{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,16]],"date-time":"2026-05-16T17:57:57Z","timestamp":1778954277785,"version":"3.51.4"},"reference-count":34,"publisher":"MDPI AG","issue":"20","license":[{"start":{"date-parts":[[2025,10,13]],"date-time":"2025-10-13T00:00:00Z","timestamp":1760313600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100000038","name":"NSERC","doi-asserted-by":"publisher","award":["RGPIN-2020-06338"],"award-info":[{"award-number":["RGPIN-2020-06338"]}],"id":[{"id":"10.13039\/501100000038","id-type":"DOI","asserted-by":"publisher"}]},{"name":"the Canadian Frailty Network","award":["ECR2022-008"],"award-info":[{"award-number":["ECR2022-008"]}]},{"DOI":"10.13039\/501100000142","name":"the Arthritis Society of Canada through a PhD Salary Award","doi-asserted-by":"publisher","award":["22-0000000169"],"award-info":[{"award-number":["22-0000000169"]}],"id":[{"id":"10.13039\/501100000142","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Sensors"],"abstract":"<jats:p>Wearable sensors have become valuable tools for assessing gait in both laboratory and free-living environments. However, detection of walking in free-living environments remains challenging, especially in clinical populations. Machine learning models may offer more robust gait identification, but most are trained on healthy participants, limiting their generalizability to other populations. To extend a previously validated machine learning model, an updated model was trained using an open dataset (PAMAP2), before progressively including training datasets with additional healthy participants and a clinical osteoarthritis population. The performance of the model in identifying walking was also evaluated using a frequency-based gait detection algorithm. The results showed that the model trained with all three datasets performed best in terms of activity classification, ultimately achieving a high accuracy of 96% on held-out test data. The model generally performed on par with the heuristic, frequency-based method for walking bout identification. However, for patients with slower gait speeds (&lt;0.8 m\/s), the machine learning model maintained high recall (&gt;0.89), while the heuristic method performed poorly, with recall as low as 0.38. This study demonstrates the enhancement of existing model architectures by training with diverse datasets, highlighting the importance of dataset diversity when developing more robust models for clinical applications.<\/jats:p>","DOI":"10.3390\/s25206318","type":"journal-article","created":{"date-parts":[[2025,10,15]],"date-time":"2025-10-15T08:34:26Z","timestamp":1760517266000},"page":"6318","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":2,"title":["Augmenting a ResNet + BiLSTM Deep Learning Model with Clinical Mobility Data Helps Outperform a Heuristic Frequency-Based Model for Walking Bout Segmentation"],"prefix":"10.3390","volume":"25","author":[{"given":"Matthew C.","family":"Ruder","sequence":"first","affiliation":[{"name":"Department of Kinesiology, McMaster University, Hamilton, ON L8S 4L8, Canada"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-6223-2996","authenticated-orcid":false,"given":"Vincenzo E.","family":"Di Bacco","sequence":"additional","affiliation":[{"name":"Department of Kinesiology, McMaster University, Hamilton, ON L8S 4L8, Canada"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Kushang","family":"Patel","sequence":"additional","affiliation":[{"name":"Department of Computing and Software, McMaster University, Hamilton, ON L8S 4L8, Canada"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Rong","family":"Zheng","sequence":"additional","affiliation":[{"name":"Department of Computing and Software, McMaster University, Hamilton, ON L8S 