A fall detection method based on acceleration data and hidden Markov model

Huiqiang Cao; Shuicai Wu; Zhuhuang Zhou; Chung Chih Lin; Chih Yu Yang; Shih Tseng Lee; Chieh Tsai Wu

doi:10.1109/SIPROCESS.2016.7888350

A fall detection method based on acceleration data and hidden Markov model

Huiqiang Cao, Shuicai Wu, Zhuhuang Zhou, Chung Chih Lin, Chih Yu Yang, Shih Tseng Lee, Chieh Tsai Wu

Department of Computer Science and Information Engineering

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

17 Scopus citations

Abstract

Falls have been a major health risk that diminishes the quality of life among the elderly. In this paper, we propose a new method using acceleration data and hidden Markov model (HMM) to detect fall events. A wearable device integrating a tri-axial accelerometer was used to collect acceleration data of human chest. Feature sequences (FSs) were extracted from the acceleration data and used as sequence of observations to train an HMM of fall detection. The probability of the input FS generated by the model was calculated as the detection standard. Experimental results showed that the accuracy of the proposed method was 97.2%, the sensitivity was 91.7%, and the specificity was 100%, demonstrating desired performance of our method in detecting fall events.

Original language	English
Title of host publication	2016 IEEE International Conference on Signal and Image Processing, ICSIP 2016
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	684-689
Number of pages	6
ISBN (Electronic)	9781509023769
DOIs	https://doi.org/10.1109/SIPROCESS.2016.7888350
State	Published - 27 03 2017
Event	2016 IEEE International Conference on Signal and Image Processing, ICSIP 2016 - Beijing, China Duration: 13 08 2016 → 15 08 2016

Publication series

Name	2016 IEEE International Conference on Signal and Image Processing, ICSIP 2016

Conference

Conference	2016 IEEE International Conference on Signal and Image Processing, ICSIP 2016
Country/Territory	China
City	Beijing
Period	13/08/16 → 15/08/16

Bibliographical note

Publisher Copyright:
© 2016 IEEE.

Keywords

Fall detection
acceleration
hidden Markov model

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

10.1109/SIPROCESS.2016.7888350

Cite this

Cao, H., Wu, S., Zhou, Z., Lin, C. C., Yang, C. Y., Lee, S. T., & Wu, C. T. (2017). A fall detection method based on acceleration data and hidden Markov model. In 2016 IEEE International Conference on Signal and Image Processing, ICSIP 2016 (pp. 684-689). Article 7888350 (2016 IEEE International Conference on Signal and Image Processing, ICSIP 2016). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/SIPROCESS.2016.7888350

@inproceedings{bd7678d399494e2795f9c069dc62f8fb,

title = "A fall detection method based on acceleration data and hidden Markov model",

abstract = "Falls have been a major health risk that diminishes the quality of life among the elderly. In this paper, we propose a new method using acceleration data and hidden Markov model (HMM) to detect fall events. A wearable device integrating a tri-axial accelerometer was used to collect acceleration data of human chest. Feature sequences (FSs) were extracted from the acceleration data and used as sequence of observations to train an HMM of fall detection. The probability of the input FS generated by the model was calculated as the detection standard. Experimental results showed that the accuracy of the proposed method was 97.2%, the sensitivity was 91.7%, and the specificity was 100%, demonstrating desired performance of our method in detecting fall events.",

keywords = "Fall detection, acceleration, hidden Markov model",

author = "Huiqiang Cao and Shuicai Wu and Zhuhuang Zhou and Lin, {Chung Chih} and Yang, {Chih Yu} and Lee, {Shih Tseng} and Wu, {Chieh Tsai}",

note = "Publisher Copyright: {\textcopyright} 2016 IEEE.; 2016 IEEE International Conference on Signal and Image Processing, ICSIP 2016 ; Conference date: 13-08-2016 Through 15-08-2016",

year = "2017",

month = mar,

day = "27",

doi = "10.1109/SIPROCESS.2016.7888350",

language = "英语",

series = "2016 IEEE International Conference on Signal and Image Processing, ICSIP 2016",

