On-body inertial sensor location recognition

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademic

33 Downloads (Pure)

Abstract

Introduction and past research: In previous work we presented an algorithm for automatically identifying the body segment to which an inertial sensor is attached during walking [1]. Using this method, the set-up of inertial motion capture systems becomes easier and attachment errors are avoided. The user can place (wireless) inertial sensors on arbitrary body segments. Then, after walking for a few steps, the segment to which each sensor is attached is identified automatically. To classify the sensors, a decision tree was trained using ranked features extracted from magnitudes, x- y- and z-components of accelerations, angular velocities and angular accelerations. Method: Drawback of using ranking and correlation coefficients as features is that information from different sensors needs to be combined. Therefore we started looking into a new method using the same data and the same extracted features as in [1], but without using the ranking and the correlation coefficients between different sensors. Instead of a decision tree, we used logistic regression for classifying the sensors [2]. Unlike decision trees, with logistic regression a probability is calculated for each body part on which the sensor can be placed. To develop a method that works for different activities of daily living, we recorded 18 activities of ten healthy subjects using 17 inertial sensors. Walking at different speeds, sit to stand, lying down, grasping objects, jumping, walking stairs and cycling were recorded. The goal is – based on the data of single sensor — to predict the body segment to which this sensor is attached, for different activities of daily living. Results: A logistic regression classifier was developed and tested with 10-fold crossvalidation using 31 walking trials of ten healthy subjects. In the case of a full-body configuration 482 of a total of 527 (31 x 17) sensors were correctly classified (91.5%). Discussion: Using our algorithm it is possible to create an intelligent sensor, which can determine its own location on the body. The data of the measurements of different daily-life activities is currently being analysed. In addition, we will look into the possibility of simultaneously predicting the on-body location of each sensor and the performed activity.
Original languageUndefined
Title of host publication5th Dutch Conference on Bio-Medical Engineering, BME 2015
Place of PublicationEgmond aan Zee, The Netherlands
PublisherBME
Pages-
Number of pages1
ISBN (Print)not assigned
Publication statusPublished - 22 Jan 2015
Event5th Dutch Bio-Medical Engineering Conference, BME 2015 - Hotel Zuiderduin, Egmond aan Zee, Netherlands
Duration: 22 Jan 201523 Jan 2015
Conference number: 5
http://www.bme2015.nl/

Publication series

Name
PublisherBME

Conference

Conference5th Dutch Bio-Medical Engineering Conference, BME 2015
Abbreviated titleBME 2015
CountryNetherlands
CityEgmond aan Zee
Period22/01/1523/01/15
Internet address

Keywords

  • EWI-25690
  • METIS-312493
  • IR-94112

Cite this

Weenk, D., van Beijnum, B. J. F., Goaied, S., Baten, C. T. M., Hermens, H. J., & Veltink, P. H. (2015). On-body inertial sensor location recognition. In 5th Dutch Conference on Bio-Medical Engineering, BME 2015 (pp. -). Egmond aan Zee, The Netherlands: BME.
Weenk, D. ; van Beijnum, Bernhard J.F. ; Goaied, Salma ; Baten, Christian T.M. ; Hermens, Hermanus J. ; Veltink, Petrus H. / On-body inertial sensor location recognition. 5th Dutch Conference on Bio-Medical Engineering, BME 2015. Egmond aan Zee, The Netherlands : BME, 2015. pp. -
@inproceedings{93555d05c0ff468b936d8d7042103237,
title = "On-body inertial sensor location recognition",
abstract = "Introduction and past research: In previous work we presented an algorithm for automatically identifying the body segment to which an inertial sensor is attached during walking [1]. Using this method, the set-up of inertial motion capture systems becomes easier and attachment errors are avoided. The user can place (wireless) inertial sensors on arbitrary body segments. Then, after walking for a few steps, the segment to which each sensor is attached is identified automatically. To classify the sensors, a decision tree was trained using ranked features extracted from magnitudes, x- y- and z-components of accelerations, angular velocities and angular accelerations. Method: Drawback of using ranking and correlation coefficients as features is that information from different sensors needs to be combined. Therefore we started looking into a new method using the same data and the same extracted features as in [1], but without using the ranking and the correlation coefficients between different sensors. Instead of a decision tree, we used logistic regression for classifying the sensors [2]. Unlike decision trees, with logistic regression a probability is calculated for each body part on which the sensor can be placed. To develop a method that works for different activities of daily living, we recorded 18 activities of ten healthy subjects using 17 inertial sensors. Walking at different speeds, sit to stand, lying down, grasping objects, jumping, walking stairs and cycling were recorded. The goal is – based on the data of single sensor — to predict the body segment to which this sensor is attached, for different activities of daily living. Results: A logistic regression classifier was developed and tested with 10-fold crossvalidation using 31 walking trials of ten healthy subjects. In the case of a full-body configuration 482 of a total of 527 (31 x 17) sensors were correctly classified (91.5{\%}). Discussion: Using our algorithm it is possible to create an intelligent sensor, which can determine its own location on the body. The data of the measurements of different daily-life activities is currently being analysed. In addition, we will look into the possibility of simultaneously predicting the on-body location of each sensor and the performed activity.",
keywords = "EWI-25690, METIS-312493, IR-94112",
author = "D. Weenk and {van Beijnum}, {Bernhard J.F.} and Salma Goaied and Baten, {Christian T.M.} and Hermens, {Hermanus J.} and Veltink, {Petrus H.}",
year = "2015",
month = "1",
day = "22",
language = "Undefined",
isbn = "not assigned",
publisher = "BME",
pages = "--",
booktitle = "5th Dutch Conference on Bio-Medical Engineering, BME 2015",

}

Weenk, D, van Beijnum, BJF, Goaied, S, Baten, CTM, Hermens, HJ & Veltink, PH 2015, On-body inertial sensor location recognition. in 5th Dutch Conference on Bio-Medical Engineering, BME 2015. BME, Egmond aan Zee, The Netherlands, pp. -, 5th Dutch Bio-Medical Engineering Conference, BME 2015, Egmond aan Zee, Netherlands, 22/01/15.

