{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,24]],"date-time":"2026-09-24T09:04:12Z","timestamp":1790240652352,"version":"4.1.0"},"reference-count":30,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,9,24]],"date-time":"2026-09-24T00:00:00Z","timestamp":1790208000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Bioeng. Biotechnol."],"abstract":"<jats:sec>\n                    <jats:title>Background<\/jats:title>\n                    <jats:p>Balance dysfunction is a common health issue among older adults and patients with neurological disorders. The Mini-Balance Evaluation Systems Test (Mini-BESTest) is a widely used clinical tool for balance assessment. However, its full implementation requires a trained clinician to administer 14 tasks, and the scoring is subject to inter-rater variability.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Objective<\/jats:title>\n                    <jats:p>This study aims to develop an end-to-end, lightweight multimodal deep learning framework (with approximately 55\u00a0K parameters) to automatically predict the total Mini-BESTest score and its four subscale scores, utilizing only center of pressure (COP) signals from a single stance condition alongside clinical metadata.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>COP signals and clinical metadata from 163 subjects in a public dataset (Santos et al., PeerJ, 2016, 4) were utilized. We proposed a lightweight multimodal deep learning model comprising three dilated convolution branches with varying dilation rates, a channel attention mechanism, and a metadata fusion module. Three independent comparative experiments were conducted to investigate the effects of stance condition, trial number, and time window length on predictive performance. Based on these results, a suitable data configuration was selected for subsequent experiments. An adaptive task weighting scheme based on homoscedastic uncertainty was employed to balance the five regression targets with varying score ranges. Five-fold cross-validation was used to evaluate the models, and out-of-fold (OOF) predictions were constructed to provide a full-sample assessment of predictive performance.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>\n                      Under the selected data configuration (foam surface with eyes open, second trial, first 30\u00a0s of data), the five-fold cross-validation mean absolute error (MAE) for the predicted Mini-BESTest total score was 1.971 (out of 28). The full-sample OOF prediction MAE was 1.979, with a Pearson correlation coefficient of r = 0.758 and R\n                      <jats:sup>2<\/jats:sup>\n                      = 0.574. In ablation experiments, removing the metadata branch led to the largest numerical increase in total-score MAE (to 2.338), while removal of the channel attention module resulted in a smaller increase (to 2.034). A metadata-only model achieved a total-score MAE of 2.105. Paired Wilcoxon signed-rank tests did not show statistically significant differences between the full model and the ablated variants (all p &amp;gt; 0.05).\n                    <\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusion<\/jats:title>\n                    <jats:p>This study demonstrates that an end-to-end deep learning framework can predict the complete Mini-BESTest score from 30\u00a0s of quiet standing COP signals under foam surface eyes-open conditions, together with basic clinical metadata. While maintaining predictive accuracy, this lightweight framework (approx. 55\u00a0K parameters) may represent a potential direction for future deployment in portable or remote settings, subject to further engineering and clinical validation.