User-Independent Intent Recognition for Lower-Limb Prostheses Using Depth Sensing

Yerzhan Massalin, Madina Abdrakhmanova, Huseyin Atakan Varol

Research output: Contribution to journalArticle

3 Citations (Scopus)

Abstract

Objective: The intent recognizers of advanced lower limb prostheses utilize mechanical sensors on the prosthesis and/or electromyographic measurements from the residual limb. Besides the delay caused by these signals, such systems require user-specific databases to train the recognizers. In this work, our objective is the development and validation of a user-independent intent recognition framework utilizing depth sensing. Methods: We collected a depth image dataset from 12 healthy subjects engaging in a variety of routine activities. After filtering the depth images, we extracted simple features employing a recursive strategy. The feature vectors were classified using a support vector machine. For robust activity mode switching, we implemented a voting filter scheme. Results: The model selection showed that the support vector machine classifier with no dimension reduction has the highest classification accuracy. Specifically, it reached 94.1% accuracy on the testing data from four subjects. We also observed a positive trend in the accuracy of classifiers trained with data from increasing the number of subjects. Activity mode switching using a voting filter detected 732 out of 778 activity mode transitions of the four users while initiating 70 erroneous transitions during steady-state activities. Conclusion: The intent recognizer trained on multiple subjects can be used for any other subject, providing a promising solution for supervisory control of powered lower limb prostheses. Significance: A user-independent intent recognition framework has the potential to decrease or eliminate the time required for extensive data collection regiments for intent recognizer training. This could accelerate the introduction of robotic lower limb prostheses to the market.

Original languageEnglish
JournalIEEE Transactions on Biomedical Engineering
DOIs
Publication statusAccepted/In press - Nov 20 2017
Externally publishedYes

Fingerprint

Prosthetics
Support vector machines
Classifiers
Signal systems
Robotics
Sensors
Testing

Keywords

  • big data
  • Cameras
  • Databases
  • Depth image processing
  • Feature extraction
  • intent recognition
  • lower-limb prosthesis
  • pattern recognition
  • Prosthetics
  • Real-time systems
  • Robot sensing systems
  • Three-dimensional displays

ASJC Scopus subject areas

  • Biomedical Engineering

Cite this

User-Independent Intent Recognition for Lower-Limb Prostheses Using Depth Sensing. / Massalin, Yerzhan; Abdrakhmanova, Madina; Varol, Huseyin Atakan.

In: IEEE Transactions on Biomedical Engineering, 20.11.2017.

Research output: Contribution to journalArticle

@article{3450841544364ebe97ede562f4492ea2,
title = "User-Independent Intent Recognition for Lower-Limb Prostheses Using Depth Sensing",
abstract = "Objective: The intent recognizers of advanced lower limb prostheses utilize mechanical sensors on the prosthesis and/or electromyographic measurements from the residual limb. Besides the delay caused by these signals, such systems require user-specific databases to train the recognizers. In this work, our objective is the development and validation of a user-independent intent recognition framework utilizing depth sensing. Methods: We collected a depth image dataset from 12 healthy subjects engaging in a variety of routine activities. After filtering the depth images, we extracted simple features employing a recursive strategy. The feature vectors were classified using a support vector machine. For robust activity mode switching, we implemented a voting filter scheme. Results: The model selection showed that the support vector machine classifier with no dimension reduction has the highest classification accuracy. Specifically, it reached 94.1{\%} accuracy on the testing data from four subjects. We also observed a positive trend in the accuracy of classifiers trained with data from increasing the number of subjects. Activity mode switching using a voting filter detected 732 out of 778 activity mode transitions of the four users while initiating 70 erroneous transitions during steady-state activities. Conclusion: The intent recognizer trained on multiple subjects can be used for any other subject, providing a promising solution for supervisory control of powered lower limb prostheses. Significance: A user-independent intent recognition framework has the potential to decrease or eliminate the time required for extensive data collection regiments for intent recognizer training. This could accelerate the introduction of robotic lower limb prostheses to the market.",
keywords = "big data, Cameras, Databases, Depth image processing, Feature extraction, intent recognition, lower-limb prosthesis, pattern recognition, Prosthetics, Real-time systems, Robot sensing systems, Three-dimensional displays",
author = "Yerzhan Massalin and Madina Abdrakhmanova and Varol, {Huseyin Atakan}",
year = "2017",
month = "11",
day = "20",
doi = "10.1109/TBME.2017.2776157",
language = "English",
journal = "IEEE Transactions on Biomedical Engineering",
issn = "0018-9294",
publisher = "IEEE Computer Society",

