{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,9]],"date-time":"2026-10-09T17:54:20Z","timestamp":1791568460884,"version":"4.3.4"},"reference-count":56,"publisher":"Public Library of Science (PLoS)","issue":"10","license":[{"start":{"date-parts":[[2026,10,9]],"date-time":"2026-10-09T00:00:00Z","timestamp":1791504000000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["www.plosone.org"],"crossmark-restriction":false},"short-container-title":["PLoS One"],"abstract":"<jats:p>The prescription of lower limb prosthetics relies heavily on the assigned K-level. K-level assessment has been criticised for subjectivity and variability, potentially leading to inappropriate component selection and poorer functional outcomes. This study presents an original methodology for objectively determining activity levels from inertial sensor data, developed through rigorous data collection on 20 amputees who walked on various types of outdoor terrain (flat ground, stairs, ramp, grass\/uneven\/unstable surfaces), with and without a walking aid. Advanced machine learning algorithms were applied to the real-world IMU data, to establish a validated classification model for daily activity outside the laboratory. A random forest classifier achieved a mean leave\u2011some\u2011out accuracy of 65.70% (SD 6.6%; 95% CI 47.3\u201384.1%) and cross\u2011validation accuracy of 85.65% (SD 0.79%; 95% CI 83.4\u201387.9%). Per\u2011class sensitivity ranged from approximately 44\u201378% in leave\u2011some\u2011out analyses, with lower performance for ramp ascending and descending. The difference between cross\u2011validation and leave\u2011some\u2011out accuracies indicates participant\u2011specific overfitting, so the leave\u2011some\u2011out accuracy is considered the current conservative estimate of performance on unseen users and the cross-validation accuracy an estimate of potential accuracy with a more robust dataset. The research subsequently incorporated a preliminary clinical evaluation of the approach on a subset of 3 amputees using a data collection protocol for unsupervised, longitudinal real-world monitoring, with preliminary insight into perceived clinical significance of outcomes ascertained through exploratory interviews with 4 prescribing health care professionals. The findings suggest potential benefits of real-world inertial sensing, supporting a move from primarily subjective clinical assessment toward more objective, data-driven activity classification. This research demonstrates that an evidence-based tool for enhancing K-level prescription could in future provide more appropriate component selection and improve long-term mobility for prosthetic users, but current accuracies and sample sizes indicate that the system should be viewed as a promising proof\u2011of\u2011concept rather than a definitive tool for changing K\u2011level prescription.<\/jats:p>","DOI":"10.1371\/journal.pone.0358882","type":"journal-article","created":{"date-parts":[[2026,10,9]],"date-time":"2026-10-09T17:25:29Z","timestamp":1791566729000},"page":"e0358882","update-policy":"https:\/\/doi.org\/10.1371\/journal.pone.corrections_policy","source":"Crossref","is-referenced-by-count":0,"title":["Lower limb prosthetic users\u2019 activity classification through real-world inertial sensing"],"prefix":"10.1371","volume":"21","author":[{"ORCID":"https:\/\/orcid.org\/0000-0001-7288-1669","authenticated-orcid":true,"given":"Matthew","family":"Wassall","sequence":"first","affiliation":[{"name":"Norwegian University of Science and Technology","place":["Trondheim, Norway"],"department":["Department of Neuromedicine and Movement Science"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Malcolm","family":"Granat","sequence":"additional","affiliation":[{"name":"School of Health & Society, University of