{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,7]],"date-time":"2026-08-07T17:20:55Z","timestamp":1786123255407,"version":"build-2736575974"},"reference-count":25,"publisher":"Ovid Technologies (Wolters Kluwer Health)","content-domain":{"domain":["lww.com","ovid.com"],"crossmark-restriction":true},"short-container-title":[],"abstract":"<jats:sec>\n                    <jats:title>BACKGROUND:<\/jats:title>\n                    <jats:p>The stellate ganglion region is densely vascularized and innervated, making the stellate ganglion block (SGB) technically challenging under ultrasound, particularly for beginners. Deep learning can segment complex ultrasound anatomy, but its application to SGB has not been systematically assessed. We developed and validated a multilevel feature fusion UNet (MLF-UNet) to automatically delineate the SGB region on ultrasound, aiming to support accurate needle placement and improve procedural safety.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>METHODS:<\/jats:title>\n                    <jats:p>In this retrospective study, 370 patients who underwent ultrasound-guided SGB between March 1, 2023 and January 16, 2025 were included. Three expert anesthesiologists jointly annotated 730 videos (2190 images) to produce ground truth. Data were split 9:1 by patient into development and heldout test sets. MLF-UNet was trained and compared with 5 benchmark models using identical pipelines. Test-set performance was evaluated with Dice similarity coefficient (DSC), Intersection over Union (IoU), 95th percentile Hausdorff distance (95HD), and average symmetric surface distance (ASSD). Three blinded experts rated model outputs (0\u20132 scale) for topological integrity, boundary precision, and background accuracy. For clinical validation and human\u2013machine comparison, 3 additional experts and 3 nonexperts independently delineated SGB regions on the test set; spatial agreement was visualized with heat maps and assessed by Bland\u2013Altman analysis. Metrics (DSC, IoU, 95HD, and ASSD) were compared among MLF-UNet, experts, and nonexperts.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>RESULTS:<\/jats:title>\n                    <jats:p>\n                      MLF-UNet achieved the best test performance: DSC 0.856 (95% confidence interval [CI], 0.846\u20130.865), IoU 0.754 (95% CI, 0.740\u20130.768), 95HD 3.98 mm (95% CI, 3.44\u20134.52 mm), and ASSD 1.08 mm (95% CI, 0.99\u20131.18 mm). Expert ratings favored MLF-UNet over all benchmark models for topological integrity (all\n                      <jats:italic toggle=\"yes\">P &lt;<\/jats:italic>\n                      .001), boundary precision (all\n                      <jats:italic toggle=\"yes\">P &lt;<\/jats:italic>\n                      .001), background accuracy (\n                      <jats:italic toggle=\"yes\">P &lt;<\/jats:italic>\n                      .01 or\n                      <jats:italic toggle=\"yes\">P &lt;<\/jats:italic>\n                      .001), and total score (all\n                      <jats:italic toggle=\"yes\">P &lt;<\/jats:italic>\n                      .001). Bland\u2013Altman analysis showed a mean segmentation area difference between MLF-UNet and ground truth of \u221238.1 mm\u00b2 (limits of agreement \u2212278 to +202 mm\u00b2). MLF-UNet outperformed the nonexpert group on region overlap (DSC, IoU; both\n                      <jats:italic toggle=\"yes\">P &lt;<\/jats:italic>\n                      .001) and boundary precision (95HD, ASSD; both\n                      <jats:italic toggle=\"yes\">P &lt;<\/jats:italic>\n                      .001). Compared with experts, MLF-UNet showed no significant difference in overlap (DSC\n                      <jats:italic toggle=\"yes\">P =<\/jats:italic>\n                      .332; IoU\n                      <jats:italic toggle=\"yes\">P =<\/jats:italic>\n                      .125) but had slightly larger boundary precision (95HD and ASSD: both\n                      <jats:italic toggle=\"yes\">P &lt;<\/jats:italic>\n                      .001).\n                    <\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>CONCLUSIONS:<\/jats:title>\n                    <jats:p>MLFUNet outperforms 5 benchmark models and nonexpert clinicians for automated ultrasound segmentation of the SGB region, achieving expert\u2011level region overlap with a modest deficit in boundary precision.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1213\/ane.0000000000007938","type":"journal-article","created":{"date-parts":[[2026,2,16]],"date-time":"2026-02-16T22:00:14Z","timestamp":1771279214000},"update-policy":"https:\/\/doi.org\/10.1097\/lww.0000000000001000","source":"Crossref","is-referenced-by-count":0,"title":["Automated Segmentation of Stellate Ganglion Block Region in Ultrasound Images Using Deep Learning Model"],"prefix":"10.1213","author":[{"given":"Weixiong","family":"Chen","sequence":"first","affiliation":[{"name":"School of Health Science and Engineering, University of Shanghai for Science and Technology, Shanghai, China"},{"name":"Department of Anesthesiology, Eye & ENT Hospital of Fudan University, Shanghai, China."