Automatic and continuous discomfort detection for premature infants in a NICU using video-based motion analysis

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

Abstract

Frequent pain and discomfort in premature infants can lead to long-term adverse neurodevelopmental outcomes. Video-based monitoring is considered to be a promising contactless method for identification of discomfort moments. In this study, we propose a video-based method for automated detection of infant discomfort. The method is based on analyzing facial and body motion. Therefore, motion trajectories are estimated from frame to frame using optical flow. For each video segment, we further calculate the motion acceleration rate and extract 18 time- and frequency-domain features characterizing motion patterns. A support vector machine (SVM) classifier is then applied to video sequences to recognize infant status of comfort or discomfort. The method is evaluated using 183 video segments for 11 infants from 17 heel prick events. Experimental results show an AUC of 0.94 for discomfort detection and the average accuracy of 0.86 when combining all proposed features, which is promising for clinical use.
Original languageEnglish
Title of host publication2019 41st Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC)
Place of PublicationPiscataway
PublisherInstitute of Electrical and Electronics Engineers
Pages5995-5999
Number of pages5
ISBN (Electronic)978-1-5386-1311-5
DOIs
Publication statusPublished - 2019
Event41st Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC 2019)
- City Cube Berlin, Berlin, Germany
Duration: 23 Jul 201927 Jul 2019
https://embc.embs.org/2019/

Conference

Conference41st Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC 2019)
Abbreviated titleEMBC 2019
CountryGermany
CityBerlin
Period23/07/1927/07/19
Internet address

Fingerprint

Optical flows
Support vector machines
Classifiers
Trajectories
Monitoring
Motion analysis

Cite this

Sun, Y., Kommers, D., Wang, W., Joshi, R., Shan, C., Tan, T., ... With, P. H. N. D. (2019). Automatic and continuous discomfort detection for premature infants in a NICU using video-based motion analysis. In 2019 41st Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC) (pp. 5995-5999). Piscataway: Institute of Electrical and Electronics Engineers. https://doi.org/10.1109/EMBC.2019.8857597
Sun, Y. ; Kommers, D. ; Wang, W. ; Joshi, R. ; Shan, C. ; Tan, T. ; Aarts, R. M. ; Pul, C. van ; Andriessen, P. ; With, P. H. N. de. / Automatic and continuous discomfort detection for premature infants in a NICU using video-based motion analysis. 2019 41st Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC). Piscataway : Institute of Electrical and Electronics Engineers, 2019. pp. 5995-5999
@inproceedings{23cb831ed0aa4fdb88923cf4f1be5966,
title = "Automatic and continuous discomfort detection for premature infants in a NICU using video-based motion analysis",
abstract = "Frequent pain and discomfort in premature infants can lead to long-term adverse neurodevelopmental outcomes. Video-based monitoring is considered to be a promising contactless method for identification of discomfort moments. In this study, we propose a video-based method for automated detection of infant discomfort. The method is based on analyzing facial and body motion. Therefore, motion trajectories are estimated from frame to frame using optical flow. For each video segment, we further calculate the motion acceleration rate and extract 18 time- and frequency-domain features characterizing motion patterns. A support vector machine (SVM) classifier is then applied to video sequences to recognize infant status of comfort or discomfort. The method is evaluated using 183 video segments for 11 infants from 17 heel prick events. Experimental results show an AUC of 0.94 for discomfort detection and the average accuracy of 0.86 when combining all proposed features, which is promising for clinical use.",
keywords = "Feature extraction, Pediatrics, Pain, Acceleration, Motion segmentation, Monitoring, Correlation",
author = "Y. Sun and D. Kommers and W. Wang and R. Joshi and C. Shan and T. Tan and Aarts, {R. M.} and Pul, {C. van} and P. Andriessen and With, {P. H. N. de}",
year = "2019",
doi = "10.1109/EMBC.2019.8857597",
language = "English",
pages = "5995--5999",
booktitle = "2019 41st Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC)",
publisher = "Institute of Electrical and Electronics Engineers",
address = "United States",

