{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,28]],"date-time":"2026-04-28T19:37:19Z","timestamp":1777405039402,"version":"3.51.4"},"reference-count":25,"publisher":"MDPI AG","issue":"16","license":[{"start":{"date-parts":[[2023,8,14]],"date-time":"2023-08-14T00:00:00Z","timestamp":1691971200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"Karel de Grote University of Applied Sciences and Arts through funding by the Flemish government"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Sensors"],"abstract":"<jats:p>Introduction. Spatiotemporal gait parameters, e.g., gait stride length, are measurements that are classically derived from instrumented gait analysis. Today, different solutions are available for gait assessment outside the laboratory, specifically for spatiotemporal gait parameters. Such solutions are wearable devices that comprise an inertial measurement unit (IMU) sensor and a microcontroller (MCU). However, these existing wearable devices are resource-constrained. They contain a processing unit with limited processing and memory capabilities which limit the use of machine learning to estimate spatiotemporal gait parameters directly on the device. The solution for this limitation is embedded machine learning or tiny machine learning (tinyML). This study aims to create a machine-learning model for gait stride length estimation deployable on a microcontroller. Materials and Method. Starting from a dataset consisting of 4467 gait strides from 15 healthy people, measured by IMU sensor, and using state-of-the-art machine learning frameworks and machine learning operations (MLOps) tools, a multilayer 1D convolutional float32 and int8 model for gait stride length estimation was developed. Results. The developed float32 model demonstrated a mean accuracy and precision of 0.23 \u00b1 4.3 cm, and the int8 model demonstrated a mean accuracy and precision of 0.07 \u00b1 4.3 cm. The memory usage for the float32 model was 284.5 kB flash and 31.9 kB RAM. The int8 model memory usage was 91.6 kB flash and 13.6 kB RAM. Both models were able to be deployed on a Cortex-M4F 64 MHz microcontroller with 1 MB flash memory and 256 kB RAM. Conclusions. This study shows that estimating gait stride length directly on a microcontroller is feasible and demonstrates the potential of embedded machine learning, or tinyML, in designing wearable sensor devices for gait analysis.<\/jats:p>","DOI":"10.3390\/s23167166","type":"journal-article","created":{"date-parts":[[2023,8,14]],"date-time":"2023-08-14T11:07:10Z","timestamp":1692011230000},"page":"7166","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":14,"title":["Gait Stride Length Estimation Using Embedded Machine Learning"],"prefix":"10.3390","volume":"23","author":[{"given":"Joeri R.","family":"Verbiest","sequence":"first","affiliation":[{"name":"Department of Sciences and Technology, Karel de Grote (KdG) University of Applied Sciences and Arts, 2660 Antwerp, Belgium"},{"name":"REVAL Rehabilitation Research Center, Faculty of Rehabilitation Sciences, Hasselt University, 3590 Diepenbeek, Belgium"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-7729-4700","authenticated-orcid":false,"given":"Bruno","family":"Bonnech\u00e8re","sequence":"additional","affiliation":[{"name":"REVAL Rehabilitation Research Center, Faculty of Rehabilitation Sciences, Hasselt University, 3590 Diepenbeek, Belgium"},{"name":"Technology-Supported and Data-Driven Rehabilitation, Data Science Institute, Hasselt University, 3590 Diepenbeek, Belgium"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-8193-5016","authenticated-orcid":false,"given":"Wim","family":"Saeys","sequence":"additional","affiliation":[{"name":"Department of Rehabilitation Sciences and Physiotherapy, MOVANT, Faculty of Medicine and Health Sciences, University of Antwerp, Wilrijk, 