Pain is among the most pervasive and universal forms of human distress. Pain typically is measured by patient self-report or clinician impressions, either through clinical interview or the visual analog scale. However, self- reported pain is difficult to interpret and in some circumstances not possible to obtain [Hadjistavropoulos et al., 2002]. To improve the standard of care and advance pain assessment, monitoring, and intervention, we propose (1) a savvy technology based on automatic facial, head, and body movement analysis for a reliable and valid assessment of the occurrence and intensity associated with five causes of acute and chronic low back pain (LBP); (2) inform our understanding of psychosocial and behavioral indicators of chronic LBP to develop new means to prevent chronic LBP. Participants' face, head, and body movement will be recorded during clinical assessment using two synchronized high-definition digital video cameras during extension, flexion, and rotation movements. The obtained video-recordings, taken during a first visit to the clinic and 3 follow-up visits after treatment, will be used for the development of automatic measures of the occurrence and intensity of pain. To investigate the generalizability of the proposed automatic measures, we will explicitly train and test the proposed classifiers on five different types of acute and chronic LBP. To do so, face, head, and body movement will be automatically tracked using our fully- automatic methods. The tracking results will be used to train end-to-end deep-leaning based classifiers to automatically measure the occurrence and intensity of LBP. To investigate the validity of the proposed classifiers, we will compare automated measurement to the patient- and clinician- rated visual analog scale, brief pain inventory, and continuous observer ratings of pain intensity from the video recordings. MANOVA will be used to quantify the relationship between the individual modalities and their combination for the measurement of the occurrence and intensity of the five LBP conditions and for chronic and acute conditions. To inform our understanding of how LBP evolves into a chronic form, we will use Ecological Momentary Assessment (EMA) to collect behavioral and contextual information beyond the video-recordings and pain scores' assessments. Participants will be monitored for 6 months after treatment, at a frequency of 7 consecutive days per month (1 week per month), and 4 prompts per day, to identify those who evolved to chronic LBP. EMA measures will be used to investigate whether pain intensity differs by psychosocial and behavioral factors both between and within LBP groups as well as whether psychosocial and behavioral factors are associated with the development of chronic LBP.

Public Health Relevance

Pain typically is measured by patient self-report, but self-reported pain is difficult to interpret and may be impaired or not possible to obtain. We will use clinically well-characterized data to: (1) develop an automatic multimodal method for the detection of occurrence and intensity of pain in five of the most common and disabling acute and chronic LBP conditions; (2) use EMA to characterize early point psychosocial and behavioral signs that acute pain is likely to become chronic.

National Institute of Health (NIH)
National Institute of Nursing Research (NINR)
Research Project (R01)
Project #
Application #
Study Section
Biomedical Computing and Health Informatics Study Section (BCHI)
Program Officer
Hamlet, Michelle R
Project Start
Project End
Budget Start
Budget End
Support Year
Fiscal Year
Total Cost
Indirect Cost
Carnegie-Mellon University
Biostatistics & Other Math Sci
Schools of Arts and Sciences
United States
Zip Code