}

Sun, Y, Kommers, D, Wang, W, Joshi, R, Shan, C, Tan, T, Aarts, RM, Pul, CV, Andriessen, P & With, PHND 2019, Automatic and continuous discomfort detection for premature infants in a NICU using video-based motion analysis. in 2019 41st Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC). Institute of Electrical and Electronics Engineers, Piscataway, pp. 5995-5999, 41st Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC 2019)
, Berlin, Germany, 23/07/19. https://doi.org/10.1109/EMBC.2019.8857597

Automatic and continuous discomfort detection for premature infants in a NICU using video-based motion analysis. / Sun, Y.; Kommers, D.; Wang, W.; Joshi, R.; Shan, C.; Tan, T.; Aarts, R. M.; Pul, C. van; Andriessen, P.; With, P. H. N. de.

2019 41st Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC). Piscataway : Institute of Electrical and Electronics Engineers, 2019. p. 5995-5999.

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

TY - GEN

T1 - Automatic and continuous discomfort detection for premature infants in a NICU using video-based motion analysis

AU - Sun, Y.

AU - Kommers, D.

AU - Wang, W.

AU - Joshi, R.

AU - Shan, C.

AU - Tan, T.

AU - Aarts, R. M.

AU - Pul, C. van

AU - Andriessen, P.

AU - With, P. H. N. de

PY - 2019

Y1 - 2019

N2 - Frequent pain and discomfort in premature infants can lead to long-term adverse neurodevelopmental outcomes. Video-based monitoring is considered to be a promising contactless method for identification of discomfort moments. In this study, we propose a video-based method for automated detection of infant discomfort. The method is based on analyzing facial and body motion. Therefore, motion trajectories are estimated from frame to frame using optical flow. For each video segment, we further calculate the motion acceleration rate and extract 18 time- and frequency-domain features characterizing motion patterns. A support vector machine (SVM) classifier is then applied to video sequences to recognize infant status of comfort or discomfort. The method is evaluated using 183 video segments for 11 infants from 17 heel prick events. Experimental results show an AUC of 0.94 for discomfort detection and the average accuracy of 0.86 when combining all proposed features, which is promising for clinical use.

AB - Frequent pain and discomfort in premature infants can lead to long-term adverse neurodevelopmental outcomes. Video-based monitoring is considered to be a promising contactless method for identification of discomfort moments. In this study, we propose a video-based method for automated detection of infant discomfort. The method is based on analyzing facial and body motion. Therefore, motion trajectories are estimated from frame to frame using optical flow. For each video segment, we further calculate the motion acceleration rate and extract 18 time- and frequency-domain features characterizing motion patterns. A support vector machine (SVM) classifier is then applied to video sequences to recognize infant status of comfort or discomfort. The method is evaluated using 183 video segments for 11 infants from 17 heel prick events. Experimental results show an AUC of 0.94 for discomfort detection and the average accuracy of 0.86 when combining all proposed features, which is promising for clinical use.

KW - Feature extraction

KW - Pediatrics

KW - Pain

KW - Acceleration

KW - Motion segmentation

KW - Monitoring

KW - Correlation

U2 - 10.1109/EMBC.2019.8857597

DO - 10.1109/EMBC.2019.8857597

M3 - Conference contribution

SP - 5995

EP - 5999

BT - 2019 41st Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC)

PB - Institute of Electrical and Electronics Engineers

CY - Piscataway

ER -

Sun Y, Kommers D, Wang W, Joshi R, Shan C, Tan T et al. Automatic and continuous discomfort detection for premature infants in a NICU using video-based motion analysis. In 2019 41st Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC). Piscataway: Institute of Electrical and Electronics Engineers. 2019. p. 5995-5999 https://doi.org/10.1109/EMBC.2019.8857597