in Medical and Endovascular-Treated Large-Core Stroke: A Subgroup Analysis of the CHARM Randomized Clinical Trial. Ann Neurol. 2025;","DOI":"10.1002\/ana.27268"},{"key":"2025112104510726000_2025.11.18.25340504v1.8","doi-asserted-by":"publisher","DOI":"10.1212\/WNL.0000000000002860"},{"key":"2025112104510726000_2025.11.18.25340504v1.9","doi-asserted-by":"publisher","DOI":"10.1002\/ana.24818"},{"key":"2025112104510726000_2025.11.18.25340504v1.10","doi-asserted-by":"publisher","DOI":"10.1177\/0271678X15610586"},{"key":"2025112104510726000_2025.11.18.25340504v1.11","doi-asserted-by":"publisher","DOI":"10.1148\/radiol.2021203651"},{"key":"2025112104510726000_2025.11.18.25340504v1.12","doi-asserted-by":"publisher","DOI":"10.3389\/fneur.2020.00001"},{"key":"2025112104510726000_2025.11.18.25340504v1.13","doi-asserted-by":"publisher","DOI":"10.3389\/fneur.2021.669406"},{"issue":"10","key":"2025112104510726000_2025.11.18.25340504v1.14","doi-asserted-by":"crossref","first-page":"7026","DOI":"10.1007\/s00330-022-09077-8","article-title":"ASPECTS-based net water uptake predicts poor reperfusion and poor clinical outcomes in patients with ischemic stroke","volume":"32","year":"2022","journal-title":"Eur Radiol"},{"issue":"1","key":"2025112104510726000_2025.11.18.25340504v1.15","doi-asserted-by":"crossref","first-page":"53","DOI":"10.1136\/jnis-2023-021283","article-title":"Random expert sampling for deep learning segmentation of acute ischemic stroke on non-contrast CT","volume":"17","year":"2025","journal-title":"J NeuroInterventional Surg"},{"issue":"November 2021","key":"2025112104510726000_2025.11.18.25340504v1.16","first-page":"105033","article-title":"Evaluating nnU-Net for early ischemic change segmentation on non-contrast computed tomography in patients with Acute Ischemic Stroke","volume":"141","year":"2022","journal-title":". Comput Biol Med"},{"issue":"7","key":"2025112104510726000_2025.11.18.25340504v1.17","doi-asserted-by":"crossref","first-page":"1815","DOI":"10.1161\/STROKEAHA.123.042689","article-title":"Automatic Ischemic Core Estimation Based on Noncontrast-Enhanced Computed Tomography","volume":"54","year":"2023","journal-title":"Stroke"},{"issue":"1","key":"2025112104510726000_2025.11.18.25340504v1.18","doi-asserted-by":"crossref","first-page":"189","DOI":"10.1038\/s41598-023-27496-5","article-title":"Head CT deep learning model is highly accurate for early infarct estimation","volume":"13","year":"2023","journal-title":"Sci Rep"},{"key":"2025112104510726000_2025.11.18.25340504v1.19","doi-asserted-by":"crossref","unstructured":"Albers GW , Marks MP , Kemp S , Christensen S , Tsai JP , Ortega-Gutierrez S , et al. Thrombectomy for Stroke at 6 to 16 Hours with Selection by Perfusion Imaging. N Engl J Med. 2018;","DOI":"10.1056\/NEJMoa1713973"},{"key":"2025112104510726000_2025.11.18.25340504v1.20","doi-asserted-by":"crossref","unstructured":"Liang K , Han K , Li X , Cheng X , Li Y , Wang Y , et al. Symmetry-Enhanced Attention Network for Acute Ischemic Infarct Segmentation with Non-Contrast CT Images. In Strasbourg, France; 2021.","DOI":"10.1007\/978-3-030-87234-2_41"},{"key":"2025112104510726000_2025.11.18.25340504v1.21","unstructured":"Riedel EO , de la Rosa E , Baran TA , Petzsche MH , Baazaoui H , Yang K , et al. ISLES 2024: The first longitudinal multimodal multi-center real-world dataset in (sub-) acute stroke. ArXiv Prepr ArXiv240811142. 2024;"},{"issue":"2","key":"2025112104510726000_2025.11.18.25340504v1.22","doi-asserted-by":"crossref","first-page":"569","DOI":"10.1161\/STROKEAHA.121.034444","article-title":"Prediction of Stroke Infarct Growth Rates by Baseline Perfusion Imaging","volume":"53","year":"2022","journal-title":"Stroke"},{"key":"2025112104510726000_2025.11.18.25340504v1.23","doi-asserted-by":"publisher","DOI":"10.1038\/s41592-020-01008-z"},{"issue":"4","key":"2025112104510726000_2025.11.18.25340504v1.24","doi-asserted-by":"crossref","first-page":"e200127","DOI":"10.1148\/ryai.2021200127","article-title":"Optimizing Deep Learning Algorithms for Segmentation of Acute Infarcts on Non\u2013Contrast Material\u2013enhanced CT Scans of the Brain Using Simulated Lesions","volume":"3","year":"2021","journal-title":"Radiol Artif Intell"},{"key":"2025112104510726000_2025.11.18.25340504v1.25","unstructured":"Heylen E , Demeestere J , Wouters A , Seners P , Christensen S , Yuen N , et al. Uncertainty Estimation and Calibration in nnU-Net-Based Ischemic Stroke Lesion Segmentation. In: Medical Imaging with Deep Learning-Short Papers. 2025."},{"key":"2025112104510726000_2025.11.18.25340504v1.26","doi-asserted-by":"publisher","DOI":"10.1177\/096228029900800204"},{"issue":"11","key":"2025112104510726000_2025.11.18.25340504v1.27","doi-asserted-by":"crossref","first-page":"3028","DOI":"10.1177\/0271678X211023660","article-title":"Deep learning-based identification of acute ischemic core and deficit from non-contrast CT and CTA","volume":"41","year":"2021","journal-title":"J Cereb Blood Flow Metab"},{"key":"2025112104510726000_2025.11.18.25340504v1.28","doi-asserted-by":"crossref","unstructured":"Vandewalle L , Konduri PR , Christensen S , Seners P , Wouters A , Yuen N , et al. Hypodensity Beyond the Ischemic Core: Penumbral Changes Detected With Relative Noncontrast Computed Tomography. Stroke. 2025;","DOI":"10.1161\/str.56.suppl_1.DP32"}],"container-title":[],"original-title":[],"link":[{"URL":"https:\/\/syndication.highwire.org\/content\/doi\/10.1101\/2025.11.18.25340504","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,1,15]],"date-time":"2026-01-15T20:51:36Z","timestamp":1768510296000},"score":1,"resource":{"primary":{"URL":"http:\/\/medrxiv.org\/lookup\/doi\/10.1101\/2025.11.18.25340504"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,11,19]]},"references-count":28,"URL":"https:\/\/doi.org\/10.1101\/2025.11.18.25340504","relation":{},"subject":[],"published":{"date-parts":[[2025,11,19]]},"subtype":"preprint"}}