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.3389\/fbioe.2026.1936471","type":"journal-article","created":{"date-parts":[[2026,9,24]],"date-time":"2026-09-24T08:26:27Z","timestamp":1790238387000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["Automated assessment of balance function based on a lightweight end-to-end multimodal deep learning framework: predicting Mini-BESTest scores from center of pressure signals in a single stance condition"],"prefix":"10.3389","volume":"14","author":[{"given":"Yuyi","family":"Fan","sequence":"first","affiliation":[{"name":"School of Health Science and Engineering, University of Shanghai for Science and Technology","place":["Shanghai, China"]},{"name":"College of Rehabilitation Sciences, Shanghai University of Medicine and Health Sciences","place":["Shanghai, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yunfei","family":"Gui","sequence":"additional","affiliation":[{"name":"School of Health Science and Engineering, University of Shanghai for Science and Technology","place":["Shanghai, China"]},{"name":"College of Rehabilitation Sciences, Shanghai University of Medicine and Health Sciences","place":["Shanghai, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Linyan","family":"Liu","sequence":"additional","affiliation":[{"name":"College of Medical Instruments, Shanghai University of Medicine and Health Sciences","place":["Shanghai, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Linlin","family":"Zhang","sequence":"additional","affiliation":[{"name":"College of Rehabilitation Sciences, Shanghai University of Medicine and Health Sciences","place":["Shanghai, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yi","family":"Yu","sequence":"additional","affiliation":[{"name":"College of Medical Instruments, Shanghai University of Medicine and Health Sciences","place":["Shanghai, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Roger","family":"Adams","sequence":"additional","affiliation":[{"name":"Research Institute for Sport and Exercise, University of Canberra","place":["Canberra, ACT, Australia"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jia","family":"Han","sequence":"additional","affiliation":[{"name":"College of Rehabilitation Sciences, Shanghai University of Medicine and Health Sciences","place":["Shanghai, China"]},{"name":"Research Institute for Sport and Exercise, University of Canberra","place":["Canberra, ACT, Australia"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jinfeng","family":"Cao","sequence":"additional","affiliation":[{"name":"Periodicals Agency of Shanghai University, Shanghai University","place":["Shanghai, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jie","family":"Lyu","sequence":"additional","affiliation":[{"name":"Department of Orthopedics, Jinshan District Central Hospital Affiliated to Shanghai University of Medicine and Health Sciences","place":["Shanghai, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2026,9,24]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","DOI":"10.1038\/s41598-026-60745-x","article-title":"Video-based dual-task four-square step test for fall risk assessment in Parkinson\u2019s disease","author":"Aytutuldu","year":"2026","journal-title":"Sci. Rep"},{"key":"B2","doi-asserted-by":"publisher","first-page":"474","DOI":"10.1007\/s11227-026-08629-9","article-title":"Arrhythmiavision: resource-Conscious deep learning models with visual explanations for ecg arrhythmia classification","volume":"82","author":"Baig","year":"2026","journal-title":"J. Supercomput."},{"key":"B3","doi-asserted-by":"publisher","first-page":"509","DOI":"10.15585\/mmwr.mm6718a1","article-title":"Deaths from falls among persons aged \u226565 years\u2014United states, 2007\u20132016","volume":"67","author":"Burns","year":"2018","journal-title":"Morb. Mortal. Wkly. Rep."},{"key":"B4","doi-asserted-by":"publisher","first-page":"1228302","DOI":"10.3389\/fneur.2023.1228302","article-title":"Pay attention: you can fall! the Mini-BESTest scale and the turning duration of the TUG test provide valid balance measures in neurological patients: a prospective study with falls as the balance criterion","volume":"14","author":"Caronni","year":"2023","journal-title":"Front. Neurology"},{"key":"B5","doi-asserted-by":"publisher","first-page":"1292","DOI":"10.1109\/TNSRE.2024.3378112","article-title":"Deep learning model to evaluate sensorimotor system ability in patients with dizziness for postural control","volume":"32","author":"Choi","year":"2024","journal-title":"IEEE Trans. Neural Syst. Rehabilitation Eng."