{"title":"Identification of footstrike pattern using accelerometry and machine learning","authors":"","doi":"10.1016/j.jbiomech.2024.112255","DOIUrl":null,"url":null,"abstract":"<div><p>Recent reports have suggested that there may be a relationship between footstrike pattern and overuse injury incidence and type. With the recent increase in wearable sensors, it is important to identify paradigms where the footstrike pattern can be detected in real-time from minimal data. Machine learning was used to classify tibial acceleration data into three distinct footstrike patterns: rearfoot, midfoot, or forefoot. Tibial accelerometry data were collected during treadmill running from 58 participants who each ran with rearfoot, midfoot, and forefoot strike patterns. These data were used as inputs into an artificial neural network classifier. Models were created by using three distinct acceleration data sets, using the first 100%, 75%, and 40% of stance phase. All models were able to predict the footstrike pattern with up to 89.9% average accuracy. The highest error was associated with the identification of the midfoot versus forefoot strike pattern. This technique required no pre-selection of features or filtering of the data and may be easily incorporated into a wearable device to aid with real-time footstrike pattern detection.</p></div>","PeriodicalId":15168,"journal":{"name":"Journal of biomechanics","volume":null,"pages":null},"PeriodicalIF":2.4000,"publicationDate":"2024-08-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Journal of biomechanics","FirstCategoryId":"5","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S0021929024003336","RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"BIOPHYSICS","Score":null,"Total":0}
引用次数: 0
Abstract
Recent reports have suggested that there may be a relationship between footstrike pattern and overuse injury incidence and type. With the recent increase in wearable sensors, it is important to identify paradigms where the footstrike pattern can be detected in real-time from minimal data. Machine learning was used to classify tibial acceleration data into three distinct footstrike patterns: rearfoot, midfoot, or forefoot. Tibial accelerometry data were collected during treadmill running from 58 participants who each ran with rearfoot, midfoot, and forefoot strike patterns. These data were used as inputs into an artificial neural network classifier. Models were created by using three distinct acceleration data sets, using the first 100%, 75%, and 40% of stance phase. All models were able to predict the footstrike pattern with up to 89.9% average accuracy. The highest error was associated with the identification of the midfoot versus forefoot strike pattern. This technique required no pre-selection of features or filtering of the data and may be easily incorporated into a wearable device to aid with real-time footstrike pattern detection.
期刊介绍:
The Journal of Biomechanics publishes reports of original and substantial findings using the principles of mechanics to explore biological problems. Analytical, as well as experimental papers may be submitted, and the journal accepts original articles, surveys and perspective articles (usually by Editorial invitation only), book reviews and letters to the Editor. The criteria for acceptance of manuscripts include excellence, novelty, significance, clarity, conciseness and interest to the readership.
Papers published in the journal may cover a wide range of topics in biomechanics, including, but not limited to:
-Fundamental Topics - Biomechanics of the musculoskeletal, cardiovascular, and respiratory systems, mechanics of hard and soft tissues, biofluid mechanics, mechanics of prostheses and implant-tissue interfaces, mechanics of cells.
-Cardiovascular and Respiratory Biomechanics - Mechanics of blood-flow, air-flow, mechanics of the soft tissues, flow-tissue or flow-prosthesis interactions.
-Cell Biomechanics - Biomechanic analyses of cells, membranes and sub-cellular structures; the relationship of the mechanical environment to cell and tissue response.
-Dental Biomechanics - Design and analysis of dental tissues and prostheses, mechanics of chewing.
-Functional Tissue Engineering - The role of biomechanical factors in engineered tissue replacements and regenerative medicine.
-Injury Biomechanics - Mechanics of impact and trauma, dynamics of man-machine interaction.
-Molecular Biomechanics - Mechanical analyses of biomolecules.
-Orthopedic Biomechanics - Mechanics of fracture and fracture fixation, mechanics of implants and implant fixation, mechanics of bones and joints, wear of natural and artificial joints.
-Rehabilitation Biomechanics - Analyses of gait, mechanics of prosthetics and orthotics.
-Sports Biomechanics - Mechanical analyses of sports performance.