{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,19]],"date-time":"2026-09-19T00:18:49Z","timestamp":1789777129280,"version":"4.0.1"},"reference-count":55,"publisher":"MDPI AG","issue":"11","license":[{"start":{"date-parts":[[2026,5,24]],"date-time":"2026-05-24T00:00:00Z","timestamp":1779580800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Sensors"],"abstract":"<jats:p>Falls are a major contributor to morbidity and mortality among older adults, and timely fall detection can help reduce the severity of fall-related outcomes. Wearable inertial measurement unit (IMU) sensors offer a promising solution for fall detection; however, many existing approaches rely on multiple sensing locations and computationally intensive models, which can limit their practicality for resource-constrained wearable devices. This study proposes a knowledge distillation framework for efficient wrist-based fall detection using the publicly available University of M\u00e1laga fall detection dataset (UMAFall), a benchmark dataset for human activity recognition and fall detection. Although UMAFall was not collected from older adults, it provides a useful public benchmark for evaluating IMU-based fall detection methods. Knowledge distillation was implemented using a teacher\u2013student framework, in which a high-capacity teacher model trained with IMU data from four body locations (waist, wrist, ankle, and chest) provided soft targets for guiding a compact wrist-only CNN student model. In a held-out test evaluation using Subjects 2 and 5, the teacher model achieved 97.6% accuracy and an F1 score of 96.7%, with approximately 1.3 million trainable parameters. The independently trained wrist-based CNN achieved 90.2% accuracy and an F1 score of 87.1%. After applying knowledge distillation, the student model improved to 95.1% accuracy and an F1 score of 93.3% while maintaining the same lightweight architecture. A supplementary leave-one-subject-out analysis showed slightly higher and more stable AUC for KD-CNN than the independently trained CNN (0.96 \u00b1 0.03 vs. 0.94 \u00b1 0.07). These findings suggest that knowledge distillation can improve wrist-only fall detection in this feasibility evaluation, but further validation using older adults and real-world smartwatch data is needed.<\/jats:p>","DOI":"10.3390\/s26113328","type":"journal-article","created":{"date-parts":[[2026,5,25]],"date-time":"2026-05-25T01:45:08Z","timestamp":1779673508000},"page":"3328","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":1,"title":["Efficient Fall Detection from Wrist-Worn IMU Signals via Knowledge Distillation: A Lightweight CNN Approach Using the UMAFall Dataset"],"prefix":"10.3390","volume":"26","author":[{"ORCID":"https:\/\/orcid.org\/0009-0003-4869-7791","authenticated-orcid":false,"given":"Ali","family":"Taheri","sequence":"first","affiliation":[{"name":"Department of Electrical and Computer Engineering, Texas A&M University, College Station, TX 77843, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-3579-1004","authenticated-orcid":false,"given":"Mina","family":"Salehi","sequence":"additional","affiliation":[{"name":"School of Nutrition and Public Health, College of Health, Oregon State University, Corvallis, OR 97331, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-9967-5846","authenticated-orcid":false,"given":"Jeong Ho","family":"Kim","sequence":"additional","affiliation":[{"name":"Department of Environmental and Occupational Health, School of Public Health, Texas A&M University, College Station, TX 77843, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1968","published-online":{"date-parts":[[2026,5,24]]},"reference":[{"key":"ref_1","unstructured":"CDC (2025, July 01). Older Adult Falls Data, Available online: https:\/\/www.cdc.gov\/falls\/data-research\/?CDC_AAref_Val=https:\/\/www.cdc.gov\/falls\/data\/index.html."