Automatic Coding of Facial Expressions of Pain: Are We There Yet?
Journal
Pain research & management
ISSN: 1918-1523
Titre abrégé: Pain Res Manag
Pays: United States
ID NLM: 9612504
Informations de publication
Date de publication:
2022
2022
Historique:
received:
26
11
2020
revised:
11
09
2021
accepted:
16
12
2021
entrez:
24
1
2022
pubmed:
25
1
2022
medline:
25
3
2022
Statut:
epublish
Résumé
The experience of pain is regularly accompanied by facial expressions. The gold standard for analyzing these facial expressions is the Facial Action Coding System (FACS), which provides so-called action units (AUs) as parametrical indicators of facial muscular activity. Particular combinations of AUs have appeared to be pain-indicative. The manual coding of AUs is, however, too time- and labor-intensive in clinical practice. New developments in automatic facial expression analysis have promised to enable automatic detection of AUs, which might be used for pain detection. Our aim is to compare manual with automatic AU coding of facial expressions of pain. FaceReader7 was used for automatic AU detection. We compared the performance of FaceReader7 using videos of 40 participants (20 younger with a mean age of 25.7 years and 20 older with a mean age of 52.1 years) undergoing experimentally induced heat pain to manually coded AUs as gold standard labeling. Percentages of correctly and falsely classified AUs were calculated, and we computed as indicators of congruency, "sensitivity/recall," "precision," and "overall agreement (F1)." The automatic coding of AUs only showed poor to moderate outcomes regarding sensitivity/recall, precision, and F1. The congruency was better for younger compared to older faces and was better for pain-indicative AUs compared to other AUs. At the moment, automatic analyses of genuine facial expressions of pain may qualify at best as semiautomatic systems, which require further validation by human observers before they can be used to validly assess facial expressions of pain.
Identifiants
pubmed: 35069957
doi: 10.1155/2022/6635496
pmc: PMC8767386
doi:
Types de publication
Journal Article
Research Support, Non-U.S. Gov't
Langues
eng
Sous-ensembles de citation
IM
Pagination
6635496Informations de copyright
Copyright © 2022 Stefan Lautenbacher et al.
Déclaration de conflit d'intérêts
The authors declare that there are no conflicts of interest.
Références
Eur J Pain. 2018 Oct;22(9):1617-1627
pubmed: 29756314
Pain. 2012 Feb;153(2):350-358
pubmed: 22112930
Pain Manag Nurs. 2015 Oct;16(5):692-700
pubmed: 25962546
Pain. 2014 Oct;155(10):1927-1931
pubmed: 24997352
Front Psychol. 2014 Feb 04;5:30
pubmed: 24550859
Curr Alzheimer Res. 2017;14(5):486-500
pubmed: 27335035
Psychol Sci Public Interest. 2019 Jul;20(1):1-68
pubmed: 31313636
Pain. 2019 Mar;160(3):535-549
pubmed: 30335682
Eur J Pain. 2014 Jul;18(6):813-23
pubmed: 24174396
Pain Res Manag. 2014 Jan-Feb;19(1):15-22
pubmed: 24432350
Pain Manag Nurs. 2019 Oct;20(5):404-417
pubmed: 31610992
Psychol Bull. 2011 Nov;137(6):910-939
pubmed: 21639605
IEEE Trans Pattern Anal Mach Intell. 2021 Jun;43(6):1815-1831
pubmed: 31825861
Eur J Pain. 2020 Jan;24(1):192-208
pubmed: 31487411
Pain. 2006 Jun;122(3):282-288
pubmed: 16545907
J Am Med Dir Assoc. 2003 Jan-Feb;4(1):9-15
pubmed: 12807591
Int J Geriatr Psychiatry. 2018 Oct;33(10):1346-1354
pubmed: 29961948
Pain. 1992 Dec;51(3):297-306
pubmed: 1491857
Curr Alzheimer Res. 2017;14(5):501-505
pubmed: 27335044
Pain Med. 2016 Apr;17(4):685-91
pubmed: 26361368
Gerontology. 2009;55(6):674-82
pubmed: 19696473
BMC Geriatr. 2017 Jan 26;17(1):33
pubmed: 28125956
Behav Brain Sci. 2002 Aug;25(4):439-55; discussion 455-88
pubmed: 12879700
Clin J Pain. 2007 Jul-Aug;23(6):506-10
pubmed: 17575490
PLoS One. 2019 Oct 17;14(10):e0223905
pubmed: 31622426
Clin Interv Aging. 2018 Jul 16;13:1245-1258
pubmed: 30038491