Related Experiment Video
Updated: Aug 26, 2025
09:16
Multi-Modal Signals for Analyzing Pain Responses to Thermal and Electrical Stimuli
Published on: April 5, 2019
10.9K
Preliminary study: quantification of chronic pain from physiological data
Zhuowei Cheng1, Franklin Ly2, Tyler Santander3
1Departments of Computer Science.
Pain Reports
|October 10, 2022
Summary
This study shows that physiological sensors can measure chronic pain. Researchers used sensors to track pulse, temperature, and motion, correlating these with reported pain levels.
Area of Science:
- Biomedical Engineering
- Pain Medicine
- Wearable Technology
Background:
- Chronic pain quantification remains a challenge.
- Previous studies quantified acute pain using physiological sensors.
Purpose of the Study:
- To determine if physiological sensors can measure chronic pain.
- To develop a method for quantifying chronic pain using sensor data.
Main Methods:
- Collected physiological data (pulse, temperature, motion) from chronic pain patients.
- Extracted features from sensor data and applied machine learning models (linear models, random forest).
- Validated models using cross-validation and permutation testing.
Main Results:
- Individual models achieved intraclass correlation coefficients (ICCs) from 0.46 to 0.75.
- Population-level model using random forest achieved an ICC of 0.58.
- Bland-Altman analysis indicated systematic overestimation of low pain scores and underestimation of high pain scores.
Conclusions:
- First demonstration correlating physiological data with chronic pain in individuals and populations.
- Further research is needed to develop a "chronic pain meter" for clinical use.