},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"938","DOI":"10.15585\/mmwr.mm7235a1","article-title":"Nonfatal and Fatal Falls Among Adults Aged \u2265 65 Years\u2013United States, 2020\u20132021","volume":"72","author":"Kakara","year":"2023","journal-title":"MMWR Morb. Mortal. Wkly. Rep."},{"key":"ref_3","doi-asserted-by":"crossref","first-page":"69","DOI":"10.1007\/s43465-019-00037-x","article-title":"Falls in Older Adults are Serious","volume":"54","author":"Vaishya","year":"2020","journal-title":"Indian J. Orthop."},{"key":"ref_4","doi-asserted-by":"crossref","first-page":"816","DOI":"10.1080\/13607863.2013.805398","article-title":"Pathways from Fear of Falling to Quality of Life: The Mediating Effect of the Self-Concept of Health and Physical Independence","volume":"17","author":"Hsu","year":"2013","journal-title":"Aging Ment. Health"},{"key":"ref_5","doi-asserted-by":"crossref","first-page":"748","DOI":"10.1097\/TA.0b013e3181f6f5e5","article-title":"Societal Consequences of Falls in the Older Population: Injuries, Healthcare Costs, and Long-Term Reduced Quality of Life","volume":"71","author":"Hartholt","year":"2011","journal-title":"J. Trauma Acute Care Surg."},{"key":"ref_6","unstructured":"WHO (2021). Falls, World Health Organization. Available online: https:\/\/www.who.int\/news-room\/fact-sheets\/detail\/falls."},{"key":"ref_7","doi-asserted-by":"crossref","first-page":"693","DOI":"10.1111\/jgs.15304","article-title":"Medical Costs of Fatal and Nonfatal Falls in Older Adults","volume":"66","author":"Florence","year":"2018","journal-title":"J. Am. Geriatr. Soc."},{"key":"ref_8","doi-asserted-by":"crossref","unstructured":"Xing, L., Bao, Y., Wang, B., Shi, M., Wei, Y., Huang, X., Dai, Y., Shi, H., Gai, X., and Luo, Q. (2023). Falls Caused by Balance Disorders in the Elderly with Multiple Systems Involved: Pathogenic Mechanisms and Treatment Strategies. Front. Neurol., 14.","DOI":"10.3389\/fneur.2023.1128092"},{"key":"ref_9","doi-asserted-by":"crossref","first-page":"173","DOI":"10.1016\/j.cger.2019.01.008","article-title":"Balance Problems and Fall Risks in the Elderly","volume":"35","year":"2019","journal-title":"Clin. Geriatr. Med."},{"key":"ref_10","doi-asserted-by":"crossref","first-page":"21","DOI":"10.4103\/ijph.IJPH_332_17","article-title":"Prevalence, Risk Factors, Circumstances for Falls and Level of Functional Independence among Geriatric Population\u2013A Descriptive Study","volume":"63","author":"Pitchai","year":"2019","journal-title":"Indian J. Public Health"},{"key":"ref_11","doi-asserted-by":"crossref","unstructured":"Tanwar, R., Nandal, N., Zamani, M., and Manaf, A.A. (2022). Pathway of Trends and Technologies in Fall Detection: A Systematic Review. Healthcare, 10.","DOI":"10.3390\/healthcare10010172"},{"key":"ref_12","doi-asserted-by":"crossref","unstructured":"Torres-Guzman, R.A., Paulson, M.R., Avila, F.R., Maita, K., Garcia, J.P., Forte, A.J., and Maniaci, M.J. (2023). Smartphones and Threshold-Based Monitoring Methods Effectively Detect Falls Remotely: A Systematic Review. Sensors, 23.","DOI":"10.3390\/s23031323"},{"key":"ref_13","doi-asserted-by":"crossref","first-page":"77702","DOI":"10.1109\/ACCESS.2019.2922708","article-title":"Research of Fall Detection and Fall Prevention Technologies: A Systematic Review","volume":"7","author":"Ren","year":"2019","journal-title":"IEEE Access"},{"key":"ref_14","doi-asserted-by":"crossref","unstructured":"Gulline, H., Melder, A., Barker, A., Dickins, M., Smith, K., and Ayton, D. (2025). Opportunities to Digitally Enable Falls Prevention in Older Adults. Gerontologist, gnaf016.","DOI":"10.1093\/geront\/gnaf016"},{"key":"ref_15","doi-asserted-by":"crossref","first-page":"8982","DOI":"10.1007\/s10489-024-05645-1","article-title":"Deep Learning for Computer Vision Based Activity Recognition and Fall Detection of the Elderly: A Systematic Review","volume":"54","year":"2024","journal-title":"Appl. Intell."