{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,25]],"date-time":"2026-08-25T23:07:13Z","timestamp":1787699233787,"version":"build-2784847793"},"reference-count":25,"publisher":"JMIR Publications Inc.","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["J Med Internet Res"],"abstract":"Abstract<\/jats:title>\n \n Background<\/jats:title>\n Traditional ocular gaze photograph preprocessing, relying on manual cropping and head tilt correction, is time-consuming and inconsistent, limiting artificial intelligence (AI) model development and clinical application.<\/jats:p>\n <\/jats:sec>\n \n Objective<\/jats:title>\n This study aimed to address these challenges using an advanced preprocessing algorithm to enhance the accuracy, efficiency, and standardization of eye region cropping for clinical workflows and AI data preprocessing.<\/jats:p>\n <\/jats:sec>\n \n Methods<\/jats:title>\n This retrospective and prospective cross-sectional study utilized 5832 images from 648 inpatients and outpatients, capturing 3 gaze positions under diverse conditions, including obstructions and varying distances. The preprocessing algorithm, based on a rotating bounding box detection framework, was trained and evaluated using precision, recall, and mean average precision (mAP) at various intersections over union thresholds. A 5-fold cross-validation was performed on an inpatient dataset, with additional testing on an independent outpatient dataset and an external cross-population dataset of 500 images from the IMDB-WIKI collection, representing diverse ethnicities and ages. Expert validation confirmed alignment with clinical standards across 96 images (48 images from a Chinese dataset of patients with strabismus and 48 images from IMDB-WIKI). Gradient-weighted class activation mapping heatmaps were used to assess model interpretability. A control experiment with 5 optometry specialists compared manual and automated cropping efficiency. Downstream task validation involved preprocessing 1000 primary gaze photographs using the Dlib toolkit, faster region-based convolutional neural network (R-CNN; both without head tilt correction), and our model (with correction), evaluating the impact of head tilt correction via the vision transformer strabismus screening network through 5-fold cross-validation.<\/jats:p>\n <\/jats:sec>\n \n Results<\/jats:title>\n \n The model achieved exceptional performance across datasets: on the 5-fold cross-validation set, it recorded a mean precision of 1.000 (95% CI 1.000\u20101.000), recall of 1.000 (95% CI 1.000\u20101.000), mAP50 of 0.995 (95% CI 0.995\u20100.995), and mAP95 of 0.893 (95% CI 0.870\u20100.918); on the internal independent test set, precision and recall were 1.000, with mAP50 of 0.995 and mAP95 of 0.801; and on the external cross-population test set, precision and recall were 1.000, with mAP50 of 0.937 and mAP95 of 0.792. The control experiment reduced image preparation time from 10 hours for manual cropping of 900 photos to 30 seconds with the automated model. Downstream strabismus screening task validation showed our model (with head tilt correction) achieving an area under the curve of 0.917 (95% CI 0.901\u20100.933), surpassing Dlib-toolkit and faster R-CNN (both without head tilt correction) with an area under the curve of 0.856 (\n P<\/jats:italic>\n =.02) and 0.884 (\n P<\/jats:italic>\n =.05), respectively. Heatmaps highlighted core ocular focus, aligning with head tilt directions.\n <\/jats:p>\n <\/jats:sec>\n \n Conclusions<\/jats:title>\n This study delivers an AI-driven platform featuring a preprocessing algorithm that automates eye region cropping, correcting head tilt variations to improve image quality for AI development and clinical use. Integrated with electronic archives and patient-physician interaction, it enhances workflow efficiency, ensures telemedicine privacy, and supports ophthalmological research and strabismus care.