},{"key":"B6","first-page":"7482","article-title":"Multi-task learning using uncertainty to weigh losses for scene geometry and semantics","volume-title":"2018 IEEE\/CVF Conference on Computer Vision and Pattern Recognition, (Salt Lake City, UT, USA: IEEE)","author":"Cipolla","year":"2018"},{"key":"B7","doi-asserted-by":"publisher","first-page":"97","DOI":"10.1097\/MRR.0000000000000153","article-title":"The Mini-BESTest: a review of psychometric properties","volume":"39","author":"Di Carlo","year":"2016","journal-title":"Int. J. Rehabilitation Res."},{"key":"B8","doi-asserted-by":"publisher","first-page":"4","DOI":"10.3991\/ijoe.v22i04.59523","article-title":"The role of artificial intelligence and machine learning in balance classification using center of Pressure\u2014A comprehensive review","volume":"22","author":"Duangnga","year":"2026","journal-title":"Int. J. Online Biomed. Eng."},{"key":"B9","doi-asserted-by":"publisher","first-page":"183","DOI":"10.1590\/S1413-35552010000300003","article-title":"Revision of posturography based on force plate for balance evaluation","volume":"14","author":"Duarte","year":"2010","journal-title":"Braz. J. Phys. Ther."},{"key":"B10","doi-asserted-by":"publisher","first-page":"323","DOI":"10.2340\/16501977-0537","article-title":"Using psychometric techniques to improve the balance evaluation systems test: the Mini-BESTest","volume":"42","author":"Franchignoni","year":"2010","journal-title":"J. Rehabilitation Med."},{"key":"B11","doi-asserted-by":"publisher","first-page":"734","DOI":"10.1056\/NEJMcp1903252","article-title":"Prevention of falls in community-dwelling older adults","volume":"382","author":"Ganz","year":"2020","journal-title":"N. Engl. J. Med."},{"key":"B12","doi-asserted-by":"publisher","first-page":"7849","DOI":"10.3390\/e17127849","article-title":"Multiscale entropy analysis of center-of-pressure dynamics in human postural control: methodological considerations","volume":"17","author":"Gow","year":"2015","journal-title":"Entropy"},{"key":"B13","doi-asserted-by":"publisher","first-page":"484","DOI":"10.2522\/ptj.20080071","article-title":"The balance evaluation systems test (BESTest) to differentiate balance deficits","volume":"89","author":"Horak","year":"2009","journal-title":"Phys. Ther."},{"key":"B14","doi-asserted-by":"publisher","first-page":"58","DOI":"10.1519\/JPT.0000000000000047","article-title":"Reliability, validity, and minimal detectable change of balance evaluation systems test and its short versions in older cancer survivors: a pilot study","volume":"39","author":"Huang","year":"2016","journal-title":"J. Geriatric Phys. Ther."},{"key":"B15","doi-asserted-by":"publisher","first-page":"3860","DOI":"10.1109\/JBHI.2023.3331278","article-title":"LDMRes-Net: a lightweight deep learning model for medical image segmentation on edge devices","volume":"28","author":"Iqbal","year":"2024","journal-title":"IEEE J. Biomed. Health Inf."},{"key":"B16","doi-asserted-by":"publisher","first-page":"e2695","DOI":"10.7717\/peerj-cs.2695","article-title":"AI-driven balance evaluation: a comparative study between blind and non-blind individuals using the Mini-BESTest","volume":"11","author":"Ja\u00e9n-Vargas","year":"2025","journal-title":"PeerJ Comput. Sci."},{"key":"B17","first-page":"239","article-title":"The relevance of clinical balance assessment tools to differentiate balance deficits","volume":"46","author":"Mancini","year":"2010","journal-title":"Eur. Journal Physical Rehabilitation Medicine"},{"key":"B18","doi-asserted-by":"publisher","first-page":"4593","DOI":"10.1080\/09638288.2025.2456602","article-title":"Psychometric properties of the mini-balance evaluation systems test (Mini-BESTest) among multiple populations: a COSMIN systematic review and meta-analysis","volume":"47","author":"Meseguer-Henarejos","year":"2025","journal-title":"Disabil. Rehabilitation"},{"key":"B19","doi-asserted-by":"publisher","first-page":"676","DOI":"10.1038\/s41393-024-01032-2","article-title":"Reliability of the Mini-BESTest and Brief-BESTest for assessing patients with incomplete spinal cord injury","volume":"62","author":"Morooka","year":"2024","journal-title":"Spinal Cord."