},{"key":"ref_16","unstructured":"Ponce, H., Mart\u00ednez-Villase\u00f1or, L., Brieva, J., and Moya-Albor, E. (2020). Classification of Daily Life Activities for Human Fall Detection: A Systematic Review of the Techniques and Approaches. Challenges and Trends in Multimodal Fall Detection for Healthcare, Springer International Publishing."},{"key":"ref_17","doi-asserted-by":"crossref","unstructured":"Guti\u00e9rrez, J., Rodr\u00edguez, V., and Martin, S. (2021). Comprehensive Review of Vision-Based Fall Detection Systems. Sensors, 21.","DOI":"10.3390\/s21030947"},{"key":"ref_18","doi-asserted-by":"crossref","unstructured":"Usmani, S., Saboor, A., Haris, M., Khan, M.A., and Park, H. (2021). Latest Research Trends in Fall Detection and Prevention Using Machine Learning: A Systematic Review. Sensors, 21.","DOI":"10.3390\/s21155134"},{"key":"ref_19","doi-asserted-by":"crossref","first-page":"103946","DOI":"10.1016\/j.ijmedinf.2019.08.006","article-title":"Fall Detection and Fall Risk Assessment in Older Person Using Wearable Sensors: A Systematic Review","volume":"130","author":"Bet","year":"2019","journal-title":"Int. J. Med. Inform."},{"key":"ref_20","doi-asserted-by":"crossref","unstructured":"Ruiz-Ruiz, L., Jimenez, A.R., Garcia-Villamil, G., and Seco, F. (2021). Detecting Fall Risk and Frailty in Elders with Inertial Motion Sensors: A Survey of Significant Gait Parameters. Sensors, 21.","DOI":"10.3390\/s21206918"},{"key":"ref_21","doi-asserted-by":"crossref","unstructured":"Wang, X., Cao, J., Zhao, Q., Chen, M., Luo, J., Wang, H., Yu, L., Tsui, K.L., and Zhao, Y. (2024). Identifying Sensors-Based Parameters Associated with Fall Risk in Community-Dwelling Older Adults: An Investigation and Interpretation of Discriminatory Parameters. BMC Geriatr., 24.","DOI":"10.1186\/s12877-024-04723-w"},{"key":"ref_22","doi-asserted-by":"crossref","unstructured":"Chen, M., Wang, H., Yu, L., Yeung, E.H.K., Luo, J., Tsui, K.L., and Zhao, Y. (2022). A Systematic Review of Wearable Sensor-Based Technologies for Fall Risk Assessment in Older Adults. Sensors, 22.","DOI":"10.3390\/s22186752"},{"key":"ref_23","doi-asserted-by":"crossref","unstructured":"Wang, S., and Wu, J. (2023). Patch-Transformer Network: A Wearable-Sensor-Based Fall Detection Method. Sensors, 23.","DOI":"10.3390\/s23146360"},{"key":"ref_24","doi-asserted-by":"crossref","first-page":"2735","DOI":"10.1007\/s12652-021-03248-z","article-title":"Sensor-Based Fall Detection Systems: A Review","volume":"13","author":"Nooruddin","year":"2022","journal-title":"J. Ambient Intell. Humaniz. Comput."},{"key":"ref_25","doi-asserted-by":"crossref","unstructured":"Teng, S., Kim, J.Y., Jeon, S., Gil, H.W., Lyu, J., Chung, E.H., Kim, K.S., and Nam, Y. (2024). Analyzing Optimal Wearable Motion Sensor Placement for Accurate Classification of Fall Directions. Sensors, 24.","DOI":"10.3390\/s24196432"},{"key":"ref_26","doi-asserted-by":"crossref","first-page":"166678","DOI":"10.1109\/ACCESS.2020.3022971","article-title":"Sensor Location Analysis and Minimal Deployment for Fall Detection System","volume":"8","author":"Ponce","year":"2020","journal-title":"IEEE Access"},{"key":"ref_27","doi-asserted-by":"crossref","unstructured":"Rucco, R., Sorriso, A., Liparoti, M., Ferraioli, G., Sorrentino, P., Ambrosanio, M., and Baselice, F. (2018). Type and Location of Wearable Sensors for Monitoring Falls during Static and Dynamic Tasks in Healthy Elderly: A Review. Sensors, 18.","DOI":"10.3390\/s18051613"},{"key":"ref_28","doi-asserted-by":"crossref","first-page":"19806","DOI":"10.3390\/s141019806","article-title":"Survey on Fall Detection and Fall Prevention Using Wearable and External Sensors","volume":"14","author":"Delahoz","year":"2014","journal-title":"Sensors"},{"key":"ref_29","doi-asserted-by":"crossref","first-page":"103178","DOI":"10.1016\/j.aei.2025.103178","article-title":"A Hybrid CNN-LSTM Model for Involuntary Fall Detection Using Wrist-Worn Sensors","volume":"65","author":"Hu","year":"2025","journal-title":"Adv. Eng. Inform."