}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Lili","family":"Feng","sequence":"additional","affiliation":[{"name":"Department of Anesthesiology, Eye & ENT Hospital of Fudan University, Shanghai, China."}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Wenna","family":"Zhang","sequence":"additional","affiliation":[{"name":"Department of Anesthesiology, Eye & ENT Hospital of Fudan University, Shanghai, China."}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yilei","family":"Shen","sequence":"additional","affiliation":[{"name":"Department of Anesthesiology, Eye & ENT Hospital of Fudan University, Shanghai, China."}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Hongjun","family":"Liu","sequence":"additional","affiliation":[{"name":"Department of Anesthesiology, Eye & ENT Hospital of Fudan University, Shanghai, China."}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Wenxian","family":"Li","sequence":"additional","affiliation":[{"name":"School of Health Science and Engineering, University of Shanghai for Science and Technology, Shanghai, China"},{"name":"Department of Anesthesiology, Eye & ENT Hospital of Fudan University, Shanghai, China."}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yuan","family":"Han","sequence":"additional","affiliation":[{"name":"Department of Anesthesiology, Eye & ENT Hospital of Fudan University, Shanghai, China."}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yu","family":"Tian","sequence":"additional","affiliation":[{"name":"Department of Anesthesiology, Eye & ENT Hospital of Fudan University, Shanghai, China."}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Shuangshuang","family":"Li","sequence":"additional","affiliation":[{"name":"Department of Anesthesiology, Eye & ENT Hospital of Fudan University, Shanghai, China."}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"276","published-online":{"date-parts":[[2026,2,16]]},"reference":[{"key":"R1-20260216","doi-asserted-by":"crossref","first-page":"1133","DOI":"10.7205\/MILMED-D-14-00151","article-title":"Stellate ganglion block used to treat symptoms associated with combat-related post-traumatic stress disorder: a case series of 166 patients.","volume":"179","author":"Mulvaney","year":"2014","journal-title":"Mil Med"},{"key":"R2-20260216","doi-asserted-by":"crossref","first-page":"808","DOI":"10.1111\/papr.13350","article-title":"Novel ultrasound-guided supraclavicular stellate ganglion block.","volume":"24","author":"Fajardo P\u00e9rez","year":"2024","journal-title":"Pain Pract"},{"key":"R3-20260216","doi-asserted-by":"crossref","first-page":"E597","DOI":"10.36076\/ppj.2024.7.E597","article-title":"A comprehensive overview of the stellate ganglion block throughout the past three decades: a bibliometric analysis.","volume":"27","author":"Ren","year":"2024","journal-title":"Pain Physician"},{"key":"R4-20260216","doi-asserted-by":"crossref","first-page":"377","DOI":"10.1097\/AAP.0000000000000591","article-title":"Review of sympathetic blocks: anatomy, sonoanatomy, evidence, and techniques.","volume":"42","author":"Baig","year":"2017","journal-title":"Reg Anesth Pain Med"},{"key":"R5-20260216","doi-asserted-by":"crossref","first-page":"669","DOI":"10.1136\/rapm-2018-100127","article-title":"Complications associated with stellate ganglion nerve block: a systematic review.","volume":"44","author":"Goel","year":"2019","journal-title":"Reg Anesth Pain Med"},{"key":"R7-20260216","doi-asserted-by":"crossref","first-page":"453","DOI":"10.21037\/atm.2019.08.61","article-title":"Applying deep learning in recognizing the femoral nerve block region on ultrasound images.","volume":"7","author":"Huang","year":"2019","journal-title":"Ann Transl Med"},{"key":"R8-20260216","doi-asserted-by":"crossref","first-page":"82003","DOI":"10.1109\/ACCESS.2022.3196356","article-title":"Brachial plexus nerve trunk recognition from ultrasound images: a comparative study of deep learning models.","volume":"10","author":"Tian","year":"2022","journal-title":"IEEE Access"},{"key":"R15-20260216","doi-asserted-by":"crossref","first-page":"107114","DOI":"10.1016\/j.cmpb.2022.107114","article-title":"X-ray coronary centerline extraction based on C-UNet and a multifactor reconnection algorithm.","volume":"226","author":"Zhang","year":"2022","journal-title":"Comput Methods Programs Biomed"},{"key":"R16-20260216","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":"Koo","year":"2016","journal-title":"J Chiropract Med"},{"key":"R17-20260216","doi-asserted-by":"crossref","first-page":"293","DOI":"10.1111\/anae.14768","article-title":"Future directions in regional anaesthesia: not just for the cognoscenti.","volume":"75","author":"Turbitt","year":"2020","journal-title":"Anaesthesia"},{"key":"R18-20260216","doi-asserted-by":"crossref","first-page":"1063","DOI":"10.1016\/j.bja.2023.09.023","article-title":"Variability between human experts and artificial intelligence in identification of anatomical structures by