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

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.

α6GABAA Receptor Positive Modulators Alleviate Migraine-like Grimaces in Mice via Compensating GABAergic Deficits in Trigeminal Ganglia.

Migraine is caused by hyperactivity of the trigeminovascular system, where trigeminal ganglia (TG) play an important role. This hyperactivity might originate from an underfunctional GABAergic system in TG. To investigate this possibility, we adapted a mouse model of migraine by inducing migraine-like grimaces in male mice via repeated injections of nitroglycerin (NTG, 10 mg/kg, i.p.), once every 2 days, for up to 5 sessions. Migraine-like facial pain scores were measured using the mouse grimace scale. Repeated NTG treatments in mice caused significant increases in migraine-like grimaces that were aborted and prevented by two anti-migraine agents sumatriptan and topiramate, respectively. After 5 sessions of NTG injections, the GABA-synthesizing enzyme, 65-kDa glutamate decarboxylase (GAD65), but not the GABA transporter 1 (GAT1) or the α6 subunit-containing GABAA receptors (α6GABAARs), was downregulated in mouse TG tissues. Taking advantage of the unaffected TG α6GABAAR expression in NTG-treated mice, we demonstrated that an α6GABAAR-selective positive allosteric modulator (PAM), DK-I-56-1, exhibited both abortive and prophylactic effects, comparable to those of sumatriptan and topiramate, respectively, in this migraine-mimicking mouse model. The brain-impermeable furosemide significantly prevented the effects of DK-I-56-1, suggesting its peripheral site of action, likely via preventing α6GABAAR modulation in TG. Results suggest that a decreased GABA synthesis caused by the reduced GAD65 expression in TG contributes to the trigeminovascular activation in this repeated NTG-induced migraine-mimicking model and that the unaltered α6GABAARs in TG are potential targets for migraine treatment. Thus, α6GABAAR-selective PAMs are potential anti-migraine agents for both abortive and preventive therapies.