Application of neural networks for filtering non-impact transients recorded from biomechanical sensors

Shruti Motiwale; William Eppler; Dale Hollingsworth; Chad Hollingsworth; Justin Morgenthau; Reuben H. Kraft

doi:10.1109/BHI.2016.7455870

Application of neural networks for filtering non-impact transients recorded from biomechanical sensors

Shruti Motiwale, William Eppler, Dale Hollingsworth, Chad Hollingsworth, Justin Morgenthau, Reuben H. Kraft

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

9 Scopus citations

Abstract

A number of biomechanical sensor applications are available on the market today for sports, as well as military applications. These sensors are capable of measuring accelerations and velocities of impact that may be useful for understanding head and neck biomechanics; although the proper use and accuracy of these analysis techniques is still a topic of research. A challenge with these sensors is filtering the raw data to include authentic impacts and exclude false events that are sometimes registered due to the sensitivity of the sensors. In this study we have developed an algorithm based on artificial neural networks and discrete fourier transform based filtering that will be used to distinguish between a biomechanically relevant event and a non-impact transient which is of no concern. Linear accelerations and angular velocities are used as inputs to the pattern recognition algorithm and the output vector contains the probability of each impact belonging to a certain class. Using this approach we report a specificity of 47% and a sensitivity of 88% in the ability to distinguish between real impacts and non-impact transients.

Original language	English (US)
Title of host publication	3rd IEEE EMBS International Conference on Biomedical and Health Informatics, BHI 2016
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	204-207
Number of pages	4
ISBN (Electronic)	9781509024551
DOIs	https://doi.org/10.1109/BHI.2016.7455870
State	Published - Apr 18 2016
Event	3rd IEEE EMBS International Conference on Biomedical and Health Informatics, BHI 2016 - Las Vegas, United States Duration: Feb 24 2016 → Feb 27 2016

Publication series

Name	3rd IEEE EMBS International Conference on Biomedical and Health Informatics, BHI 2016

Other

Other	3rd IEEE EMBS International Conference on Biomedical and Health Informatics, BHI 2016
Country/Territory	United States
City	Las Vegas
Period	2/24/16 → 2/27/16

All Science Journal Classification (ASJC) codes

Health Informatics
Health Information Management

Access to Document

10.1109/BHI.2016.7455870

Cite this

Motiwale, S., Eppler, W., Hollingsworth, D., Hollingsworth, C., Morgenthau, J., & Kraft, R. H. (2016). Application of neural networks for filtering non-impact transients recorded from biomechanical sensors. In 3rd IEEE EMBS International Conference on Biomedical and Health Informatics, BHI 2016 (pp. 204-207). [7455870] (3rd IEEE EMBS International Conference on Biomedical and Health Informatics, BHI 2016). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/BHI.2016.7455870

Motiwale, Shruti ; Eppler, William ; Hollingsworth, Dale et al. / Application of neural networks for filtering non-impact transients recorded from biomechanical sensors. 3rd IEEE EMBS International Conference on Biomedical and Health Informatics, BHI 2016. Institute of Electrical and Electronics Engineers Inc., 2016. pp. 204-207 (3rd IEEE EMBS International Conference on Biomedical and Health Informatics, BHI 2016).

@inproceedings{1c2327b8499749c38af0317f1d0a0514,

title = "Application of neural networks for filtering non-impact transients recorded from biomechanical sensors",

abstract = "A number of biomechanical sensor applications are available on the market today for sports, as well as military applications. These sensors are capable of measuring accelerations and velocities of impact that may be useful for understanding head and neck biomechanics; although the proper use and accuracy of these analysis techniques is still a topic of research. A challenge with these sensors is filtering the raw data to include authentic impacts and exclude false events that are sometimes registered due to the sensitivity of the sensors. In this study we have developed an algorithm based on artificial neural networks and discrete fourier transform based filtering that will be used to distinguish between a biomechanically relevant event and a non-impact transient which is of no concern. Linear accelerations and angular velocities are used as inputs to the pattern recognition algorithm and the output vector contains the probability of each impact belonging to a certain class. Using this approach we report a specificity of 47% and a sensitivity of 88% in the ability to distinguish between real impacts and non-impact transients.",

author = "Shruti Motiwale and William Eppler and Dale Hollingsworth and Chad Hollingsworth and Justin Morgenthau and Kraft, {Reuben H.}",

note = "Publisher Copyright: {\textcopyright} 2016 IEEE.; 3rd IEEE EMBS International Conference on Biomedical and Health Informatics, BHI 2016 ; Conference date: 24-02-2016 Through 27-02-2016",

year = "2016",

month = apr,

day = "18",

doi = "10.1109/BHI.2016.7455870",

language = "English (US)",

series = "3rd IEEE EMBS International Conference on Biomedical and Health Informatics, BHI 2016",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "204--207",