2610 Antwerp, Belgium"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-8948-8927","authenticated-orcid":false,"given":"Patricia","family":"Van de Walle","sequence":"additional","affiliation":[{"name":"Department of Rehabilitation Sciences and Physiotherapy, MOVANT, Faculty of Medicine and Health Sciences, University of Antwerp, Wilrijk, 2610 Antwerp, Belgium"},{"name":"Clinical Gait Analysis Laboratory Antwerp, Heder, Ekeren, 2180 Antwerp, Belgium"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Steven","family":"Truijen","sequence":"additional","affiliation":[{"name":"Department of Rehabilitation Sciences and Physiotherapy, MOVANT, Faculty of Medicine and Health Sciences, University of Antwerp, Wilrijk, 2610 Antwerp, Belgium"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-2218-187X","authenticated-orcid":false,"given":"Pieter","family":"Meyns","sequence":"additional","affiliation":[{"name":"REVAL Rehabilitation Research Center, Faculty of Rehabilitation Sciences, Hasselt University, 3590 Diepenbeek, Belgium"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"1968","published-online":{"date-parts":[[2023,8,14]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","first-page":"274","DOI":"10.1016\/j.gaitpost.2020.05.031","article-title":"Clinical Efficacy of instrument gait analysis: Systematic review 2020 update","volume":"80","author":"Tishya","year":"2020","journal-title":"Gait Posture"},{"key":"ref_2","unstructured":"(2023, April 10). Physilog Digital Motion Analytics Platform. Available online: https:\/\/www.gaitup.com\/."},{"key":"ref_3","unstructured":"Alcala, E.R.D., Voerman, J.A., Konrath, J.M., and Vydhyanathan, A. (2023, April 10). Xsens DOT Wearable Sensor Platform White Paper. Available online: https:\/\/www.movella.com\/hubfs\/Downloads\/Whitepapers\/Xsens%20DOT%20WhitePaper.pdf."},{"key":"ref_4","doi-asserted-by":"crossref","first-page":"167830","DOI":"10.1109\/ACCESS.2020.3022818","article-title":"Latest Research Trends in Gait Analysis Using Wearable Sensors and Machine Learning: A Systematic Review","volume":"8","author":"Saboor","year":"2020","journal-title":"IEEE Access"},{"key":"ref_5","unstructured":"(2023, February 06). tinyML Foundation. Available online: https:\/\/www.tinyml.org."},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"98450","DOI":"10.1109\/ACCESS.2022.3206782","article-title":"Embedded Machine Learning Using Microcontrollers in Wearable and Ambulatory Systems for Health and Care Applications: A Review","volume":"10","author":"Diab","year":"2022","journal-title":"IEEE Access"},{"key":"ref_7","doi-asserted-by":"crossref","unstructured":"McGinnes, R.S., Mahadevan, N., Moon, Y., Seagers, K., Sheth, N., Wright, J.A., DiCristofaro, S., Silva, I., Jortberg, E., and Ceruolo, M. (2017). A machine learning approach for gait estimation using skin-mounted wearable sensors: From healthy controls to individuals with multiple sclerosis. PLoS ONE, 12.","DOI":"10.1371\/journal.pone.0178366"},{"key":"ref_8","doi-asserted-by":"crossref","first-page":"7380","DOI":"10.1109\/JSEN.2021.3049523","article-title":"IMU Based Deep Stride Length Estimation with Self-Suppervised Learning","volume":"21","author":"Sui","year":"2021","journal-title":"IEEE Sens. J."},{"key":"ref_9","doi-asserted-by":"crossref","first-page":"85","DOI":"10.1109\/JBHI.2016.2636456","article-title":"Sensor-Based Gait Parameter Extraction with Deep Convolutional Neural Networks","volume":"21","author":"Hannink","year":"2017","journal-title":"IEEE J. Biomed. Health Inf."},{"key":"ref_10","doi-asserted-by":"crossref","first-page":"354","DOI":"10.1109\/JBHI.2017.2679486","article-title":"Mobile Stride Length Estimation with Deep Convolutional Neural Networks","volume":"22","author":"Hannink","year":"2018","journal-title":"IEEE J. Biomed. Health Inf."