Salford","place":["Salford, United Kingdom"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-6959-632X","authenticated-orcid":true,"given":"Saeed","family":"Zahedi","sequence":"additional","affiliation":[{"name":"Blatchford Prosthetics","place":["Basingstoke, United Kingdom"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Sibylle","family":"Thies","sequence":"additional","affiliation":[{"name":"School of Health & Society, University of Salford","place":["Salford, United Kingdom"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"340","published-online":{"date-parts":[[2026,10,9]]},"reference":[{"key":"pone.0358882.ref001","volume-title":"WHO standards for prosthetics and orthotics","author":"WHO","year":"2017"},{"issue":"3","key":"pone.0358882.ref002","doi-asserted-by":"crossref","first-page":"422","DOI":"10.1016\/j.apmr.2007.11.005","article-title":"Estimating the prevalence of limb loss in the United States: 2005 to 2050","volume":"89","author":"K Ziegler-Graham","year":"2008","journal-title":"Arch Phys Med Rehabil"},{"key":"pone.0358882.ref003","article-title":"Do you know your k-level? An interview with interim president CEO Sue Stout about the importance of amputees knowing their k-levels","author":"Amputee-Coalition","year":"2013","journal-title":"inMotion"},{"key":"pone.0358882.ref004","doi-asserted-by":"crossref","DOI":"10.23970\/AHRQEPCCER213","volume-title":"Lower limb prostheses: measurement instruments, comparison of component effects by subgroups, and long-term outcomes","author":"EM Balk","year":"2018"},{"issue":"1","key":"pone.0358882.ref005","doi-asserted-by":"crossref","first-page":"33931","DOI":"10.33137\/cpoj.v3i1.33931","article-title":"The use of physical activity outcomes in rehabilitation interventions for lower limb amputees: A systematic review","volume":"3","author":"AG Jamieson","year":"2020","journal-title":"Can Prosthet Orthot J"},{"key":"pone.0358882.ref006","volume-title":"Advancing prosthetic prescriptions: developing a sensor-based system for objective activity assessment","author":"M Wassall","year":"2025"},{"issue":"6","key":"pone.0358882.ref007","doi-asserted-by":"crossref","first-page":"5683","DOI":"10.3390\/s100605683","article-title":"Gait event detection on level ground and incline walking using a rate gyroscope","volume":"10","author":"P Catalfamo","year":"2010","journal-title":"Sensors (Basel)"},{"issue":"3","key":"pone.0358882.ref008","doi-asserted-by":"crossref","first-page":"461","DOI":"10.1007\/s11517-019-02098-4","article-title":"Gait event detection using inertial measurement units in people with transfemoral amputation: a comparative study","volume":"58","author":"E Simonetti","year":"2020","journal-title":"Med Biol Eng Comput"},{"issue":"22","key":"pone.0358882.ref009","doi-asserted-by":"crossref","first-page":"8888","DOI":"10.3390\/s22228888","article-title":"Rules-based real-time gait event detection algorithm for lower-limb prosthesis users during level-ground and ramp walking","volume":"22","author":"A Gouda","year":"2022","journal-title":"Sensors (Basel)"},{"key":"pone.0358882.ref010","first-page":"5067","article-title":"Real-time gait event detection for lower limb amputees using a single wearable sensor","volume":"2016","author":"HF Maqbool","year":"2016","journal-title":"Annu Int Conf IEEE Eng Med Biol Soc"},{"issue":"12","key":"pone.0358882.ref011","doi-asserted-by":"crossref","first-page":"1168","DOI":"10.1080\/02640414.2015.1096018","article-title":"GT3X+ accelerometer placement affects the reliability of step-counts measured during running and pedal-revolution counts measured during bicycling","volume":"34","author":"AA Gatti","year":"2016","journal-title":"J Sports Sci"},{"key":"pone.0358882.ref012","doi-asserted-by":"crossref","first-page":"104927","DOI":"10.1016\/j.micpro.2023.104927","article-title":"Design