4L8, Canada"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-2861-9636","authenticated-orcid":false,"given":"Kim","family":"Madden","sequence":"additional","affiliation":[{"name":"Department of Surgery, McMaster University, Hamilton, ON L8S 4L8, Canada"},{"name":"Research Institute of St. Joe\u2019s, Hamilton, ON L8N 4A6, Canada"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Anthony","family":"Adili","sequence":"additional","affiliation":[{"name":"Department of Surgery, McMaster University, Hamilton, ON L8S 4L8, Canada"},{"name":"Research Institute of St. Joe\u2019s, Hamilton, ON L8N 4A6, Canada"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-4268-3618","authenticated-orcid":false,"given":"Dylan","family":"Kobsar","sequence":"additional","affiliation":[{"name":"Department of Kinesiology, McMaster University, Hamilton, ON L8S 4L8, Canada"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"1968","published-online":{"date-parts":[[2025,10,13]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","first-page":"107","DOI":"10.1109\/JBHI.2024.3434973","article-title":"Validation of an IMU-Based Gait Analysis Method for Assessment of Fall Risk Against Traditional Methods","volume":"29","author":"Ruiz","year":"2025","journal-title":"IEEE J. Biomed. Health Inform."},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"2091","DOI":"10.1038\/s41598-020-59002-6","article-title":"Comparison of Gait Characteristics between Clinical and Daily Life Settings in Children with Cerebral Palsy","volume":"10","author":"Carcreff","year":"2020","journal-title":"Sci. Rep."},{"key":"ref_3","doi-asserted-by":"crossref","first-page":"1765","DOI":"10.1109\/JBHI.2018.2865218","article-title":"IMU-Based Classification of Parkinson\u2019s Disease From Gait: A Sensitivity Analysis on Sensor Location and Feature Selection","volume":"22","author":"Caramia","year":"2018","journal-title":"IEEE J. Biomed. Health Inform."},{"key":"ref_4","doi-asserted-by":"crossref","first-page":"1115","DOI":"10.1002\/acr.20492","article-title":"Patient Characteristics That Predict Progression of Knee Osteoarthritis: A Systematic Review of Prognostic Studies","volume":"63","author":"Chapple","year":"2011","journal-title":"Arthritis Care Res."},{"key":"ref_5","doi-asserted-by":"crossref","first-page":"109567","DOI":"10.1016\/j.jbiomech.2019.109567","article-title":"Measuring Markers of Aging and Knee Osteoarthritis Gait Using Inertial Measurement Units","volume":"99","author":"Hafer","year":"2020","journal-title":"J. Biomech."},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"49","DOI":"10.1016\/j.gaitpost.2020.07.004","article-title":"Measuring Gait Kinematics in Patients with Severe Hip Osteoarthritis Using Wearable Sensors","volume":"81","author":"Ismailidis","year":"2020","journal-title":"Gait Posture"},{"key":"ref_7","doi-asserted-by":"crossref","first-page":"105232","DOI":"10.1016\/j.clinbiomech.2020.105232","article-title":"Quantifying Varus Thrust in Knee Osteoarthritis Using Wearable Inertial Sensors: A Proof of Concept","volume":"80","author":"Costello","year":"2020","journal-title":"Clin. Biomech."},{"key":"ref_8","doi-asserted-by":"crossref","first-page":"5936","DOI":"10.1109\/JSEN.2019.2908417","article-title":"A Wearable Sensing and Training System: Towards Gait Rehabilitation for Elderly Patients With Knee Osteoarthritis","volume":"19","author":"He","year":"2019","journal-title":"IEEE Sens. J."},{"key":"ref_9","doi-asserted-by":"crossref","unstructured":"Bolam, S.M., Batinica, B., Yeung, T.C., Weaver, S., Cantamessa, A., Vanderboor, T.C., Yeung, S., Munro, J.T., Fernandez, J.W., and Besier, T.F. (2021). Remote Patient Monitoring with Wearable Sensors Following Knee Arthroplasty. Sensors, 21.","DOI":"10.3390\/s21155143"},{"key":"ref_10","doi-asserted-by":"crossref","first-page":"6","DOI":"10.1186\/s11556-019-0214-5","article-title":"Is Every-Day Walking in Older Adults More Analogous to Dual-Task Walking or to Usual Walking? Elucidating the Gaps between Gait Performance in the Lab and during 24\/7 Monitoring","volume":"16","author":"Hillel","year":"2019","journal-title":"Eur. Rev. Aging Phys. Act."