Pain is a symptom that can neither be felt nor seen. This has posed major problems for doctors involved in pain management since there are no accurate objective measures to establish the extent of pain a patient is suffering from. The commonly used procedure is self-reporting, where the doctor asks the patient to rate the level of pain he or she is experiencing on a scale of one to ten – this method has obvious discrepancies, especially when it comes to children.
Relying on Facial Expressions for Accurate Pain Assessment
Researchers at the UC San Diego School of Medicine are hopeful that a fusion of artificial intelligence and facial recognition could help in developing a more accurate system of detecting the extent of pain.
The Journal of Pediatrics published a study this week which suggested that such a technological advancement could indeed be precisely predictive of the level of pain a patient suffers from. The researchers used specially designed software to register the facial expressions of 50 children between the ages of 5 and 18. The children enrolled in the study were recovering from laparoscopic appendectomies. Analysis from the video data, combined with clinical reports by caregivers was used to determine the extent of pain for each child. The system was found to produce ‘good-to-excellent’ results in accurately assessing pain levels.
The researchers stated in the study that, “Current pain assessment methods in youth are suboptimal and vulnerable to bias and under-recognition of clinical pain. Facial expressions are a sensitive, specific biomarker of the presence and severity of pain, and computer vision (CV) and machine-learning (ML) techniques enable reliable, valid measurement of pain-related facial expressions from video.”
The video analysis techniques used in the study are based on the Facial Action Coding System (FACS). This visual technique uses 46 anatomically-based component movements to measure even the slightest change in facial expressions.
Besides solving the problem of misinterpreting pain levels, the system also deals with the inaccuracy of pain assessment protocols. Nursing staff is incapable of checking in on the patient after every minute and might miss certain important cues. However, a facial recognition system provides constant surveillance, detecting even the slightest sense of pain via facial appearance. Moreover, the system can also determine the intervals between periodic episodes of pain, which can help improve the relief measures.