and implementation of IMU-based locomotion mode recognition system on Zynq SoC","volume":"102","author":"L Madaoui","year":"2023","journal-title":"Microprocessors and Microsystems"},{"issue":"2","key":"pone.0358882.ref013","doi-asserted-by":"crossref","first-page":"235","DOI":"10.3390\/machines11020235","article-title":"Novel feature extraction and locomotion mode classification using intelligent lower-limb prosthesis","volume":"11","author":"Y Liu","year":"2023","journal-title":"Machines"},{"issue":"2","key":"pone.0358882.ref014","doi-asserted-by":"crossref","first-page":"436","DOI":"10.1109\/TMRB.2021.3075096","article-title":"A Classification approach based on directed acyclic graph to predict locomotion activities with one inertial sensor on the thigh","volume":"3","author":"V Papapicco","year":"2021","journal-title":"IEEE Trans Med Robot Bionics"},{"key":"pone.0358882.ref015","doi-asserted-by":"crossref","first-page":"923164","DOI":"10.3389\/fnbot.2022.923164","article-title":"Comparison of machine learning and deep learning-based methods for locomotion mode recognition using a single inertial measurement unit","volume":"16","author":"HTT Vu","year":"2022","journal-title":"Front Neurorobot"},{"issue":"4","key":"pone.0358882.ref016","doi-asserted-by":"crossref","first-page":"1232","DOI":"10.3390\/prosthesis5040085","article-title":"EMG and IMU data fusion for locomotion mode classification in transtibial amputees","volume":"5","author":"OA Gonzales-Huisa","year":"2023","journal-title":"Prosthesis"},{"key":"pone.0358882.ref017","doi-asserted-by":"crossref","first-page":"1079","DOI":"10.1109\/TNSRE.2021.3086843","article-title":"IMU-based deep neural networks: Prediction of locomotor and transition intentions of an osseointegrated transfemoral amputee","volume":"29","author":"J Bruinsma","year":"2021","journal-title":"IEEE Trans Neural Syst Rehabil Eng"},{"issue":"1","key":"pone.0358882.ref018","doi-asserted-by":"crossref","first-page":"279","DOI":"10.1109\/TMECH.2023.3278315","article-title":"Adaptive locomotion transition recognition with wearable sensors for lower limb robotic prosthesis","volume":"29","author":"E Zheng","year":"2024","journal-title":"IEEE\/ASME Trans Mechatron"},{"key":"pone.0358882.ref019","doi-asserted-by":"crossref","unstructured":"Jamieson A. Development of a clinically-targeted human activity recognition system to aid the prosthetic rehabilitation of individuals with lower limb amputation in free living conditions. 2021. http:\/\/purl.org\/coar\/resource_type\/c_db06","DOI":"10.3390\/s21248377"},{"issue":"21","key":"pone.0358882.ref020","doi-asserted-by":"crossref","first-page":"6345","DOI":"10.3390\/s20216345","article-title":"Machine learning approaches for activity recognition and\/or activity prediction in locomotion assistive devices-a systematic review","volume":"20","author":"F Labarri\u00e8re","year":"2020","journal-title":"Sensors (Basel)"},{"issue":"1","key":"pone.0358882.ref021","doi-asserted-by":"crossref","first-page":"224","DOI":"10.1007\/s42235-023-00435-w","article-title":"A comparison of four neural networks algorithms on locomotion intention recognition of lower limb exoskeleton based on multi-source information","volume":"21","author":"D Wang","year":"2023","journal-title":"J Bionic Eng"},{"issue":"22","key":"pone.0358882.ref022","doi-asserted-by":"crossref","first-page":"27941","DOI":"10.1109\/JSEN.2023.3320274","article-title":"Early prediction of lower limb prostheses locomotion mode transition based on terrain recognition","volume":"23","author":"S Luo","year":"2023","journal-title":"IEEE Sensors J"},{"key":"pone.0358882.ref023","doi-asserted-by":"crossref","unstructured":"Yin S, Chen X, Ma T, Wang Y, Guo Y, Leng Y. Environmental Recognition and Multimode Continuous-Phase Control