Application of deep learning-based automated pain assessment in critically ill patients

A recent study published in Frontiers in Medicine established the deep learning-based pain classifier based on facial expression focusing on the area nearby eyebrow. The accuracy to detect tense/grimacing and grimacing were ~80 and 90%, respectively. This study, entitled ‘Deep Learning-Based Pain Classifier Based on the Facial Expression in Critically Ill Patients‘, indicates a real-world application of AI-based pain assessment based on the facial expression in ICU.

This research was approved by the Institutional Review Board approval of the Taichung Veterans General Hospital (Taiwan). The researchers conducted this prospective study by enrolling patients who were admitted to medical and surgical ICUs at TCVGH, between 2020-Nov and 2021-Nov. Thee Critical-Care Pain Observation Tool (CPOT) has been developed to grade the pain through assessing behavior alternations, such as facial expressions, among critically. In detail, a score of 0 is given if there is no observed muscle tension in the face, and the score of 1 is composed of a tensed muscle contraction. The score of 2 consists of grimacing, which is a contraction of facial muscles.

A total of 341 participants were enrolled, and there were 7,813 qualified videos, of which the number of scores 0, 1, and 2 were 5,717, 1,714, and 382, respectively. To reduce the need for an extremely high number of labeled but unrelated images for learning, this team employed a relation network architecture for the image-based pain classifier. Therefore, they used the data of the 63 participants who had images of all of 0, 1, 2 labeled images. With regard to the video-based pain classifier, the taiwanese scientists employed a many-to-one sequence model given that the output of this study is a one pain grade. Similar to the image-based pain classifier, they used a Siamese network architecture as feature extractors.

Francesc Arcas Ruscalleda