Emerging Clinical Technology: Application of Machine Learning to Chronic Pain Assessments Based on Emotional Body Maps

Neurotherapeutics. 2020 Jul;17(3):774-783. doi: 10.1007/s13311-020-00886-7.

Abstract

Depression and anxiety co-occur with chronic pain, and all three are thought to be caused by dysregulation of shared brain systems related to emotional processing associated with body sensations. Understanding the connection between emotional states, pain, and bodily sensations may help understand chronic pain conditions. We developed a mobile platform for measuring pain, emotions, and associated bodily feelings in chronic pain patients in their daily life conditions. Sixty-five chronic back pain patients reported the intensity of their pain, 11 emotional states, and the corresponding body locations. These variables were used to predict pain 2 weeks later. Applying machine learning, we developed two predictive models of future pain, emphasizing interpretability. One model excluded pain-related features as predictors of future pain, and the other included pain-related predictors. The best predictors of future pain were interactive effects of (a) body maps of fatigue with negative affect and (b) positive affect with past pain. Our findings emphasize the contribution of emotions, especially emotional experience felt in the body, to understanding chronic pain above and beyond the mere tracking of pain levels. The results may contribute to the generation of a novel artificial intelligence framework to help in the development of better diagnostic and therapeutic approaches to chronic pain.

Keywords: Chronic pain; bodily sensation map; emotions; interoception; low back pain; pain assessment.

Publication types

  • Randomized Controlled Trial

MeSH terms

  • Adult
  • Aged
  • Back Pain / diagnosis*
  • Back Pain / psychology
  • Biomedical Technology / methods
  • Biomedical Technology / trends*
  • Chronic Pain / diagnosis*
  • Chronic Pain / psychology
  • Emotions* / physiology
  • Female
  • Humans
  • Machine Learning / trends*
  • Male
  • Middle Aged
  • Pain Measurement / methods
  • Pain Measurement / psychology
  • Pain Measurement / trends*
  • Young Adult