},{"key":"ref_30","doi-asserted-by":"crossref","first-page":"12900","DOI":"10.3390\/s140712900","article-title":"Automatic Fall Monitoring: A Review","volume":"14","author":"Pannurat","year":"2014","journal-title":"Sensors"},{"key":"ref_31","doi-asserted-by":"crossref","first-page":"2193","DOI":"10.1249\/MSS.0b013e31829736d6","article-title":"Activity Recognition Using a Single Accelerometer Placed at the Wrist or Ankle","volume":"45","author":"Mannini","year":"2013","journal-title":"Med. Sci. Sports Exerc."},{"key":"ref_32","doi-asserted-by":"crossref","unstructured":"Fula, V., and Moreno, P. (2024). Wrist-Based Fall Detection: Towards Generalization across Datasets. Sensors, 24.","DOI":"10.3390\/s24051679"},{"key":"ref_33","doi-asserted-by":"crossref","unstructured":"Haque, S.T., Debnath, M., Yasmin, A., Mahmud, T., and Ngu, A.H.H. (2024). Experimental Study of Long Short-Term Memory and Transformer Models for Fall Detection on Smartwatches. Sensors, 24.","DOI":"10.3390\/s24196235"},{"key":"ref_34","doi-asserted-by":"crossref","first-page":"114226","DOI":"10.1016\/j.eswa.2020.114226","article-title":"A Multimodal Approach Using Deep Learning for Fall Detection","volume":"168","author":"Ferreira","year":"2021","journal-title":"Expert Syst. Appl."},{"key":"ref_35","doi-asserted-by":"crossref","unstructured":"Mauldin, T.R., Canby, M.E., Metsis, V., Ngu, A.H.H., and Rivera, C.C. (2018). SmartFall: A Smartwatch-Based Fall Detection System Using Deep Learning. Sensors, 18.","DOI":"10.3390\/s18103363"},{"key":"ref_36","doi-asserted-by":"crossref","unstructured":"Maray, N., Ngu, A.H.N., Ni, J., Debnath, M., and Wang, L. (2023). Transfer Learning on Small Datasets for Improved Fall Detection. Sensors, 23.","DOI":"10.3390\/s23031105"},{"key":"ref_37","doi-asserted-by":"crossref","first-page":"42195","DOI":"10.1109\/JSEN.2024.3456577","article-title":"MECKD: Deep Learning-Based Fall Detection in Multilayer Mobile Edge Computing with Knowledge Distillation","volume":"24","author":"Mao","year":"2024","journal-title":"IEEE Sens. J."},{"key":"ref_38","doi-asserted-by":"crossref","unstructured":"Chi, T.-H., Liu, K.-C., Hsieh, C.-Y., Tsao, Y., and Chan, C.-T. (2023, January 4\u201310). Prefallkd: Pre-Impact Fall Detection via CNN-ViT Knowledge Distillation. Proceedings of the IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), Rhodes Island, Greece.","DOI":"10.1109\/ICASSP49357.2023.10094979"},{"key":"ref_39","doi-asserted-by":"crossref","unstructured":"Bello, H., Gei\u00dfler, D., Suh, S., Zhou, B., and Lukowicz, P. (2024). TSAK: Two-Stage Semantic-Aware Knowledge Distillation for Efficient Wearable Modality and Model Optimization in Manufacturing Lines. arXiv.","DOI":"10.1007\/978-3-031-78389-0_14"},{"key":"ref_40","doi-asserted-by":"crossref","unstructured":"Casilari, E., Santoyo-Ram\u00f3n, J.-A., and Cano-Garc\u00eda, J.-M. (2017). Analysis of Public Datasets for Wearable Fall Detection Systems. Sensors, 17.","DOI":"10.3390\/s17071513"},{"key":"ref_41","doi-asserted-by":"crossref","unstructured":"Casilari, E., Santoyo-Ram\u00f3n, J.A., and Cano-Garc\u00eda, J.M. (2016). Analysis of a Smartphone-Based Architecture with Multiple Mobility Sensors for Fall Detection. PLoS ONE, 11.","DOI":"10.1371\/journal.pone.0168069"},{"key":"ref_42","first-page":"1283","article-title":"Tripping-Induced Fall Detection Using Dual-Torso Inertial Measurement Units: An Experimental Evaluation with Long Short-Term Memory","volume":"38","author":"Zhang","year":"2026","journal-title":"Sens. Mater."},{"key":"ref_43","doi-asserted-by":"crossref","first-page":"186","DOI":"10.1007\/s10791-025-09700-4","article-title":"TCN-BiSRU-V2 Fall Detection Model with Performance Evaluation and Comparative Analysis","volume":"28","author":"Ahirwar","year":"2025","journal-title":"Discov. Comput."