},{"key":"B20","doi-asserted-by":"publisher","first-page":"51","DOI":"10.1123\/jsr.11.1.51","article-title":"Center-of-pressure parameters used in the assessment of postural control","volume":"11","author":"Palmieri","year":"2002","journal-title":"J. Sport Rehabilitation"},{"key":"B21","doi-asserted-by":"publisher","first-page":"1339","DOI":"10.1016\/j.chaos.2005.04.111","article-title":"Considerations on the application of the chaos paradigm to describe the postural sway","volume":"27","author":"Pascolo","year":"2006","journal-title":"Chaos, Solit. and Fractals"},{"key":"B22","doi-asserted-by":"publisher","first-page":"3460","DOI":"10.1109\/TBME.2022.3170850","article-title":"Deep learning-based prediction of postural stability using short-term center of pressure data","volume":"69","author":"Phan","year":"2022","journal-title":"IEEE Trans. Biomed. Eng."},{"key":"B23","doi-asserted-by":"publisher","first-page":"658","DOI":"10.1109\/TNSRE.2020.2966784","article-title":"Assessment of balance control subsystems by artificial intelligence","volume":"28","author":"Ren","year":"2020","journal-title":"IEEE Trans. Neural Syst. Rehabilitation Eng."},{"key":"B24","doi-asserted-by":"publisher","first-page":"126","DOI":"10.1080\/09593985.2019.1622161","article-title":"Reliability and minimal detectable change of the Mini-BESTest in adults with spinal cord injury in a rehabilitation setting","volume":"37","author":"Roy","year":"2021","journal-title":"Physiother. Theory Pract."},{"key":"B25","doi-asserted-by":"publisher","first-page":"358","DOI":"10.1007\/s00586-010-1543-2","article-title":"Center of pressure excursion as a measure of balance performance in patients with non-specific low back pain compared to healthy controls: a systematic review of the literature","volume":"20","author":"Ruhe","year":"2011","journal-title":"Eur. Spine J."},{"key":"B26","doi-asserted-by":"publisher","first-page":"e2648","DOI":"10.7717\/peerj.2648","article-title":"A public data set of human balance evaluations","volume":"4","author":"Santos","year":"2016","journal-title":"PeerJ"},{"key":"B27","doi-asserted-by":"publisher","first-page":"290","DOI":"10.1016\/j.gaitpost.2012.07.009","article-title":"Clinical stabilometry standardization: basic Definitions\u2014Acquisition Interval\u2014Sampling frequency","volume":"37","author":"Scoppa","year":"2013","journal-title":"Gait and Posture"},{"key":"B28","doi-asserted-by":"publisher","first-page":"122","DOI":"10.1016\/j.apmr.2014.06.021","article-title":"Using the systems framework for postural control to analyze the components of balance evaluated in standardized balance measures: a scoping review","volume":"96","author":"Sibley","year":"2015","journal-title":"Archives Phys. Med. Rehabilitation"},{"key":"B29","doi-asserted-by":"publisher","first-page":"1102","DOI":"10.2522\/ptj.20120454","article-title":"Psychometric properties of the mini-balance evaluation systems test (Mini-BESTest) in community-dwelling individuals with chronic stroke","volume":"93","author":"Tsang","year":"2013","journal-title":"Phys. Ther."},{"key":"B30","doi-asserted-by":"publisher","first-page":"64","DOI":"10.1519\/JPT.0000000000000050","article-title":"The mini-balance evaluation systems test (Mini-BESTest) demonstrates higher accuracy in identifying older adult participants with history of falls than do the BESTest, berg balance scale, or timed up and Go test","volume":"39","author":"Yingyongyudha","year":"2016","journal-title":"J. Geriatric Phys. Ther."}],"container-title":["Frontiers in Bioengineering and Biotechnology"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fbioe.2026.1936471\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,24]],"date-time":"2026-09-24T08:26:28Z","timestamp":1790238388000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fbioe.2026.1936471\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,9,24]]},"references-count":30,"alternative-id":["10.3389\/fbioe.2026.1936471"],"URL":"https:\/\/doi.org\/10.3389\/fbioe.2026.1936471","relation":{},"ISSN":["2296-4185"],"issn-type":[{"value":"2296-4185","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,9,24]]},"article-number":"1936471"}}