Evaluating Sparse Inertial Measurement Unit Configurations for Inferring Treadmill Running Motion

Inertial measurement units (IMUs) are used to analyze running performance. While leveraging one sensor to estimate kinematic and kinetic variables is common, sparsity limits the number of digital biomarkers that can be evaluated. Shallow recurrent decoder networks (SHRED) can reconstruct a dense set...

Full description

Saved in:
Bibliographic Details
Main Authors: Mackenzie N. Pitts, Megan R. Ebers, Cristine E. Agresta, Katherine M. Steele
Format: Article
Language:English
Published: MDPI AG 2025-03-01
Series:Sensors
Subjects:
Online Access:https://www.mdpi.com/1424-8220/25/7/2105
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850213229169475584
author Mackenzie N. Pitts
Megan R. Ebers
Cristine E. Agresta
Katherine M. Steele
author_facet Mackenzie N. Pitts
Megan R. Ebers
Cristine E. Agresta
Katherine M. Steele
author_sort Mackenzie N. Pitts
collection DOAJ
description Inertial measurement units (IMUs) are used to analyze running performance. While leveraging one sensor to estimate kinematic and kinetic variables is common, sparsity limits the number of digital biomarkers that can be evaluated. Shallow recurrent decoder networks (SHRED) can reconstruct a dense set of time-series signals from a single input sensor and have been successful in human mobility applications, highlighting the potential for this algorithm to monitor running. We trained and tested subject-specific SHRED models of nine subjects running on a treadmill to map from one input sensor to the remaining three IMUs. We varied the type of input to reflect experimental parameters that are important in running studies—sensor location, sensor type, sampling rate, and running speed—and compared the error of inferred signals from each input type. Sensor location and type did not impact SHRED inference accuracy, while decreasing the sampling rate affected the accuracy of ankle measurements. All ankle acceleration inferences from these models remained below the minimal detectable change threshold of 12.0 m/s<sup>2</sup>. SHRED models trained and tested at multiple speeds did not accurately infer IMU measurements below this threshold. SHRED may broaden the scope of motion analysis by expanding datasets with fewer sensors.
format Article
id doaj-art-5de64fa139ef4c45aa37dbfeeb80883d
institution OA Journals
issn 1424-8220
language English
publishDate 2025-03-01
publisher MDPI AG
record_format Article
series Sensors
spelling doaj-art-5de64fa139ef4c45aa37dbfeeb80883d2025-08-20T02:09:11ZengMDPI AGSensors1424-82202025-03-01257210510.3390/s25072105Evaluating Sparse Inertial Measurement Unit Configurations for Inferring Treadmill Running MotionMackenzie N. Pitts0Megan R. Ebers1Cristine E. Agresta2Katherine M. Steele3Mechanical Engineering, University of Washington, Seattle, WA 98195, USAApplied Mathematics, University of Washington, Seattle, WA 98195, USARehabilitation Medicine, University of Washington, Seattle, WA 98195, USAMechanical Engineering, University of Washington, Seattle, WA 98195, USAInertial measurement units (IMUs) are used to analyze running performance. While leveraging one sensor to estimate kinematic and kinetic variables is common, sparsity limits the number of digital biomarkers that can be evaluated. Shallow recurrent decoder networks (SHRED) can reconstruct a dense set of time-series signals from a single input sensor and have been successful in human mobility applications, highlighting the potential for this algorithm to monitor running. We trained and tested subject-specific SHRED models of nine subjects running on a treadmill to map from one input sensor to the remaining three IMUs. We varied the type of input to reflect experimental parameters that are important in running studies—sensor location, sensor type, sampling rate, and running speed—and compared the error of inferred signals from each input type. Sensor location and type did not impact SHRED inference accuracy, while decreasing the sampling rate affected the accuracy of ankle measurements. All ankle acceleration inferences from these models remained below the minimal detectable change threshold of 12.0 m/s<sup>2</sup>. SHRED models trained and tested at multiple speeds did not accurately infer IMU measurements below this threshold. SHRED may broaden the scope of motion analysis by expanding datasets with fewer sensors.https://www.mdpi.com/1424-8220/25/7/2105IMUmachine learningrunningsparse sensingaccelerometersampling rate
spellingShingle Mackenzie N. Pitts
Megan R. Ebers
Cristine E. Agresta
Katherine M. Steele
Evaluating Sparse Inertial Measurement Unit Configurations for Inferring Treadmill Running Motion
Sensors
IMU
machine learning
running
sparse sensing
accelerometer
sampling rate
title Evaluating Sparse Inertial Measurement Unit Configurations for Inferring Treadmill Running Motion
title_full Evaluating Sparse Inertial Measurement Unit Configurations for Inferring Treadmill Running Motion
title_fullStr Evaluating Sparse Inertial Measurement Unit Configurations for Inferring Treadmill Running Motion
title_full_unstemmed Evaluating Sparse Inertial Measurement Unit Configurations for Inferring Treadmill Running Motion
title_short Evaluating Sparse Inertial Measurement Unit Configurations for Inferring Treadmill Running Motion
title_sort evaluating sparse inertial measurement unit configurations for inferring treadmill running motion
topic IMU
machine learning
running
sparse sensing
accelerometer
sampling rate
url https://www.mdpi.com/1424-8220/25/7/2105
work_keys_str_mv AT mackenzienpitts evaluatingsparseinertialmeasurementunitconfigurationsforinferringtreadmillrunningmotion
AT meganrebers evaluatingsparseinertialmeasurementunitconfigurationsforinferringtreadmillrunningmotion
AT cristineeagresta evaluatingsparseinertialmeasurementunitconfigurationsforinferringtreadmillrunningmotion
AT katherinemsteele evaluatingsparseinertialmeasurementunitconfigurationsforinferringtreadmillrunningmotion