{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,14]],"date-time":"2026-08-14T14:55:12Z","timestamp":1786719312437,"version":"build-2736575974"},"reference-count":50,"publisher":"Wiley","issue":"4","license":[{"start":{"date-parts":[[2026,4,3]],"date-time":"2026-04-03T00:00:00Z","timestamp":1775174400000},"content-version":"vor","delay-in-days":2,"URL":"http:\/\/onlinelibrary.wiley.com\/termsAndConditions#vor"},{"start":{"date-parts":[[2026,4,1]],"date-time":"2026-04-01T00:00:00Z","timestamp":1775001600000},"content-version":"tdm","delay-in-days":0,"URL":"http:\/\/doi.wiley.com\/10.1002\/tdm_license_1.1"}],"funder":[{"DOI":"10.13039\/501100001407","name":"Department of Biotechnology, Ministry of Science and Technology, India","doi-asserted-by":"publisher","id":[{"id":"10.13039\/501100001407","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["aapm.onlinelibrary.wiley.com"],"crossmark-restriction":true},"short-container-title":["Medical Physics"],"published-print":{"date-parts":[[2026,4]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:sec>\n                    <jats:title>Background<\/jats:title>\n                    <jats:p>Median nerve, a major peripheral nerve, connects the hand to the central nervous system, facilitating upper limb motor function and sensation by transmitting sensory data from the palm and fingers. Damage to this nerve can result in motor and sensory deficits, with carpal tunnel syndrome (CTS) causing compression, leading to tingling and numbness in the thumb, index, middle, and lateral ring\u00a0fingers.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Purpose<\/jats:title>\n                    <jats:p>This study aimed to develop an accurate deep\u2010learning\u2010based segmentation method for measuring the cross\u2010sectional area (CSA) of the median nerve to facilitate the diagnosis of nerve entrapment syndromes and aid in surgical planning, with a focus on\u00a0CTS.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>\n                      This study introduces MNSeg\u2010Net, a novel lightweight multiscale feature fusion network with 2.46M parameters for median nerve segmentation in ultrasound (US) frames, specifically designed to enable a fully automated, end\u2010to\u2010end clinical setup supporting real\u2010time segmentation and CSA computation. The dataset comprised 100 subjects and 30\u00a0000 ultrasound frames, which were split into training (80%), validation (10%), and testing (10%) subsets with subject\u2010wise separation to avoid data leakage. MNSeg\u2010Net was benchmarked against state\u2010of\u2010the\u2010art segmentation models, including UNet and its variants (UNet++ and U2Net). The performance was assessed using metrics such as the Dice similarity coefficient (DSC) and CSA difference. The statistical significance of performance differences was evaluated using paired\n                      <jats:italic>t<\/jats:italic>\n                      \u2010tests, effect size (Cohen's\n                      <jats:italic>d<\/jats:italic>\n                      ), and one\u2010way ANOVA with Tukey's HSD correction for multiple comparisons at a \u2010value threshold of 0.05, while statistical equivalence between models within predefined margins was formally assessed using the two one\u2010sided test (TOST) procedure. Following quantitative validation, the model was deployed in a real\u2010time clinical setup utilizing an Av.io HD Epiphan frame grabber to stream ultrasound images from the ultrasound machine to a GPU\u2010equipped system. A secondary display running parallel to the original ultrasound screen visualized the segmented median nerve and computed the CSA values in real\u00a0time.