},{"key":"ref_44","doi-asserted-by":"crossref","first-page":"20991","DOI":"10.1007\/s11042-021-10600-0","article-title":"Sensor-Data Augmentation for Human Activity Recognition with Time-Warping and Data Masking","volume":"80","author":"Jeong","year":"2021","journal-title":"Multimed. Tools Appl."},{"key":"ref_45","doi-asserted-by":"crossref","unstructured":"Tu, Y.-C., Lin, C.-Y., Liu, C.-P., and Chan, C.-T. (2025). Performance Analysis of Data Augmentation Approaches for Improving Wrist-Based Fall Detection System. Sensors, 25.","DOI":"10.3390\/s25072168"},{"key":"ref_46","unstructured":"Chi, H., Chi, S., Chan, S., and Ramani, K. (June, January 29). Pose Relation Transformer Refine Occlusions for Human Pose Estimation. Proceedings of the IEEE International Conference on Robotics and Automation (ICRA), London, UK."},{"key":"ref_47","doi-asserted-by":"crossref","unstructured":"Arslan, M., Guzel, M., Demirci, M., and Ozdemir, S. (2019, January 11\u201315). SMOTE and Gaussian Noise Based Sensor Data Augmentation. Proceedings of the 4th International Conference on Computer Science and Engineering (UBMK), Samsun, Turkey.","DOI":"10.1109\/UBMK.2019.8907003"},{"key":"ref_48","doi-asserted-by":"crossref","first-page":"2267","DOI":"10.1007\/s00779-023-01776-3","article-title":"Lightweight Transformers for Human Activity Recognition on Mobile Devices","volume":"27","author":"Ek","year":"2023","journal-title":"Pers. Ubiquitous Comput."},{"key":"ref_49","doi-asserted-by":"crossref","unstructured":"Lin, T.-Y., Goyal, P., Girshick, R., He, K., and Doll\u00e1r, P. (2018). Focal Loss for Dense Object Detection. arXiv.","DOI":"10.1109\/ICCV.2017.324"},{"key":"ref_50","unstructured":"Loshchilov, I., and Hutter, F. (2019). Decoupled Weight Decay Regularization. arXiv."},{"key":"ref_51","unstructured":"Freire, P., Srivallapanondh, S., Napoli, A., Prilepsky, J.E., and Turitsyn, S.K. (2022). Computational Complexity Evaluation of Neural Network Applications in Signal Processing. arXiv."},{"key":"ref_52","doi-asserted-by":"crossref","first-page":"381","DOI":"10.1007\/s10044-020-00901-9","article-title":"Spatio-Temporal Adversarial Learning for Detecting Unseen Falls","volume":"24","author":"Khan","year":"2021","journal-title":"Pattern Anal. Appl."},{"key":"ref_53","doi-asserted-by":"crossref","unstructured":"Zhang, J., Li, J., and Wang, W. (2021). A Class-Imbalanced Deep Learning Fall Detection Algorithm Using Wearable Sensors. Sensors, 21.","DOI":"10.3390\/s21196511"},{"key":"ref_54","doi-asserted-by":"crossref","first-page":"110945","DOI":"10.1016\/j.measurement.2022.110945","article-title":"A Study of the Influence of the Sensor Sampling Frequency on the Performance of Wearable Fall Detectors","volume":"193","author":"Casilari","year":"2022","journal-title":"Measurement"},{"key":"ref_55","doi-asserted-by":"crossref","unstructured":"Mondal, R., and Ghosal, P. (2023). Recall-Driven Precision Refinement: Unveiling Accurate Fall Detection Using LSTM. Proceedings of the IFIP International Internet of Things Conference, Springer Nature.","DOI":"10.1007\/978-3-031-45882-8_6"}],"container-title":["Sensors"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/1424-8220\/26\/11\/3328\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,5,25]],"date-time":"2026-05-25T01:53:06Z","timestamp":1779673986000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/1424-8220\/26\/11\/3328"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,5,24]]},"references-count":55,"journal-issue":{"issue":"11","published-online":{"date-parts":[[2026,6]]}},"alternative-id":["s26113328"],"URL":"https:\/\/doi.org\/10.3390\/s26113328","relation":{},"ISSN":["1424-8220"],"issn-type":[{"value":"1424-8220","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,5,24]]}}}