Fully automated deep learning models with smartphone applicability for prediction of pain using the Feline Grimace Scale

Abstract

This study used deep neural networks and machine learning models to predict facial landmark positions and pain scores using the Feline Grimace Scale© (FGS). A total of 3447 face images of cats were annotated with 37 landmarks. Convolutional neural networks (CNN) were trained and selected according to size, prediction time, predictive performance (normalized root mean squared error, NRMSE) and suitability for smartphone technology. Geometric descriptors (n = 35) were computed. XGBoost models were trained and selected according to predictive performance (accuracy; mean square error, MSE). For prediction of facial landmarks, the best CNN model had NRMSE of 16.76% (ShuffleNetV2). For prediction of FGS scores, the best XGBoost model had accuracy of 95.5% and MSE of 0.0096. Models showed excellent predictive performance and accuracy to discriminate painful and non-painful cats. This technology can now be used for the development of an automated, smartphone application for acute pain assessment in cats.

The grimace scale: a useful tool for assessing pain in laboratory animals

“…In addition, attempting to extrapolate certain types of automated analytical models under different conditions, such as variations in administration methods or different drugs, has resulted in low accuracy when tested with actual test data [ 82 ]. Other CNNs in use include You Only Look Once (YOLOv3, YOLOv5), Contrastive Language-Image Pretraining (CLIP), Matlab 2016b, ShuffleNetV2, EfficentNetB0, and MobileNetV3, among others [ 85 , 86 , 87 , 88 , 89 ]. With these limitations in mind, it is currently challenging to achieve a perfect assessment of animal pain using automated technology.…”

Feline Aortic Thromboembolism: recent advances and future prospects

“…1,2 can readily be managed with short-acting opioids, which allows for further investigation and stabilization, and client communication. After initial stabilization, and ideally in the light of objective measures (eg, Feline Grimace Scale), [20][21][22] additional sedation, analgesia and anxiolytics may be needed.…”

From facial expressions to algorithms: a narrative review of animal pain recognition technologies

“…This has led to the categorization of pain labeling methods in animal APR into behavior-based or stimulus-based annotations ( 90 ). The former relies solely on observed behaviors and is typically assessed by human experts ( 5 , 6 , 97 , 99 , 104 – 106 ). In contrast, the latter determines the ground truth based on whether the data were recorded during an ongoing stimulus or not ( 5 , 10 , 49 , 76 , 94 – 96 , 99 , 100 , 107 – 109 ).…”

“…Another notable DL model is the deep recurrent video model used by Broomé et al ( 96 , 98 ), which utilizes a ConvLSTM layer to analyze spatial and temporal features simultaneously, yielding better results in spatiotemporal representations. Steagall et al ( 106 ) and Martvel et al ( 114 ) introduced a landmark detection CNN-based model to predict facial landmark positions and pain scores based on the manually annotated FGS.…”