{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,26]],"date-time":"2026-06-26T18:45:29Z","timestamp":1782499529962,"version":"3.54.5"},"reference-count":23,"publisher":"Public Library of Science (PLoS)","issue":"6","license":[{"start":{"date-parts":[[2026,6,26]],"date-time":"2026-06-26T00:00:00Z","timestamp":1782432000000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100002241","name":"Japan Science and Technology Agency","doi-asserted-by":"publisher","award":["JPMJST2171"],"award-info":[{"award-number":["JPMJST2171"]}],"id":[{"id":"10.13039\/501100002241","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100008731","name":"Nakatomi Foundation","doi-asserted-by":"publisher","award":["20211279"],"award-info":[{"award-number":["20211279"]}],"id":[{"id":"10.13039\/100008731","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["www.plosdigitalhealth.org"],"crossmark-restriction":false},"short-container-title":["PLOS Digit Health"],"abstract":"<jats:p>Locomotive syndrome (LS), characterized by declining mobility due to musculoskeletal disorders, significantly affects older adults and is highly prevalent in Japan. Early detection and intervention are crucial to mitigate disease progression and preserve quality of life. Current diagnostic methods rely on subjective self-reports and labor-intensive clinical assessments, underscoring the need for automated, objective, and scalable screening tools. Building on our previous work, this study presents an updated deep learning-based computer vision model for LS screening using single-camera walking videos. We collected 511 walking videos from 178 participants clinically classified into four LS stages (non-LS, stage 1, stage 2, and stage 3). We used the MMPose framework to extract 2D body keypoints and generated silhouette-based Gait Energy Images (GEIs) from seven frames automatically sampled from fixed 40-frame walking windows to capture complementary gait-shape information. Model performance was evaluated using 10-fold cross-validation based on held-out evaluation-window predictions, while ensuring that windows from the same participant and the same walking video were not split across training, validation, and test sets. We also evaluated an independent external validation cohort of 33 cases (LS0\u20131, n\u2009=\u200915; LS2\u20133, n\u2009=\u200918). In internal validation, the pose-estimation model achieved sensitivity of 0.9440 and an AUC of 0.9668, whereas the silhouette model showed specificity of 0.9085 and an AUC of 0.9191. Fusion models outperformed the individual approaches, with the highest F1-score (0.908) at a 60:40 pose-to-silhouette ratio and the highest AUC (0.971) at a 70:30 ratio. In external validation, the score-fusion model achieved 90.9% accuracy, 88.9% sensitivity, 93.3% specificity, a macro-F1 score of 0.909, and an AUC of 0.967. In conclusion, integrating pose-estimation and silhouette-based features substantially improves the accuracy and clinical utility of automated LS screening. This low-burden approach provides a practical bridge between population-level screening and definitive clinical assessment.<\/jats:p>","DOI":"10.1371\/journal.pdig.0001530","type":"journal-article","created":{"date-parts":[[2026,6,26]],"date-time":"2026-06-26T17:45:26Z","timestamp":1782495926000},"page":"e0001530","update-policy":"https:\/\/doi.org\/10.1371\/journal.pdig.corrections_policy","source":"Crossref","is-referenced-by-count":0,"title":["A single-camera video-based assessment of locomotive syndrome using pose-silhouette fusion