for a Powered Transfemoral Prosthesis. In: 2023 IEEE International Conference on Development and Learning (ICDL); 2023 9\u201311 Nov. 2023. https:\/\/ieeexplore.ieee.org\/document\/1234567","DOI":"10.1109\/ICDL55364.2023.10364513"},{"issue":"6","key":"pone.0358882.ref024","doi-asserted-by":"crossref","first-page":"3980","DOI":"10.1109\/TSMC.2021.3084036","article-title":"Efficient environmental context prediction for lower limb prostheses","volume":"52","author":"B Zhong","year":"2022","journal-title":"IEEE Trans Syst Man Cybern, Syst"},{"issue":"1","key":"pone.0358882.ref025","doi-asserted-by":"crossref","first-page":"281","DOI":"10.1109\/TMRB.2024.3349624","article-title":"Terrain-adaptive exoskeleton control with predictive gait mode recognition: A pilot study during level walking and stair ascent","volume":"6","author":"Y Qian","year":"2024","journal-title":"IEEE Trans Med Robot Bionics"},{"key":"pone.0358882.ref026","doi-asserted-by":"crossref","first-page":"2251","DOI":"10.1109\/TRO.2025.3539206","article-title":"Ambilateral activity recognition and continuous adaptation with a powered knee-ankle prosthesis","volume":"41","author":"S Cheng","year":"2025","journal-title":"IEEE Trans Robot"},{"issue":"8","key":"pone.0358882.ref027","doi-asserted-by":"crossref","first-page":"2638","DOI":"10.3390\/app10082638","article-title":"A smart terrain identification technique based on electromyography, ground reaction force, and machine learning for lower limb rehabilitation","volume":"10","author":"S Gao","year":"2020","journal-title":"Appl Sci"},{"issue":"3","key":"pone.0358882.ref028","doi-asserted-by":"crossref","first-page":"559","DOI":"10.1007\/s13534-024-00351-w","article-title":"A multimodal framework based on deep belief network for human locomotion intent prediction","volume":"14","author":"J Li","year":"2024","journal-title":"Biomed Eng Lett"},{"issue":"9","key":"pone.0358882.ref029","doi-asserted-by":"crossref","first-page":"1082","DOI":"10.3390\/bioengineering10091082","article-title":"Multivariate CNN model for human locomotion activity recognition with a wearable exoskeleton robot","volume":"10","author":"C-S Son","year":"2023","journal-title":"Bioengineering (Basel)"},{"issue":"11","key":"pone.0358882.ref030","doi-asserted-by":"crossref","first-page":"7955","DOI":"10.1109\/JBHI.2025.3583319","article-title":"Enhancing Locomotion-Mode Recognition and Transition Prediction With (Bio)Mechanical Sensor Fusion for Intelligent Prosthetic Knees","volume":"29","author":"X Wang","year":"2025","journal-title":"IEEE J Biomed Health Inform"},{"issue":"1","key":"pone.0358882.ref031","doi-asserted-by":"crossref","first-page":"142","DOI":"10.1186\/s12984-025-01672-2","article-title":"Real-time locomotion mode detection in individuals with transfemoral amputation and osseointegration","volume":"22","author":"B Ahkami","year":"2025","journal-title":"J Neuroeng Rehabil"},{"issue":"6","key":"pone.0358882.ref032","doi-asserted-by":"crossref","first-page":"1334","DOI":"10.1109\/TNSRE.2020.2987155","article-title":"IMU-based locomotion mode identification for transtibial prostheses, orthoses, and exoskeletons","volume":"28","author":"F Gao","year":"2020","journal-title":"IEEE Trans Neural Syst Rehabil Eng"},{"key":"pone.0358882.ref033","doi-asserted-by":"crossref","unstructured":"Mai J, Xu D, Li H, Zhang S, Tan J, Wang Q. Implementing a SoC-FPGA Based Acceleration System for On-Board SVM Training for Robotic Transtibial Prostheses. In: 2018 IEEE International Conference on Real-time Computing and Robotics (RCAR); 2018 1\u20135 Aug. 2018.","DOI":"10.1109\/RCAR.2018.8621732"},{"issue":"2","key":"pone.0358882.ref034","doi-asserted-by":"crossref","DOI":"10.1371\/journal.pone.0192950","article-title":"Identifying classifier input signals to predict a cross-slope during transtibial amputee walking","volume":"13","author":"CE Shell","year":"2018","journal-title":"PLoS One"},{"key":"pone.0358882.ref035","doi-asserted-by":"crossref","first-page":"103414","DOI":"10.1016\/j.apergo.2021.103414","article-title":"Applying deep neural networks and inertial measurement unit in recognizing irregular walking differences in the real world","volume":"96","author":"B Hu","year":"2021","journal-title":"Appl Ergon"},{"key":"pone.0358882.ref036","doi-asserted-by":"crossref","unstructured":"Lu Z, Narayan A, Yu H. A Deep Learning Based End-to-End Locomotion Mode Detection Method for Lower Limb Wearable Robot Control. In: 2020 IEEE\/RSJ International Conference on Intelligent Robots and Systems (IROS), 2020. 4091\u20137. https:\/\/doi.org\/10.1109\/iros45743.2020.9341183","DOI":"10.1109\/IROS45743.2020.9341183"},{"issue":"12","key":"pone.0358882.ref037","doi-asserted-by":"crossref","first-page":"1078","DOI":"10.1177\/1545968320962483","article-title":"Improving accelerometry-based measurement of functional use of the upper extremity after stroke: machine learning versus counts threshold method","volume":"34","author":"PS Lum","year":"2020","journal-title":"Neurorehabil Neural Repair"},{"key":"pone.0358882.ref038","doi-asserted-by":"crossref","unstructured":"Choudhary R, Gianey HK. Comprehensive review on supervised machine learning algorithms. In: International Conference on Machine Learning and Data Science (MLDS); 2017 14\u201315 Dec. 2017.","DOI":"10.1109\/MLDS.2017.11"},{"issue":"4","key":"pone.0358882.ref039","article-title":"Divergent component of motion-based gait intention detection method using motion information from single leg","volume":"107","author":"H-W Oh","year":"2023","journal-title":"J Intell Robot Syst"},{"issue":"18","key":"pone.0358882.ref040","first-page":"8943","article-title":"Classifying the walking pattern of humans on different surfaces using convolutional features and shallow machine learning classifiers","volume":"238","author":"P Chauhan","year":"2024","journal-title":"Proceedings of the Institution of Mechanical Engineers, Part C: J Mech En Sci"},{"issue":"11","key":"pone.0358882.ref041","doi-asserted-by":"crossref","first-page":"5483","DOI":"10.3390\/app12115483","article-title":"FSM-HSVM-based locomotion mode recognition for exoskeleton robot","volume":"12","author":"Z Qi","year":"2022","journal-title":"Appl Sci"},{"issue":"22","key":"pone.0358882.ref042","doi-asserted-by":"crossref","first-page":"8871","DOI":"10.3390\/s22228871","article-title":"IMU-based classification of locomotion modes, transitions, and gait phases with convolutional recurrent neural networks","volume":"22","author":"D Marcos Mazon","year":"2022","journal-title":"Sensors (Basel)"},{"issue":"8","key":"pone.0358882.ref043","doi-asserted-by":"crossref","first-page":"2785","DOI":"10.3390\/s21082785","article-title":"Locomotion mode recognition algorithm based on Gaussian mixture model using IMU sensors","volume":"21","author":"D Shin","year":"2021","journal-title":"Sensors"},{"issue":"1","key":"pone.0358882.ref044","doi-asserted-by":"crossref","first-page":"282","DOI":"10.1038\/s41597-021-01057-9","article-title":"Lower-limb kinematics and kinetics during continuously varying human locomotion","volume":"8","author":"E Reznick","year":"2021","journal-title":"Sci Data"},{"issue":"24","key":"pone.0358882.ref045","doi-asserted-by":"crossref","first-page":"8377","DOI":"10.3390\/s21248377","article-title":"Human activity recognition of individuals with lower limb amputation in free-living conditions: A pilot study","volume":"21","author":"A Jamieson","year":"2021","journal-title":"Sensors (Basel)"},{"issue":"1","key":"pone.0358882.ref046","doi-asserted-by":"crossref","DOI":"10.1093\/geroni\/igz008","article-title":"Smartwatches can detect walker