ultrasound in regional anaesthesia: a framework for evaluation of assistive artificial intelligence.","volume":"132","author":"Bowness","year":"2024","journal-title":"Br J Anaesth"},{"key":"R19-20260216","doi-asserted-by":"crossref","first-page":"2736","DOI":"10.1007\/s00330-024-11167-8","article-title":"Minimally interactive segmentation of soft-tissue tumors on CT and MRI using deep learning.","volume":"35","author":"Spaanderman","year":"2025","journal-title":"Eur Radiol"},{"key":"R20-20260216","doi-asserted-by":"crossref","first-page":"5087","DOI":"10.1007\/s00330-023-09410-9","article-title":"Automated prostate multi-regional segmentation in magnetic resonance using fully convolutional neural networks.","volume":"33","author":"Jimenez-Pastor","year":"2023","journal-title":"Eur Radiol"},{"key":"R21-20260216","doi-asserted-by":"crossref","first-page":"17","DOI":"10.1186\/s12871-024-02402-2","article-title":"Convolutional neural network for brachial plexus segmentation at the interscalene level.","volume":"24","author":"Xi","year":"2024","journal-title":"BMC Anesthesiol"},{"key":"R22-20260216","doi-asserted-by":"crossref","first-page":"2512","DOI":"10.1016\/j.ultrasmedbio.2022.08.003","article-title":"Application of an improved U2-Net model in ultrasound median neural image segmentation.","volume":"48","author":"Shao","year":"2022","journal-title":"Ultrasound Med Biol"},{"key":"R23-20260216","doi-asserted-by":"crossref","first-page":"e0232586","DOI":"10.1371\/journal.pone.0232586","article-title":"Bi-national survey of Korea and Japan related to the injection site for ultrasound-guided stellate ganglion blocks and anatomic comparisons using cadaver dissection:.","volume":"15","author":"Won","year":"2020","journal-title":"PLoS One"},{"key":"R24-20260216","doi-asserted-by":"crossref","first-page":"998937","DOI":"10.3389\/fnins.2022.998937","article-title":"Visualizing stellate ganglion with US imaging for guided SGB treatment: a feasibility study with healthy adults.","volume":"16","author":"Li","year":"2022","journal-title":"Front Neurosci"},{"key":"R25-20260216","doi-asserted-by":"crossref","first-page":"13","DOI":"10.3390\/jimaging10010013","article-title":"Convolutional neural network approaches in median nerve morphological assessment from ultrasound images.","volume":"10","author":"Ando","year":"2024","journal-title":"J Imag"},{"key":"R26-20260216","doi-asserted-by":"crossref","first-page":"526","DOI":"10.1093\/bja\/aeq206","article-title":"Lateral ultrasound-guided paravertebral blockade: an anatomical-based description of a new technique.","volume":"105","author":"Marhofer","year":"2010","journal-title":"Br J Anaesth"},{"key":"R27-20260216","doi-asserted-by":"crossref","first-page":"522","DOI":"10.1038\/s41551-021-00711-2","article-title":"Prospective assessment of breast cancer risk from multimodal multiview ultrasound images via clinically applicable deep learning.","volume":"5","author":"Qian","year":"2021","journal-title":"Nat Biomed Eng"},{"key":"R28-20260216","doi-asserted-by":"crossref","first-page":"889","DOI":"10.1111\/anae.13921","article-title":"Improving needle tip identification during ultrasound-guided procedures in anaesthetic practice.","volume":"72","author":"Scholten","year":"2017","journal-title":"Anaesthesia"},{"key":"R29-20260216","doi-asserted-by":"crossref","first-page":"6828","DOI":"10.1007\/s00330-023-09700-2","article-title":"Artificial intelligence\u2013based prediction of cervical lymph node metastasis in papillary thyroid cancer with CT.","volume":"33","author":"Wang","year":"2023","journal-title":"Eur Radiol"},{"key":"R30-20260216","doi-asserted-by":"crossref","first-page":"517","DOI":"10.1001\/jama.2017.7797","article-title":"Unintended consequences of machine learning in medicine.","volume":"318","author":"Cabitza","year":"2017","journal-title":"JAMA"},{"key":"R31-20260216","doi-asserted-by":"crossref","first-page":"981","DOI":"10.1056\/NEJMp1714229","article-title":"Implementing machine learning in health care\u2014addressing ethical challenges.","volume":"378","author":"Char","year":"2018","journal-title":"N Engl J Med"},{"key":"R32-20260216","doi-asserted-by":"crossref","first-page":"238","DOI":"10.1136\/bmjqs-2018-008551","article-title":"Framing the challenges of artificial intelligence in medicine.","volume":"28","author":"Yu","year":"2019","journal-title":"BMJ Qual Saf"}],"container-title":["Anesthesia &amp; Analgesia"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.ovid.com\/10.1213\/ANE.0000000000007938","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,8,7]],"date-time":"2026-08-07T14:31:41Z","timestamp":1786113101000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.ovid.com\/10.1213\/ANE.0000000000007938"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,2,16]]},"references-count":25,"URL":"https:\/\/doi.org\/10.1213\/ane.0000000000007938","relation":{},"ISSN":["0003-2999","1526-7598"],"issn-type":[{"value":"0003-2999","type":"print"},{"value":"1526-7598","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,2,16]]}}}