model"],"prefix":"10.1371","volume":"5","author":[{"ORCID":"https:\/\/orcid.org\/0009-0000-4027-4688","authenticated-orcid":true,"given":"Yu","family":"Moriguchi","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-0738-1569","authenticated-orcid":true,"given":"Junichi","family":"Kushioka","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Noriko","family":"Takemura","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Hiroki","family":"Hagizawa","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Masafumi","family":"Kashii","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ichiro","family":"Nakahara","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Taichi","family":"Kimura","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Keisuke","family":"Akiyama","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Taku","family":"Fujimoto","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Hajime","family":"Nagahara","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Keiko","family":"Yamada","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Rodrigo","family":"Navarro-Ramirez","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Takenori","family":"Oda","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Hideki","family":"Mochizuki","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ken","family":"Nakata","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Seiji","family":"Okada","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Satoru","family":"Tada","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"340","published-online":{"date-parts":[[2026,6,26]]},"reference":[{"key":"pdig.0001530.ref001","doi-asserted-by":"crossref","first-page":"56","DOI":"10.1007\/s12018-016-9208-2","article-title":"Locomotive syndrome: definition and management","volume":"14","author":"K Nakamura","year":"2016","journal-title":"Clin Rev Bone Miner Metab"},{"issue":"1","key":"pdig.0001530.ref002","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1080\/14397595.2016.1226471","article-title":"Epidemiology of the locomotive syndrome: the research on osteoarthritis\/osteoporosis against disability study 2005-2015","volume":"27","author":"N Yoshimura","year":"2017","journal-title":"Mod Rheumatol"},{"issue":"5","key":"pdig.0001530.ref003","doi-asserted-by":"crossref","first-page":"921","DOI":"10.1080\/14397595.2019.1665616","article-title":"Differences of locomotive syndrome and frailty in community-dwelling middle-aged and elderly people: pain, osteoarthritis, spinal alignment, body balance, and quality of life","volume":"30","author":"S Imagama","year":"2020","journal-title":"Mod Rheumatol"},{"issue":"6","key":"pdig.0001530.ref004","doi-asserted-by":"crossref","first-page":"458","DOI":"10.1111\/ggi.14162","article-title":"Relationship between locomotive syndrome, frailty and sarcopenia: locomotive syndrome overlapped in the majority of frailty and sarcopenia patients","volume":"21","author":"K Ide","year":"2021","journal-title":"Geriatr Gerontol Int"},{"issue":"1","key":"pdig.0001530.ref005","doi-asserted-by":"crossref","first-page":"96","DOI":"10.1093\/ageing\/afaa219","article-title":"Prevalence of frailty in 62 countries across the world: a systematic review and meta-analysis of population-level studies","volume":"50","author":"R O\u2019Caoimh","year":"2021","journal-title":"Age Ageing"},{"issue":"4","key":"pdig.0001530.ref006","doi-asserted-by":"crossref","first-page":"678","DOI":"10.1016\/j.jos.2020.06.012","article-title":"Locomotive syndrome: prevalence, surgical outcomes, and physical performance of patients treated to correct adult spinal deformity","volume":"26","author":"T Ohba","year":"2021","journal-title":"J Orthop Sci"},{"key":"pdig.0001530.ref007","doi-asserted-by":"crossref","first-page":"819","DOI":"10.2147\/CIA.S148683","article-title":"Locomotive syndrome: clinical perspectives","volume":"13","author":"T Ikemoto","year":"2018","journal-title":"Clin Interv Aging"},{"issue":"11","key":"pdig.0001530.ref008","doi-asserted-by":"crossref","DOI":"10.1371\/journal.pdig.0000668","article-title":"Deep learning-based screening for locomotive syndrome using single-camera walking video: development and validation study","volume":"3","author":"J Kushioka","year":"2024","journal-title":"PLOS Digit Health"},{"issue":"1","key":"pdig.0001530.ref009","doi-asserted-by":"crossref","DOI":"10.1186\/s41074-019-0054-2","article-title":"Gait-based age estimation using multi-stage convolutional neural network","volume":"11","author":"A Sakata","year":"2019","journal-title":"IPSJ T Comput Vis Appl"},{"issue":"1","key":"pdig.0001530.ref010","article-title":"Spatio-temporal silhouette sequence reconstruction for gait recognition against occlusion","volume":"11","author":"MdZ Uddin","year":"2019","journal-title":"IPSJ T Comput Vis Appl"},{"issue":"1","key":"pdig.0001530.ref011","doi-asserted-by":"crossref","first-page":"172","DOI":"10.1109\/TPAMI.2019.2929257","article-title":"OpenPose: realtime multi-person 2D pose estimation using part affinity fields","volume":"43","author":"Z Cao","year":"2021","journal-title":"IEEE Trans Pattern Anal Mach Intell"},{"key":"pdig.0001530.ref012","author":"MMPose C","year":"2020"},{"key":"pdig.0001530.ref013","doi-asserted-by":"crossref","first-page":"1678","DOI":"10.1587\/transinf.2020ZDP7502","article-title":"Health indicator estimation by video-based gait analysis","author":"R Liao","year":"2021","journal-title":"IEICE Trans Inf Syst"},{"issue":"2","key":"pdig.0001530.ref014","doi-asserted-by":"crossref","first-page":"316","DOI":"10.1109\/TPAMI.2006.38","article-title":"Individual recognition using gait energy image","volume":"28","author":"J Han","year":"2006","journal-title":"IEEE Trans Pattern Anal Mach Intell"},{"key":"pdig.0001530.ref015","doi-asserted-by":"crossref","unstructured":"He K, Zhang X, Ren S, Sun J. Deep residual learning for image recognition. In: 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2016. pp. 770\u20138. Available from: https:\/\/doi.org\/10.1109\/cvpr.2016.90","DOI":"10.1109\/CVPR.2016.90"},{"key":"pdig.0001530.ref016","article-title":"Disentangling and unifying graph convolutions for skeleton-based action recognition","author":"Z Liu","year":"2020","journal-title":"arXiv"},{"key":"pdig.0001530.ref017","article-title":"GPGait: generalized pose-based gait recognition","author":"Y Fu","year":"2023","journal-title":"arXiv"},{"key":"pdig.0001530.ref018","doi-asserted-by":"crossref","first-page":"102623","DOI":"10.1109\/ACCESS.2023.3317437","article-title":"Effective fusion method on silhouette and pose for gait recognition","volume":"11","author":"Y Zhao","year":"2023","journal-title":"IEEE Access"},{"issue":"1","key":"pdig.0001530.ref019","doi-asserted-by":"crossref","first-page":"105","DOI":"10.1038\/s41746-022-00654-1","article-title":"Silhouette images enable estimation of body fat distribution and associated cardiometabolic risk","volume":"5","author":"MDR Klarqvist","year":"2022","journal-title":"NPJ Digit Med"},{"key":"pdig.0001530.ref020","doi-asserted-by":"crossref","first-page":"4189","DOI":"10.1109\/TNSRE.2023.3325215","article-title":"A systematic review of gait analysis in the context of multimodal sensing fusion and AI","volume":"31","author":"R Katmah","year":"2023","journal-title":"IEEE Trans Neural Syst Rehabil Eng"},{"issue":"1","key":"pdig.0001530.ref021","doi-asserted-by":"crossref","DOI":"10.1371\/journal.pone.0315453","article-title":"Parkinson\u2019s disease screening using a fusion of gait point cloud and silhouette features","volume":"20","author":"T Connie","year":"2025","journal-title":"PLoS One"},{"issue":"5","key":"pdig.0001530.ref022","doi-asserted-by":"crossref","first-page":"888","DOI":"10.1007\/s00776-015-0737-1","article-title":"Development of a screening program to assess motor function in the adult population: a cross-sectional observational study","volume":"20","author":"T Ogata","year":"2015","journal-title":"J Orthop Sci"},{"key":"pdig.0001530.ref023","article-title":"Graphonomy: universal human parsing via graph transfer learning","author":"K Gong","year":"2019","journal-title":"arXiv"}],"container-title":["PLOS Digital Health"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pdig.0001530","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,6,26]],"date-time":"2026-06-26T17:45:33Z","timestamp":1782495933000},"score":1,"resource":{"primary":{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pdig.0001530"}},"subtitle":[],"editor":[{"given":"Kaisaierjiang","family":"Kadier","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"editor"}]}],"short-title":[],"issued":{"date-parts":[[2026,6,26]]},"references-count":23,"journal-issue":{"issue":"6","published-online":{"date-parts":[[2026,6,26]]}},"URL":"https:\/\/doi.org\/10.1371\/journal.pdig.0001530","relation":{},"ISSN":["2767-3170"],"issn-type":[{"value":"2767-3170","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,6,26]]}}}