<\/jats:p>\n <\/jats:sec>","DOI":"10.2196\/74402","type":"journal-article","created":{"date-parts":[[2025,6,3]],"date-time":"2025-06-03T03:00:10Z","timestamp":1748919610000},"page":"e74402-e74402","source":"Crossref","is-referenced-by-count":4,"title":["Deep Learning\u2013Based Precision Cropping of Eye Regions in Strabismus Photographs: Algorithm Development and Validation Study for Workflow Optimization"],"prefix":"10.2196","volume":"27","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-4938-3044","authenticated-orcid":false,"given":"Dawen","family":"Wu","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0003-4563-7515","authenticated-orcid":false,"given":"Yanfei","family":"Li","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0007-2327-9082","authenticated-orcid":false,"given":"Zeyi","family":"Yang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0003-1254-6705","authenticated-orcid":false,"given":"Teng","family":"Yin","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-0165-3607","authenticated-orcid":false,"given":"Xiaohang","family":"Chen","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0000-7153-397X","authenticated-orcid":false,"given":"Jingyu","family":"Liu","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0001-5148-9358","authenticated-orcid":false,"given":"Wenyi","family":"Shang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0004-1144-9900","authenticated-orcid":false,"given":"Bin","family":"Xie","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-7314-836X","authenticated-orcid":false,"given":"Guoyuan","family":"Yang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-9821-508X","authenticated-orcid":false,"given":"Haixian","family":"Zhang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-2265-4625","authenticated-orcid":false,"given":"Longqian","family":"Liu","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1010","published-online":{"date-parts":[[2025,7,17]]},"reference":[{"issue":"10","key":"R1","doi-asserted-by":"publisher","first-page":"1337","DOI":"10.1136\/bjo.2007.116905","article-title":"Comitant horizontal strabismus: an Asian perspective","volume":"91","author":"Chia","journal-title":"Br J Ophthalmol"},{"issue":"2","key":"R2","doi-asserted-by":"publisher","first-page":"402","DOI":"10.1016\/s0161-6420(99)00035-4","article-title":"Impact of large angle horizontal strabismus on ability to obtain employment","volume":"107","author":"Coats","journal-title":"Ophthalmology"},{"issue":"6","key":"R3","doi-asserted-by":"publisher","first-page":"484","DOI":"10.3760\/cma.j.cn112142-20231111-00226","article-title":"Challenges and prospects in the application of artificial intelligence for ocular disease screening and diagnosis","volume":"60","author":"Liu","journal-title":"Zhonghua Yan Ke Za Zhi"},{"key":"R4","doi-asserted-by":"publisher","DOI":"10.2196\/60226","article-title":"Performance of ChatGPT in ophthalmic registration and clinical diagnosis: cross-sectional study","volume":"26","author":"Ming","journal-title":"J Med Internet Res"},{"issue":"16","key":"R5","doi-asserted-by":"publisher","first-page":"3101","DOI":"10.1038\/s41433-024-03228-5","article-title":"An artificial intelligence platform for the screening and managing of strabismus","volume":"38","author":"Wu","journal-title":"Eye (Lond)"},{"key":"R6","doi-asserted-by":"publisher","DOI":"10.3389\/ebm.2024.10320","article-title":"Integrating artificial intelligence in strabismus management: current research landscape and future directions","volume":"249","author":"Wu","journal-title":"Exp Biol Med (Maywood)"},{"key":"R7","doi-asserted-by":"crossref","unstructured":"Kazemi V Sullivan J . One millisecond face alignment with an ensemble of regression trees. In: Kazemi V Sullivan J , editors. Presented at: 2014 IEEE Conference on Computer Vision and Pattern Recognition (CVPR); Jun 23-28, 2014; Columbus, OH. [doi: 10.1109\/CVPR.2014.241]","DOI":"10.1109\/CVPR.2014.241"},{"key":"R8","author":"King","journal-title":"J Mach Learn Res"},{"key":"R9","doi-asserted-by":"crossref","unstructured":"Girshick R . Fast r-CNN. Presented at: 2015 IEEE International Conference on Computer Vision (ICCV); Dec 7-13, 2015; Santiago, Chile. 2015.