On-body inertial sensor location recognition. / Weenk, D.; van Beijnum, Bernhard J.F.; Goaied, Salma; Baten, Christian T.M.; Hermens, Hermanus J.; Veltink, Petrus H.

5th Dutch Conference on Bio-Medical Engineering, BME 2015. Egmond aan Zee, The Netherlands : BME, 2015. p. -.

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademic

TY - GEN

T1 - On-body inertial sensor location recognition

AU - Weenk, D.

AU - van Beijnum, Bernhard J.F.

AU - Goaied, Salma

AU - Baten, Christian T.M.

AU - Hermens, Hermanus J.

AU - Veltink, Petrus H.

PY - 2015/1/22

Y1 - 2015/1/22

N2 - Introduction and past research: In previous work we presented an algorithm for automatically identifying the body segment to which an inertial sensor is attached during walking [1]. Using this method, the set-up of inertial motion capture systems becomes easier and attachment errors are avoided. The user can place (wireless) inertial sensors on arbitrary body segments. Then, after walking for a few steps, the segment to which each sensor is attached is identified automatically. To classify the sensors, a decision tree was trained using ranked features extracted from magnitudes, x- y- and z-components of accelerations, angular velocities and angular accelerations. Method: Drawback of using ranking and correlation coefficients as features is that information from different sensors needs to be combined. Therefore we started looking into a new method using the same data and the same extracted features as in [1], but without using the ranking and the correlation coefficients between different sensors. Instead of a decision tree, we used logistic regression for classifying the sensors [2]. Unlike decision trees, with logistic regression a probability is calculated for each body part on which the sensor can be placed. To develop a method that works for different activities of daily living, we recorded 18 activities of ten healthy subjects using 17 inertial sensors. Walking at different speeds, sit to stand, lying down, grasping objects, jumping, walking stairs and cycling were recorded. The goal is – based on the data of single sensor — to predict the body segment to which this sensor is attached, for different activities of daily living. Results: A logistic regression classifier was developed and tested with 10-fold crossvalidation using 31 walking trials of ten healthy subjects. In the case of a full-body configuration 482 of a total of 527 (31 x 17) sensors were correctly classified (91.5%). Discussion: Using our algorithm it is possible to create an intelligent sensor, which can determine its own location on the body. The data of the measurements of different daily-life activities is currently being analysed. In addition, we will look into the possibility of simultaneously predicting the on-body location of each sensor and the performed activity.

AB - Introduction and past research: In previous work we presented an algorithm for automatically identifying the body segment to which an inertial sensor is attached during walking [1]. Using this method, the set-up of inertial motion capture systems becomes easier and attachment errors are avoided. The user can place (wireless) inertial sensors on arbitrary body segments. Then, after walking for a few steps, the segment to which each sensor is attached is identified automatically. To classify the sensors, a decision tree was trained using ranked features extracted from magnitudes, x- y- and z-components of accelerations, angular velocities and angular accelerations. Method: Drawback of using ranking and correlation coefficients as features is that information from different sensors needs to be combined. Therefore we started looking into a new method using the same data and the same extracted features as in [1], but without using the ranking and the correlation coefficients between different sensors. Instead of a decision tree, we used logistic regression for classifying the sensors [2]. Unlike decision trees, with logistic regression a probability is calculated for each body part on which the sensor can be placed. To develop a method that works for different activities of daily living, we recorded 18 activities of ten healthy subjects using 17 inertial sensors. Walking at different speeds, sit to stand, lying down, grasping objects, jumping, walking stairs and cycling were recorded. The goal is – based on the data of single sensor — to predict the body segment to which this sensor is attached, for different activities of daily living. Results: A logistic regression classifier was developed and tested with 10-fold crossvalidation using 31 walking trials of ten healthy subjects. In the case of a full-body configuration 482 of a total of 527 (31 x 17) sensors were correctly classified (91.5%). Discussion: Using our algorithm it is possible to create an intelligent sensor, which can determine its own location on the body. The data of the measurements of different daily-life activities is currently being analysed. In addition, we will look into the possibility of simultaneously predicting the on-body location of each sensor and the performed activity.

KW - EWI-25690

KW - METIS-312493

KW - IR-94112

M3 - Conference contribution

SN - not assigned

SP - -

BT - 5th Dutch Conference on Bio-Medical Engineering, BME 2015

PB - BME

CY - Egmond aan Zee, The Netherlands

ER -

Weenk D, van Beijnum BJF, Goaied S, Baten CTM, Hermens HJ, Veltink PH. On-body inertial sensor location recognition. In 5th Dutch Conference on Bio-Medical Engineering, BME 2015. Egmond aan Zee, The Netherlands: BME. 2015. p. -