Deep Learning-Based Grimace Scoring Is Comparable to Human Scoring in a Mouse Migraine Model

Chih-Yi Chiang; Yueh-Peng Chen; Hung-Ruei Tzeng; Man-Hsin Chang; Lih-Chu Chiou; Yu-Cheng Pei

doi:10.3390/jpm12060851

Deep Learning-Based Grimace Scoring Is Comparable to Human Scoring in a Mouse Migraine Model

J Pers Med. 2022 May 24;12(6):851. doi: 10.3390/jpm12060851.

Authors

Chih-Yi Chiang¹, Yueh-Peng Chen^{2

3}, Hung-Ruei Tzeng⁴, Man-Hsin Chang⁴, Lih-Chu Chiou^{4

5

6}, Yu-Cheng Pei^{1

3

7

8}

Affiliations

¹ Department of Physical Medicine and Rehabilitation, Chang Gung Memorial Hospital at Linkou, Taoyuan 33305, Taiwan.
² Center for Artificial Intelligence in Medicine, Chang Gung Memorial Hospital at Linkou, Taoyuan 33305, Taiwan.
³ Master of Science Degree Program in Innovation for Smart Medicine, College of Management, Chang Gung University, Taoyuan 33302, Taiwan.
⁴ Graduate Institute of Pharmacology, College of Medicine, National Taiwan University, Taipei 10617, Taiwan.
⁵ Graduate Institute of Brain and Mind Sciences, College of Medicine, National Taiwan University, Taipei 10617, Taiwan.
⁶ Graduate Institute of Acupuncture Science, China Medical University, Taichung 40447, Taiwan.
⁷ School of Medicine, Chang Gung University, Taoyuan 33302, Taiwan.
⁸ Center of Vascularized Tissue Allograft, Chang Gung Memorial Hospital at Linkou, Taoyuan 33305, Taiwan.

Abstract

Pain assessment is essential for preclinical and clinical studies on pain. The mouse grimace scale (MGS), consisting of five grimace action units, is a reliable measurement of spontaneous pain in mice. However, MGS scoring is labor-intensive and time-consuming. Deep learning can be applied for the automatic assessment of spontaneous pain. We developed a deep learning model, the DeepMGS, that automatically crops mouse face images, predicts action unit scores and total scores on the MGS, and finally infers whether pain exists. We then compared the performance of DeepMGS with that of experienced and apprentice human scorers. The DeepMGS achieved an accuracy of 70-90% in identifying the five action units of the MGS, and its performance (correlation coefficient = 0.83) highly correlated with that of an experienced human scorer in total MGS scores. In classifying pain and no pain conditions, the DeepMGS is comparable to the experienced human scorer and superior to the apprentice human scorers. Heatmaps generated by gradient-weighted class activation mapping indicate that the DeepMGS accurately focuses on MGS-relevant areas in mouse face images. These findings support that the DeepMGS can be applied for quantifying spontaneous pain in mice, implying its potential application for predicting other painful conditions from facial images.

Keywords: deep machine learning; facial expression; migraine animal model; mouse grimace scale; spontaneous pain.

Grants and funding

This research is supported by the Maintenance Project of the Center for Artificial Intelligence in Medicine (Grant CLRPG3H0013 to Y.P.C. and Y.C.P.) from Chang Gung Memorial Hospital, research grants (CMRPG3K2151-2, CMRPG3K0231-2, and BMRPB67 to Y.P.C. and Y.C.P.) from Chang Gung Medical Foundation, the research grant from the Ministry of Science and Technology, Taiwan (MOST 108-2320-B-002-029-MY3 to L.C.C.), and the Innovative Research Grant from the National Health Research Institutes, Taiwan (NHRI-EX111-11114NI to L.C.C.).