[doi: 10.1109\/ICCV.2015.169]","DOI":"10.1109\/ICCV.2015.169"},{"key":"R10","unstructured":"G Jocher AC Qiu J . YOLO by ultralytics. GitHub. URL: https:\/\/github.com\/ultralytics\/ultralytics2023 [Accessed 10-01-2023]"},{"key":"R11","doi-asserted-by":"crossref","unstructured":"Timofte R Rothe R Van Gool L . DEX: deep expectation of apparent age from a single image. Presented at: IEEE International Conference on Computer Vision (ICCV); Dec 11-18, 2015; Santiago, Chile. [doi: 10.1109\/ICCVW.2015.41]","DOI":"10.1109\/ICCVW.2015.41"},{"key":"R12","unstructured":"roLabelImg. GitHub. URL: https:\/\/github.com\/cgvict\/roLabelImg [Accessed 03-07-2025]"},{"issue":"4","key":"R13","doi-asserted-by":"publisher","first-page":"1680","DOI":"10.3390\/make5040083","article-title":"A comprehensive review of YOLO architectures in computer vision: from YOLOv1 to YOLOv8 and YOLO-NAS","volume":"5","author":"Terven","journal-title":"MAKE"},{"issue":"2","key":"R14","doi-asserted-by":"publisher","first-page":"336","DOI":"10.1007\/s11263-019-01228-7","article-title":"Grad-CAM: visual explanations from deep networks via gradient-based localization","volume":"128","author":"Selvaraju","journal-title":"Int J Comput Vis"},{"issue":"8","key":"R15","doi-asserted-by":"publisher","DOI":"10.1371\/journal.pone.0255643","article-title":"An automatic screening method for strabismus detection based on image processing","volume":"16","author":"Huang","journal-title":"PLoS One"},{"key":"R16","unstructured":"Dai JF Li Y He KM Sun J . R-FCN: object detection via region-based fully convolutional networks. Presented at: 30th Conference on Neural Information Processing Systems (NIPS); Dec 5-10, 2016; Barcelona, Spain. [doi: 10.5555\/3157096.3157139]"},{"key":"R17","doi-asserted-by":"crossref","unstructured":"He K Gkioxari G Dollar P Girshick R . Mask r-CNN. Presented at: 2017 IEEE International Conference on Computer Vision (ICCV); Oct 22-29, 2017; Venice. [doi: 10.1109\/ICCV.2017.322]","DOI":"10.1109\/ICCV.2017.322"},{"issue":"1","key":"R18","doi-asserted-by":"publisher","first-page":"33","DOI":"10.1167\/tvst.10.1.33","article-title":"Detection of referable horizontal strabismus in children\u2019s primary gaze photographs using deep learning","volume":"10","author":"Zheng","journal-title":"Transl Vis Sci Technol"},{"key":"R19","doi-asserted-by":"crossref","unstructured":"Redmon J Divvala S Girshick R Farhadi A . You only look once: unified, real-time object detection. Presented at: 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR); Jun 27-30, 2016; Las Vegas, NV. [doi: 10.1109\/CVPR.2016.91]","DOI":"10.1109\/CVPR.2016.91"},{"key":"R20","doi-asserted-by":"crossref","unstructured":"Liu W Anguelov D Erhan D et al. SSD: single shot multibox detector. Presented at: 14th European Conference on Computer Vision (ECCV); Oct 8-16, 2016; Amsterdam, Netherlands. [doi: 10.1007\/978-3-319-46448-0_2]","DOI":"10.1007\/978-3-319-46448-0_2"},{"key":"R21","unstructured":"Qijie Z Tao S Yongtao W Zhi T Ying C Ling C et al. M2Det: a single-shot object detector based on multi-level feature pyramid network. arXiv. Preprint posted online on Nov 12, 2018. [doi: 10.48550\/arXiv.1811.04533]"},{"key":"R22","doi-asserted-by":"crossref","unstructured":"Nie Y Sommella P O\u2019Nils M Liguori C Lundgren J . Automatic detection of melanoma with yolo deep convolutional neural networks. Presented at: 2019 E-Health and Bioengineering Conference (EHB); Nov 21-23, 2019; Iasi, Romania. [doi: 10.1109\/EHB47216.2019.8970033]","DOI":"10.1109\/EHB47216.2019.8970033"},{"issue":"3","key":"R23","doi-asserted-by":"publisher","DOI":"10.3390\/diagnostics9030072","article-title":"Skin lesion segmentation in dermoscopic images with combination of YOLO and GrabCut algorithm","volume":"9","author":"\u00dcnver","journal-title":"Diagnostics (Basel)"},{"issue":"1","key":"R24","doi-asserted-by":"publisher","DOI":"10.1186\/s12911-021-01691-8","article-title":"Comparison of RetinaNet, SSD, and YOLO v3 for real-time pill identification","volume":"21","author":"Tan","journal-title":"BMC Med Inform Decis Mak"},{"key":"R25","unstructured":"Yolov8 for oriented object detection. GitHub. URL: https:\/\/github.com\/victor250214384\/yolov8obb-for-strbismus [Accessed 03-07-2025]"}],"container-title":["Journal of Medical Internet Research"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.jmir.org\/2025\/1\/e74402","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,6,30]],"date-time":"2026-06-30T18:05:34Z","timestamp":1782842734000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.jmir.org\/2025\/1\/e74402"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,7,17]]},"references-count":25,"URL":"https:\/\/doi.org\/10.2196\/74402","relation":{},"ISSN":["1438-8871"],"issn-type":[{"value":"1438-8871","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,7,17]]},"article-number":"v27i10e74402"}}