Towards a Physiology-Based Measure of Pain: Patterns of Human Brain Activity Distinguish Painful from Non-Painful Thermal Stimulation
2011
Using Brain Activity to Measure Pain
Sample size: 24
publication
Evidence: high
Author Information
Author(s): Justin E. Brown, Neil Chatterjee, Jarred Younger, Sean Mackey
Primary Institution: Stanford University
Hypothesis
An SVM trained on fMRI data can assess pain in the absence of self-report.
Conclusion
The study demonstrates that fMRI with SVM learning can accurately assess pain without requiring communication from the person being tested.
Supporting Evidence
- The SVM model was 81% accurate at distinguishing painful from non-painful stimuli.
- Accuracy increased to 84% when excluding the most difficult stimuli.
- Performance was primarily affected by activity in pain-processing regions of the brain.
Takeaway
Scientists can use brain scans to tell if someone is in pain, even if they can't say it.
Methodology
The study used fMRI and SVM to classify brain activity patterns in response to painful and non-painful thermal stimuli.
Limitations
The study's findings may not generalize to all types of pain or populations.
Participant Demographics
16 participants, average age 22.7 years, 10 men and 6 women.
Statistical Information
P-Value
p<0.00002
Statistical Significance
p<0.0000001
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website