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

pages = "684--689",

booktitle = "2016 IEEE International Conference on Signal and Image Processing, ICSIP 2016",

address = "美国",

}

Cao, H, Wu, S, Zhou, Z, Lin, CC, Yang, CY, Lee, ST & Wu, CT 2017, A fall detection method based on acceleration data and hidden Markov model. in 2016 IEEE International Conference on Signal and Image Processing, ICSIP 2016., 7888350, 2016 IEEE International Conference on Signal and Image Processing, ICSIP 2016, Institute of Electrical and Electronics Engineers Inc., pp. 684-689, 2016 IEEE International Conference on Signal and Image Processing, ICSIP 2016, Beijing, China, 13/08/16. https://doi.org/10.1109/SIPROCESS.2016.7888350

A fall detection method based on acceleration data and hidden Markov model. / Cao, Huiqiang; Wu, Shuicai; Zhou, Zhuhuang et al.
2016 IEEE International Conference on Signal and Image Processing, ICSIP 2016. Institute of Electrical and Electronics Engineers Inc., 2017. p. 684-689 7888350 (2016 IEEE International Conference on Signal and Image Processing, ICSIP 2016).

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

TY - GEN

T1 - A fall detection method based on acceleration data and hidden Markov model

AU - Cao, Huiqiang

AU - Wu, Shuicai

AU - Zhou, Zhuhuang

AU - Lin, Chung Chih

AU - Yang, Chih Yu

AU - Lee, Shih Tseng

AU - Wu, Chieh Tsai

PY - 2017/3/27

Y1 - 2017/3/27

N2 - Falls have been a major health risk that diminishes the quality of life among the elderly. In this paper, we propose a new method using acceleration data and hidden Markov model (HMM) to detect fall events. A wearable device integrating a tri-axial accelerometer was used to collect acceleration data of human chest. Feature sequences (FSs) were extracted from the acceleration data and used as sequence of observations to train an HMM of fall detection. The probability of the input FS generated by the model was calculated as the detection standard. Experimental results showed that the accuracy of the proposed method was 97.2%, the sensitivity was 91.7%, and the specificity was 100%, demonstrating desired performance of our method in detecting fall events.

AB - Falls have been a major health risk that diminishes the quality of life among the elderly. In this paper, we propose a new method using acceleration data and hidden Markov model (HMM) to detect fall events. A wearable device integrating a tri-axial accelerometer was used to collect acceleration data of human chest. Feature sequences (FSs) were extracted from the acceleration data and used as sequence of observations to train an HMM of fall detection. The probability of the input FS generated by the model was calculated as the detection standard. Experimental results showed that the accuracy of the proposed method was 97.2%, the sensitivity was 91.7%, and the specificity was 100%, demonstrating desired performance of our method in detecting fall events.

KW - Fall detection

KW - acceleration

KW - hidden Markov model

UR - http://www.scopus.com/inward/record.url?scp=85018702752&partnerID=8YFLogxK

U2 - 10.1109/SIPROCESS.2016.7888350

DO - 10.1109/SIPROCESS.2016.7888350

M3 - 会议稿件

AN - SCOPUS:85018702752

T3 - 2016 IEEE International Conference on Signal and Image Processing, ICSIP 2016

SP - 684

EP - 689

BT - 2016 IEEE International Conference on Signal and Image Processing, ICSIP 2016

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 2016 IEEE International Conference on Signal and Image Processing, ICSIP 2016

Y2 - 13 August 2016 through 15 August 2016

ER -

Cao H, Wu S, Zhou Z, Lin CC, Yang CY, Lee ST et al. A fall detection method based on acceleration data and hidden Markov model. In 2016 IEEE International Conference on Signal and Image Processing, ICSIP 2016. Institute of Electrical and Electronics Engineers Inc. 2017. p. 684-689. 7888350. (2016 IEEE International Conference on Signal and Image Processing, ICSIP 2016). doi: 10.1109/SIPROCESS.2016.7888350

A fall detection method based on acceleration data and hidden Markov model

Abstract

Publication series

Conference

Bibliographical note

Keywords

UN SDGs

Access to Document

Other files and links

Fingerprint

Cite this