},{"key":"ref_11","doi-asserted-by":"crossref","unstructured":"Zrenner, M., Gradl, S., Jensen, U., Ulrich, M., and Eskofier, B.M. (2018). Comparison of Different Algorithms for Calculating Velocity and Stride Length in Running Using Inertial Measurement Units. Sensors, 18.","DOI":"10.3390\/s18124194"},{"key":"ref_12","doi-asserted-by":"crossref","unstructured":"Trautmann, J., Zhou, L., Brahms, C.M., Tunca, C., Ersoy, C., Granacher, U., and Arnrich, B. (2021). TRIPOD\u2014A treadmill Walking Dataset with IMU, Pressure-Distribution and Photoelectric Data for Gait Analysis. Data, 6.","DOI":"10.3390\/data6090095"},{"key":"ref_13","unstructured":"(2023, February 06). Python. Available online: https:\/\/www.python.org."},{"key":"ref_14","unstructured":"(2023, February 06). SciPy. Available online: https:\/\/docs.scipy.org\/doc\/scipy-1.9.3\/."},{"key":"ref_15","doi-asserted-by":"crossref","first-page":"1500","DOI":"10.1109\/TNSRE.2016.2636367","article-title":"A Real-time Gait Event Detection for Lower Limb Prosthesis Control and Evaluation","volume":"25","author":"Maqbool","year":"2017","journal-title":"IEEE Trans. Neural Syst. Rehabil. Eng."},{"key":"ref_16","unstructured":"(2023, February 06). Edge Impulse Data Acquisition Format. Available online: https:\/\/docs.edgeimpulse.com\/reference\/data-ingestion\/data-acquisition-format."},{"key":"ref_17","unstructured":"Biewald, L. (2023, February 06). Experiment Tracking with Weights and Biases, Weights & Biases. Available online: http:\/\/wandb.com\/."},{"key":"ref_18","unstructured":"(2023, February 06). Keras. Available online: https:\/\/keras.io\/."},{"key":"ref_19","unstructured":"Hymel, S., Banbury, C., Situnayake, D., Elium, A., Ward, C., Kelcey, M., Baaijens, M., Majchrzycki, M., Plunkett, J., and Tishler, D. (2022). Edge Impulse: An MLOps Platform for Tiny Machine Learning. arXiv, Available online: https:\/\/arxiv.org\/abs\/2212.03332."},{"key":"ref_20","unstructured":"(2023, February 06). Post-Training Quantisation. Available online: https:\/\/www.tensorflow.org\/lite\/performance\/post_training_quantization."},{"key":"ref_21","doi-asserted-by":"crossref","first-page":"931","DOI":"10.1016\/j.ijnurstu.2009.10.001","article-title":"Statistical methods for assessing agreement between two methods of clinical measurements","volume":"47","author":"Bland","year":"2010","journal-title":"Int. J. Nurs. Stud."},{"key":"ref_22","doi-asserted-by":"crossref","first-page":"055012","DOI":"10.1088\/1361-6579\/ab86d6","article-title":"A comprehensive guideline for Bland-Altman and intra class correlation calculations to properly compare two methods of measurements and interpret findings","volume":"41","author":"Haghayegh","year":"2020","journal-title":"Physiol. Meas."},{"key":"ref_23","unstructured":"(2023, February 06). PlatformIO. Available online: https:\/\/platformio.org."},{"key":"ref_24","unstructured":"(2023, February 06). SparkFun MicroMod nRF52840 Processor Board. Available online: https:\/\/www.sparkfun.com\/products\/16984."},{"key":"ref_25","unstructured":"(2023, February 06). SparkFun MicroMod Data Logging Carrier Board. Available online: https:\/\/www.sparkfun.com\/products\/16829."}],"container-title":["Sensors"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/1424-8220\/23\/16\/7166\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,10]],"date-time":"2025-10-10T20:33:32Z","timestamp":1760128412000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/1424-8220\/23\/16\/7166"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2023,8,14]]},"references-count":25,"journal-issue":{"issue":"16","published-online":{"date-parts":[[2023,8]]}},"alternative-id":["s23167166"],"URL":"https:\/\/doi.org\/10.3390\/s23167166","relation":{},"ISSN":["1424-8220"],"issn-type":[{"value":"1424-8220","type":"electronic"}],"subject":[],"published":{"date-parts":[[2023,8,14]]}}}