},{"key":"ref_11","doi-asserted-by":"crossref","first-page":"81","DOI":"10.1109\/TNSRE.2004.843176","article-title":"Automated Estimation of Initial and Terminal Contact Timing Using Accelerometers; Development and Validation in Transtibial Amputees and Controls","volume":"13","author":"Selles","year":"2005","journal-title":"IEEE Trans. Neural Syst. Rehabil. Eng."},{"key":"ref_12","first-page":"1869","article-title":"Detection of Gait From Continuous Inertial Sensor Data Using Harmonic Frequencies","volume":"24","author":"Ullrich","year":"2020","journal-title":"IEEE J. Biomed. Health Inform."},{"key":"ref_13","doi-asserted-by":"crossref","first-page":"479","DOI":"10.1016\/j.gaitpost.2015.07.063","article-title":"Gait Variability and Motor Control in People with Knee Osteoarthritis","volume":"42","author":"Alkjaer","year":"2015","journal-title":"Gait Posture"},{"key":"ref_14","doi-asserted-by":"crossref","first-page":"17966","DOI":"10.1038\/s41598-019-54399-1","article-title":"Open-Source Remote Gait Analysis: A Post-Surgery Patient Monitoring Application","volume":"9","author":"Gurchiek","year":"2019","journal-title":"Sci. Rep."},{"key":"ref_15","doi-asserted-by":"crossref","first-page":"93","DOI":"10.1186\/s12984-021-00883-7","article-title":"Hidden Markov Model Based Stride Segmentation on Unsupervised Free-Living Gait Data in Parkinson\u2019s Disease Patients","volume":"18","author":"Roth","year":"2021","journal-title":"J. Neuroeng. Rehabil."},{"key":"ref_16","doi-asserted-by":"crossref","first-page":"250","DOI":"10.1016\/j.neucom.2020.08.079","article-title":"Wearables-Based Multi-Task Gait and Activity Segmentation Using Recurrent Neural Networks","volume":"432","author":"Martindale","year":"2021","journal-title":"Neurocomputing"},{"key":"ref_17","doi-asserted-by":"crossref","unstructured":"Li, Y., and Wang, L. (2022). Human Activity Recognition Based on Residual Network and BiLSTM. Sensors, 22.","DOI":"10.3390\/s22020635"},{"key":"ref_18","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1016\/j.jbiomech.2018.09.009","article-title":"Machine Learning in Human Movement Biomechanics: Best Practices, Common Pitfalls, and New Opportunities","volume":"81","author":"Halilaj","year":"2018","journal-title":"J. Biomech."},{"key":"ref_19","unstructured":"Reiss, A. (2025, October 10). PAMAP2 Physical Activity Monitoring 2012. Available online: https:\/\/archive.ics.uci.edu\/dataset\/231\/pamap2+physical+activity+monitoring."},{"key":"ref_20","doi-asserted-by":"crossref","unstructured":"Kobsar, D., Masood, Z., Khan, H., Khalil, N., Kiwan, M.Y., Ridd, S., and Tobis, M. (2020). Wearable Inertial Sensors for Gait Analysis in Adults with Osteoarthritis\u2014A Scoping Review. Sensors, 20.","DOI":"10.3390\/s20247143"},{"key":"ref_21","doi-asserted-by":"crossref","unstructured":"Mihy, J.A., Wagatsuma, M., Cain, S.M., and Hafer, J.F. (2022). A Functional Sensor-to-Segment Orientation Method to Reduce the Effects of Varied Sensor Placement: Implications for IMU Best Practices. medRxiv.","DOI":"10.1101\/2022.11.29.22282894"},{"key":"ref_22","doi-asserted-by":"crossref","first-page":"229","DOI":"10.1016\/j.gaitpost.2012.07.012","article-title":"Quantitative Estimation of Foot-Flat and Stance Phase of Gait Using Foot-Worn Inertial Sensors","volume":"37","author":"Mariani","year":"2013","journal-title":"Gait Posture"},{"key":"ref_23","doi-asserted-by":"crossref","first-page":"47","DOI":"10.1016\/j.jphys.2022.11.005","article-title":"Normal Gait Speed Varies by Age and Sex but Not by Geographical Region: A Systematic Review","volume":"69","author":"Andrews","year":"2023","journal-title":"J. Physiother."