}

TY - JOUR

T1 - User-Independent Intent Recognition for Lower-Limb Prostheses Using Depth Sensing

AU - Massalin, Yerzhan

AU - Abdrakhmanova, Madina

AU - Varol, Huseyin Atakan

PY - 2017/11/20

Y1 - 2017/11/20

N2 - Objective: The intent recognizers of advanced lower limb prostheses utilize mechanical sensors on the prosthesis and/or electromyographic measurements from the residual limb. Besides the delay caused by these signals, such systems require user-specific databases to train the recognizers. In this work, our objective is the development and validation of a user-independent intent recognition framework utilizing depth sensing. Methods: We collected a depth image dataset from 12 healthy subjects engaging in a variety of routine activities. After filtering the depth images, we extracted simple features employing a recursive strategy. The feature vectors were classified using a support vector machine. For robust activity mode switching, we implemented a voting filter scheme. Results: The model selection showed that the support vector machine classifier with no dimension reduction has the highest classification accuracy. Specifically, it reached 94.1% accuracy on the testing data from four subjects. We also observed a positive trend in the accuracy of classifiers trained with data from increasing the number of subjects. Activity mode switching using a voting filter detected 732 out of 778 activity mode transitions of the four users while initiating 70 erroneous transitions during steady-state activities. Conclusion: The intent recognizer trained on multiple subjects can be used for any other subject, providing a promising solution for supervisory control of powered lower limb prostheses. Significance: A user-independent intent recognition framework has the potential to decrease or eliminate the time required for extensive data collection regiments for intent recognizer training. This could accelerate the introduction of robotic lower limb prostheses to the market.

AB - Objective: The intent recognizers of advanced lower limb prostheses utilize mechanical sensors on the prosthesis and/or electromyographic measurements from the residual limb. Besides the delay caused by these signals, such systems require user-specific databases to train the recognizers. In this work, our objective is the development and validation of a user-independent intent recognition framework utilizing depth sensing. Methods: We collected a depth image dataset from 12 healthy subjects engaging in a variety of routine activities. After filtering the depth images, we extracted simple features employing a recursive strategy. The feature vectors were classified using a support vector machine. For robust activity mode switching, we implemented a voting filter scheme. Results: The model selection showed that the support vector machine classifier with no dimension reduction has the highest classification accuracy. Specifically, it reached 94.1% accuracy on the testing data from four subjects. We also observed a positive trend in the accuracy of classifiers trained with data from increasing the number of subjects. Activity mode switching using a voting filter detected 732 out of 778 activity mode transitions of the four users while initiating 70 erroneous transitions during steady-state activities. Conclusion: The intent recognizer trained on multiple subjects can be used for any other subject, providing a promising solution for supervisory control of powered lower limb prostheses. Significance: A user-independent intent recognition framework has the potential to decrease or eliminate the time required for extensive data collection regiments for intent recognizer training. This could accelerate the introduction of robotic lower limb prostheses to the market.

KW - big data

KW - Cameras

KW - Databases

KW - Depth image processing

KW - Feature extraction

KW - intent recognition

KW - lower-limb prosthesis

KW - pattern recognition

KW - Prosthetics

KW - Real-time systems

KW - Robot sensing systems

KW - Three-dimensional displays

UR - http://www.scopus.com/inward/record.url?scp=85035808915&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85035808915&partnerID=8YFLogxK

U2 - 10.1109/TBME.2017.2776157

DO - 10.1109/TBME.2017.2776157

M3 - Article

AN - SCOPUS:85035808915

JO - IEEE Transactions on Biomedical Engineering

JF - IEEE Transactions on Biomedical Engineering

SN - 0018-9294

ER -