and cane use in older adults","volume":"3","author":"SA Antos","year":"2019","journal-title":"Innov Aging"},{"key":"pone.0358882.ref047","doi-asserted-by":"crossref","DOI":"10.1007\/978-3-319-41264-1_46","volume-title":"Walking aid identification using wearables","author":"T Moder","year":"2016"},{"issue":"3","key":"pone.0358882.ref048","first-page":"100209","article-title":"Effects of walking with a cane on frontal plane hip joint loading in patients with late-stage unilateral hip osteoarthritis","volume":"4","author":"M Tazawa","year":"2022","journal-title":"Arch Rehabil Res Clin Transl"},{"key":"pone.0358882.ref049","doi-asserted-by":"crossref","first-page":"111426","DOI":"10.1016\/j.jbiomech.2022.111426","article-title":"Effects of cane use on the kinematic and kinetic of lower-extremity joints in inexperienced users","volume":"146","author":"Y Zhang","year":"2023","journal-title":"J Biomech"},{"issue":"1","key":"pone.0358882.ref050","doi-asserted-by":"crossref","first-page":"146","DOI":"10.1016\/j.gaitpost.2008.06.001","article-title":"Gait and kinematics of the trunk and lower extremities in stair ascent using quadricane in healthy subjects","volume":"29","author":"B-J Hsue","year":"2009","journal-title":"Gait Posture"},{"key":"pone.0358882.ref051","unstructured":"HM-Goverment. Protection from falling, collision and impact: Approved Document K. 2013."},{"issue":"8","key":"pone.0358882.ref052","doi-asserted-by":"crossref","first-page":"10173","DOI":"10.1109\/TPAMI.2023.3250241","article-title":"Normalization techniques in training DNNs: methodology, analysis and application","volume":"45","author":"L Huang","year":"2023","journal-title":"IEEE Trans Pattern Anal Mach Intell"},{"issue":"17","key":"pone.0358882.ref053","doi-asserted-by":"crossref","first-page":"5571","DOI":"10.3390\/s24175571","article-title":"Machine learning based abnormal gait classification with IMU considering joint impairment","volume":"24","author":"S Hwang","year":"2024","journal-title":"Sensors"},{"issue":"2","key":"pone.0358882.ref054","doi-asserted-by":"crossref","first-page":"77","DOI":"10.1191\/1478088706qp063oa","article-title":"Using thematic analysis in psychology","volume":"3","author":"V Braun","year":"2006","journal-title":"Qualitative Res Psychology"},{"issue":"3","key":"pone.0358882.ref055","doi-asserted-by":"crossref","first-page":"394","DOI":"10.1016\/j.apmr.2004.03.026","article-title":"Partial weight-bearing gait using conventional assistive devices","volume":"86","author":"JW Youdas","year":"2005","journal-title":"Arch Phys Med Rehabil"},{"issue":"1","key":"pone.0358882.ref056","doi-asserted-by":"crossref","first-page":"734","DOI":"10.1186\/s12877-023-04443-7","article-title":"Evaluation of a novel biomechanics-informed walking frame, developed through a Knowledge Transfer Partnership between biomechanists and design engineers","volume":"23","author":"SB Thies","year":"2023","journal-title":"BMC Geriatr"}],"container-title":["PLOS One"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0358882","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,10,9]],"date-time":"2026-10-09T17:25:39Z","timestamp":1791566739000},"score":1,"resource":{"primary":{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0358882"}},"subtitle":[],"editor":[{"given":"Gary","family":"Guerra","sequence":"first","affiliation":[{"name":"St Mary's University","place":["UNITED STATES OF AMERICA"]}],"role":[{"vocabulary":"crossref","role":"editor"}]}],"short-title":[],"issued":{"date-parts":[[2026,10,9]]},"references-count":56,"journal-issue":{"issue":"10","published-online":{"date-parts":[[2026,10,9]]}},"URL":"https:\/\/doi.org\/10.1371\/journal.pone.0358882","relation":{},"ISSN":["1932-6203"],"issn-type":[{"value":"1932-6203","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,10,9]]}}}