booktitle = "3rd IEEE EMBS International Conference on Biomedical and Health Informatics, BHI 2016",

address = "United States",

}

Motiwale, S, Eppler, W, Hollingsworth, D, Hollingsworth, C, Morgenthau, J & Kraft, RH 2016, Application of neural networks for filtering non-impact transients recorded from biomechanical sensors. in 3rd IEEE EMBS International Conference on Biomedical and Health Informatics, BHI 2016., 7455870, 3rd IEEE EMBS International Conference on Biomedical and Health Informatics, BHI 2016, Institute of Electrical and Electronics Engineers Inc., pp. 204-207, 3rd IEEE EMBS International Conference on Biomedical and Health Informatics, BHI 2016, Las Vegas, United States, 2/24/16. https://doi.org/10.1109/BHI.2016.7455870

Application of neural networks for filtering non-impact transients recorded from biomechanical sensors. / Motiwale, Shruti; Eppler, William; Hollingsworth, Dale et al.
3rd IEEE EMBS International Conference on Biomedical and Health Informatics, BHI 2016. Institute of Electrical and Electronics Engineers Inc., 2016. p. 204-207 7455870 (3rd IEEE EMBS International Conference on Biomedical and Health Informatics, BHI 2016).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

TY - GEN

T1 - Application of neural networks for filtering non-impact transients recorded from biomechanical sensors

AU - Motiwale, Shruti

AU - Eppler, William

AU - Hollingsworth, Dale

AU - Hollingsworth, Chad

AU - Morgenthau, Justin

AU - Kraft, Reuben H.

PY - 2016/4/18

Y1 - 2016/4/18

N2 - A number of biomechanical sensor applications are available on the market today for sports, as well as military applications. These sensors are capable of measuring accelerations and velocities of impact that may be useful for understanding head and neck biomechanics; although the proper use and accuracy of these analysis techniques is still a topic of research. A challenge with these sensors is filtering the raw data to include authentic impacts and exclude false events that are sometimes registered due to the sensitivity of the sensors. In this study we have developed an algorithm based on artificial neural networks and discrete fourier transform based filtering that will be used to distinguish between a biomechanically relevant event and a non-impact transient which is of no concern. Linear accelerations and angular velocities are used as inputs to the pattern recognition algorithm and the output vector contains the probability of each impact belonging to a certain class. Using this approach we report a specificity of 47% and a sensitivity of 88% in the ability to distinguish between real impacts and non-impact transients.

AB - A number of biomechanical sensor applications are available on the market today for sports, as well as military applications. These sensors are capable of measuring accelerations and velocities of impact that may be useful for understanding head and neck biomechanics; although the proper use and accuracy of these analysis techniques is still a topic of research. A challenge with these sensors is filtering the raw data to include authentic impacts and exclude false events that are sometimes registered due to the sensitivity of the sensors. In this study we have developed an algorithm based on artificial neural networks and discrete fourier transform based filtering that will be used to distinguish between a biomechanically relevant event and a non-impact transient which is of no concern. Linear accelerations and angular velocities are used as inputs to the pattern recognition algorithm and the output vector contains the probability of each impact belonging to a certain class. Using this approach we report a specificity of 47% and a sensitivity of 88% in the ability to distinguish between real impacts and non-impact transients.

UR - http://www.scopus.com/inward/record.url?scp=84968644504&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84968644504&partnerID=8YFLogxK

U2 - 10.1109/BHI.2016.7455870

DO - 10.1109/BHI.2016.7455870

M3 - Conference contribution

AN - SCOPUS:84968644504

T3 - 3rd IEEE EMBS International Conference on Biomedical and Health Informatics, BHI 2016

SP - 204

EP - 207

BT - 3rd IEEE EMBS International Conference on Biomedical and Health Informatics, BHI 2016

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 3rd IEEE EMBS International Conference on Biomedical and Health Informatics, BHI 2016

Y2 - 24 February 2016 through 27 February 2016

ER -

Motiwale S, Eppler W, Hollingsworth D, Hollingsworth C, Morgenthau J, Kraft RH. Application of neural networks for filtering non-impact transients recorded from biomechanical sensors. In 3rd IEEE EMBS International Conference on Biomedical and Health Informatics, BHI 2016. Institute of Electrical and Electronics Engineers Inc. 2016. p. 204-207. 7455870. (3rd IEEE EMBS International Conference on Biomedical and Health Informatics, BHI 2016). doi: 10.1109/BHI.2016.7455870

Application of neural networks for filtering non-impact transients recorded from biomechanical sensors

Abstract

Publication series

Other

All Science Journal Classification (ASJC) codes

Access to Document

Other files and links

Fingerprint

Cite this