},{"key":"ref_24","doi-asserted-by":"crossref","first-page":"1334","DOI":"10.1111\/pme.12478","article-title":"Correlates of Gait Speed in Advanced Knee Osteoarthritis","volume":"15","author":"Marcum","year":"2014","journal-title":"Pain Med"},{"key":"ref_25","doi-asserted-by":"crossref","first-page":"322","DOI":"10.5312\/wjo.v8.i4.322","article-title":"Abnormal Ground Reaction Forces Lead to a General Decline in Gait Speed in Knee Osteoarthritis Patients","volume":"8","author":"Wiik","year":"2017","journal-title":"WJO"},{"key":"ref_26","doi-asserted-by":"crossref","first-page":"1390","DOI":"10.1016\/j.apmr.2010.06.015","article-title":"Differences in Preferred Walking Speeds in a Gait Laboratory Compared With the Real World After Total Hip Replacement","volume":"91","author":"Foucher","year":"2010","journal-title":"Arch. Phys. Med. Rehabil."},{"key":"ref_27","doi-asserted-by":"crossref","first-page":"3496","DOI":"10.1038\/s41598-019-39695-0","article-title":"Relationship between Daily and In-Laboratory Gait Speed among Healthy Community-Dwelling Older Adults","volume":"9","author":"Takayanagi","year":"2019","journal-title":"Sci. Rep."},{"key":"ref_28","doi-asserted-by":"crossref","unstructured":"Romijnders, R., Warmerdam, E., Hansen, C., Schmidt, G., and Maetzler, W. (2022). A Deep Learning Approach for Gait Event Detection from a Single Shank-Worn IMU: Validation in Healthy and Neurological Cohorts. Sensors, 22.","DOI":"10.3390\/s22103859"},{"key":"ref_29","doi-asserted-by":"crossref","first-page":"100069","DOI":"10.1016\/j.ocarto.2020.100069","article-title":"Machine Learning in Knee Osteoarthritis: A Review","volume":"2","author":"Kokkotis","year":"2020","journal-title":"Osteoarthr. Cartil. Open"},{"key":"ref_30","doi-asserted-by":"crossref","first-page":"7316954","DOI":"10.1155\/2018\/7316954","article-title":"Deep Residual Bidir-LSTM for Human Activity Recognition Using Wearable Sensors","volume":"2018","author":"Zhao","year":"2018","journal-title":"Math. Probl. Eng."},{"key":"ref_31","doi-asserted-by":"crossref","unstructured":"Nafea, O., Abdul, W., Muhammad, G., and Alsulaiman, M. (2021). Sensor-Based Human Activity Recognition with Spatio-Temporal Deep Learning. Sensors, 21.","DOI":"10.3390\/s21062141"},{"key":"ref_32","doi-asserted-by":"crossref","first-page":"49","DOI":"10.1186\/s42836-023-00204-4","article-title":"Advances in the Application of Wearable Sensors for Gait Analysis after Total Knee Arthroplasty: A Systematic Review","volume":"5","author":"Feng","year":"2023","journal-title":"Arthroplasty"},{"key":"ref_33","doi-asserted-by":"crossref","first-page":"2093","DOI":"10.1002\/jor.24950","article-title":"Comparison of Wearable Sensor to Traditional Methods in Functional Outcome Measures: A Systematic Review","volume":"39","author":"Follis","year":"2021","journal-title":"J. Orthop. Res."},{"key":"ref_34","doi-asserted-by":"crossref","first-page":"e39396","DOI":"10.2196\/39396","article-title":"The Role of Wearable Technology in Measuring and Supporting Patient Outcomes Following Total Joint Replacement: Review of the Literature","volume":"6","author":"Iovanel","year":"2023","journal-title":"JMIR Perioper. Med."}],"container-title":["Sensors"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/1424-8220\/25\/20\/6318\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,15]],"date-time":"2025-10-15T08:43:01Z","timestamp":1760517781000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/1424-8220\/25\/20\/6318"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,10,13]]},"references-count":34,"journal-issue":{"issue":"20","published-online":{"date-parts":[[2025,10]]}},"alternative-id":["s25206318"],"URL":"https:\/\/doi.org\/10.3390\/s25206318","relation":{},"ISSN":["1424-8220"],"issn-type":[{"value":"1424-8220","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,10,13]]}}}