\n                    <\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>MNSeg\u2010Net achieved high segmentation performance, with average DSC scores of 94.7% at the wrist and 83.4% from the wrist to the elbow, and the lowest Hausdorff distance, matching the performance of the best\u2010performing 44\u2010million\u2010parameter heavy U2Net model. Compared to U2Net, MNSeg\u2010Net showed no statistically significant difference in DSC performance (; Cohen's ; mean difference = \u20130.001), with formal equivalence testing confirming equivalence across all tested margins (). For CSA estimation, MNSeg\u2010Net also showed no statistically significant difference from clinician\u2010annotated values (; Cohen's ; mean difference = \u20130.081), and equivalence was established at the  margin, confirming a strong alignment with expert clinical assessments. MNSeg\u2010Net demonstrated real\u2010time performance by processing up to 43 frames per second on a single GPU, successfully segmenting the median nerve and computing CSA from ultrasound\u00a0frames.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusion<\/jats:title>\n                    <jats:p>The developed MNSeg\u2010Net\u2010based clinical system represents an important step toward real\u2010time median nerve assessment, enabling a fully automated solution for CTS diagnosis. By combining a lightweight architecture, real\u2010time processing capability, and successful clinical deployment, it represents a substantial advancement in the CTS detection and\u00a0management.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1002\/mp.70414","type":"journal-article","created":{"date-parts":[[2026,4,4]],"date-time":"2026-04-04T03:31:56Z","timestamp":1775273516000},"update-policy":"https:\/\/doi.org\/10.1002\/crossmark_policy","source":"Crossref","is-referenced-by-count":1,"title":["Performance benchmarking of deep learning models for real\u2010time median nerve segmentation and cross\u2010sectional area measurement in ultrasound imaging"],"prefix":"10.1002","volume":"53","author":[{"given":"Vaddadi","family":"Venkatesh","sequence":"first","affiliation":[{"name":"Department of Computational and Data Sciences Indian Institute of Science Bangalore India"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Lokesh","family":"Bathala","sequence":"additional","affiliation":[{"name":"Aster CMI Hospital Bangalore India"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Raji Susan","family":"Mathew","sequence":"additional","affiliation":[{"name":"School of Data Science Indian Institute of Science Education and Research Thiruvananthapuram India"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Phaneendra K.","family":"Yalavarthy","sequence":"additional","affiliation":[{"name":"Department of Computational and Data Sciences Indian Institute of Science Bangalore India"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"311","published-online":{"date-parts":[[2026,4,3]]},"reference":[{"key":"e_1_2_10_2_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.hansur.2019.10.197"},{"key":"e_1_2_10_3_1","doi-asserted-by":"publisher","DOI":"10.4103\/0970-0358.129632"},{"key":"e_1_2_10_4_1","doi-asserted-by":"publisher","DOI":"10.1371\/journal.pone.0136477"},{"key":"e_1_2_10_5_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.cnp.2017.09.001"},{"key":"e_1_2_10_6_1","doi-asserted-by":"publisher","DOI":"10.1109\/TNSRE.2018.2833141"},{"key":"e_1_2_10_7_1","doi-asserted-by":"crossref","unstructured":"HopkinsPM.Ultrasound guidance as a gold standard in regional anaesthesia.2007.","DOI":"10.1093\/bja\/ael387"},{"key":"e_1_2_10_8_1","doi-asserted-by":"publisher","DOI":"10.1093\/bja\/aei002"},{"key":"e_1_2_10_9_1","doi-asserted-by":"crossref","unstructured":"HadjerciO HafianeA ConteD MakrisP VieyresP DelbosA.Ultrasound median nerve localization by classification based on despeckle filtering and feature selection. In:2015 IEEE International Conference on Image Processing (ICIP).2015:4155\u20104159.","DOI":"10.1109\/ICIP.2015.7351588"},{"key":"e_1_2_10_10_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.imu.2016.06.003"},{"key":"e_1_2_10_11_1","unstructured":"HafianeA VieyresP DelbosA.Deep learning with spatiotemporal consistency for nerve segmentation in ultrasound images.arXiv preprint arXiv:1706.05870.2017."},{"key":"e_1_2_10_12_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.compmedimag.2019.101687"},{"key":"e_1_2_10_13_1","doi-asserted-by":"publisher","DOI":"10.21037\/qims-21-1074"},{"key":"e_1_2_10_14_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.ultrasmedbio.2021.03.018"},{"key":"e_1_2_10_15_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.ultrasmedbio.2020.03.017"},{"key":"e_1_2_10_16_1","doi-asserted-by":"publisher","DOI":"10.3390\/diagnostics11101893"},{"key":"e_1_2_10_17_1","doi-asserted-by":"publisher","DOI":"10.1007\/s11517-022-02662-5"},{"key":"e_1_2_10_18_1","doi-asserted-by":"publisher","DOI":"10.1109\/TIM.2024.3451569"},{"key":"e_1_2_10_19_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.ultrasmedbio.2022.12.014"},{"key":"e_1_2_10_20_1","doi-asserted-by":"publisher","DOI":"10.1109\/TNNLS.2023.3238381"},{"key":"e_1_2_10_21_1","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1109\/TIM.2024.3421436","article-title":"EU2\u2010Net: A parameter efficient ensemble model with attention\u2010aided triple feature fusion for tumor segmentation in breast ultrasound images","volume":"73","author":"Roy A","year":"2024","journal-title":"IEEE Trans Instrum Meas"},{"key":"e_1_2_10_22_1","doi-asserted-by":"publisher","DOI":"10.1002\/mp.15492"},{"key":"e_1_2_10_23_1","article-title":"Automatic calculation method for stenosis ratio based on dialysis access ultrasound image segmentation","author":"Shi F","journal-title":"Med Phys"},{"key":"e_1_2_10_24_1","doi-asserted-by":"publisher","DOI":"10.1109\/TUFFC.2023.3330539"},{"key":"e_1_2_10_25_1","doi-asserted-by":"crossref","unstructured":"RonnebergerO FischerP BroxT.U\u2010net: convolutional networks for biomedical image segmentation. In:International Conference on Medical Image Computing and Computer\u2010Assisted Intervention.Springer;2015:234\u2010241.","DOI":"10.1007\/978-3-319-24574-4_28"},{"key":"e_1_2_10_26_1","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2016.2644615"},{"key":"e_1_2_10_27_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.isprsjprs.2020.01.013"},{"key":"e_1_2_10_28_1","unstructured":"OktayO SchlemperJ FolgocLL et\u00a0al.Attention U\u2010Net: learning where to look for the pancreas.arXiv preprint arXiv:1804.03999.2018."},{"key":"e_1_2_10_29_1","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-00889-5_1"},{"key":"e_1_2_10_30_1","doi-asserted-by":"crossref","unstructured":"QinX ZhangZ HuangC et\u00a0al.BASNet: boundary\u2010aware salient object detection. In:Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition.2019:7479\u20107489.","DOI":"10.1109\/CVPR.2019.00766"},{"key":"e_1_2_10_31_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.patcog.2020.107404"},{"key":"e_1_2_10_32_1","first-page":"155","article-title":"ImageJ user guide","volume":"1","author":"Ferreira T","year":"2012","journal-title":"ImageJ\/Fiji"},{"key":"e_1_2_10_33_1","first-page":"164","article-title":"Determination of cut\u2010off point of cross\u2010sectional area of median nerve at the wrist for diagnosing carpal tunnel syndrome","volume":"16","author":"Ghasemi M","year":"2017","journal-title":"Iran J Neurol"},{"key":"e_1_2_10_34_1","doi-asserted-by":"publisher","DOI":"10.1177\/1558944718788642"},{"key":"e_1_2_10_35_1","doi-asserted-by":"publisher","DOI":"10.3390\/diagnostics12112621"},{"key":"e_1_2_10_36_1","doi-asserted-by":"publisher","DOI":"10.1016\/S1076-6332(03)00671-8"},{"key":"e_1_2_10_37_1","doi-asserted-by":"publisher","DOI":"10.1109\/34.232073"},{"key":"e_1_2_10_38_1","unstructured":"KingmaDP BaJ.Adam: A method for stochastic optimization.arXiv preprint arXiv:1412.6980.2014."},{"key":"e_1_2_10_39_1","unstructured":"SudreCH LiW VercauterenT et\u00a0al.Generalised dice overlap as a deep learning loss function for highly unbalanced segmentations in deep learning in medical image analysis and multimodal learning for clinical decision support. In:Third International Workshop DLMIA 2017 and 7th International Workshop ML\u2010CDS 2017 Held in Conjunction with MICCAI 2017 Qu\u00e9bec City QC Canada September 14 Proceedings. Vol.3 Springer;2017:240\u2010248."},{"key":"e_1_2_10_40_1","doi-asserted-by":"publisher","DOI":"10.1007\/s10479-005-5724-z"},{"key":"e_1_2_10_41_1","doi-asserted-by":"crossref","unstructured":"JadonS.A survey of loss functions for semantic segmentation. In:2020 IEEE Conference on Computational Intelligence in Bioinformatics and Computational Biology (CIBCB).IEEE;2020:1\u20107.","DOI":"10.1109\/CIBCB48159.2020.9277638"},{"key":"e_1_2_10_42_1","doi-asserted-by":"publisher","DOI":"10.1109\/TIP.2003.819861"},{"key":"e_1_2_10_43_1","unstructured":"AV.Io HD. Now available: the new and improved AV.Io HD+. Epiphan video website. Accessed March 19 2026."},{"key":"e_1_2_10_44_1","doi-asserted-by":"crossref","unstructured":"Van WoudenbergN LiaoZ AbdiAH et\u00a0al.Quantitative echocardiography: real\u2010time quality estimation and view classification implemented on a mobile android device. In:Simulation Image Processing and Ultrasound Systems for Assisted Diagnosis and Navigation: International Workshops POCUS 2018 BIVPCS 2018 Cu\u2010RIOUS 2018 and CPM 2018 Held in Conjunction with MICCAI 2018 Granada Spain September 16\u201020 2018 Proceedings.Springer;2018:74\u201081.","DOI":"10.1007\/978-3-030-01045-4_9"},{"key":"e_1_2_10_45_1","doi-asserted-by":"publisher","DOI":"10.1186\/s13075-022-02729-6"},{"key":"e_1_2_10_46_1","doi-asserted-by":"publisher","DOI":"10.2307\/2331554"},{"key":"e_1_2_10_47_1","volume-title":"Statistical Power Analysis for the Behavioral Sciences","author":"Cohen J","year":"1988"},{"key":"e_1_2_10_48_1","volume-title":"Statistical Methods for Research Workers","author":"Fisher RA","year":"1925"},{"key":"e_1_2_10_49_1","doi-asserted-by":"publisher","DOI":"10.2307\/3001913"},{"key":"e_1_2_10_50_1","doi-asserted-by":"publisher","DOI":"10.1007\/BF01068419"},{"key":"e_1_2_10_51_1","doi-asserted-by":"publisher","DOI":"10.3389\/fninf.2022.911679"}],"container-title":["Medical Physics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/aapm.onlinelibrary.wiley.com\/doi\/pdf\/10.1002\/mp.70414","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/aapm.onlinelibrary.wiley.com\/doi\/full-xml\/10.1002\/mp.70414","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/aapm.onlinelibrary.wiley.com\/doi\/pdf\/10.1002\/mp.70414","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,7,3]],"date-time":"2026-07-03T21:27:19Z","timestamp":1783114039000},"score":1,"resource":{"primary":{"URL":"https:\/\/aapm.onlinelibrary.wiley.com\/doi\/10.1002\/mp.70414"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,4]]},"references-count":50,"journal-issue":{"issue":"4","published-print":{"date-parts":[[2026,4]]}},"alternative-id":["10.1002\/mp.70414"],"URL":"https:\/\/doi.org\/10.1002\/mp.70414","archive":["Portico"],"relation":{},"ISSN":["0094-2405","2473-4209"],"issn-type":[{"value":"0094-2405","type":"print"},{"value":"2473-4209","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,4]]},"assertion":[{"value":"2025-01-24","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2026-03-14","order":2,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2026-04-03","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}],"article-number":"e70414"}}