Abstract
Chronic musculoskeletal pain condition often shows poor correlations between tissue abnormalities and clinical pain. Therefore, classification of pain conditions like chronic low back pain, osteoarthritis, and fibromyalgia depends mostly on self report and less on objective findings like X-ray or magnetic resonance imaging (MRI) changes. However, recent advances in structural and functional brain imaging have identified brain abnormalities in chronic pain conditions that can be used for illness classification. Because the analysis of complex and multivariate brain imaging data is challenging, machine learning techniques have been increasingly utilized for this purpose. The goal of machine learning is to train specific classifiers to best identify variables of interest on brain MRIs (i.e., biomarkers). This report describes classification techniques capable of separating MRI-based brain biomarkers of chronic pain patients from healthy controls with high accuracy (70–92%) using machine learning, as well as critical scientific, practical, and ethical considerations related to their potential clinical application. Although self-report remains the gold standard for pain assessment, machine learning may aid in the classification of chronic pain disorders like chronic back pain and fibromyalgia as well as provide mechanistic information regarding their neural correlates.
Similar content being viewed by others
References
Papers of particular interest, published recently, have been highlighted as: • Of importance •• Of major importance
Institute of Medicine. Relieving pain in America: a blueprint for transforming prevention, care, education, and research. Washington, DC: National Academies Press; 2011.
Wolfe F, Clauw DJ, Fitzcharles MA, Goldenberg DL, Katz RS, Mease P, et al. The American College of Rheumatology preliminary diagnostic criteria for fibromyalgia and measurement of symptom severity. Arthritis Care Res. 2010;62(5):600–10. doi:10.1002/acr.20140.
Shamir L, Ling SM, Scott Jr WW, Bos A, Orlov N, Macura TJ, et al. Knee x-ray image analysis method for automated detection of osteoarthritis. IEEE Trans Biomed Eng. 2009;56(2):407–15. doi:10.1109/TBME.2008.2006025.
Kirkwood RN, Resende RA, Magalhaes CM, Gomes HA, Mingoti SA, Sampaio RF. Application of principal component analysis on gait kinematics in elderly women with knee osteoarthritis. Rev Bras Fis. 2011;15(1):52–8.
Frize M, Ogungbemile A. Estimating rheumatoid arthritis activity with infrared image analysis. Stud Health Technol Inform. 2012;180:594–8.
Frampton D, Kerr J, Harrison TJ, Kellam P. Assessment of a 44 gene classifier for the evaluation of chronic fatigue syndrome from peripheral blood mononuclear cell gene expression. PLoS One. 2011;6(3):e16872. doi:10.1371/journal.pone.0016872.
Apkarian AV, Hashmi JA, Baliki MN. Pain and the brain: specificity and plasticity of the brain in clinical chronic pain. Pain. 2011;152(3 Suppl):S49–64. doi:10.1016/j.pain.2010.11.010.
• Robinson ME, Staud R, Price DD. Pain measurement and brain activity: will neuroimages replace pain ratings? J Pain: Off J Am Pain Soc. 2013;14(4):323–7. doi:10.1016/j.jpain.2012.05.007. This critical review summarized the limitations of neuroimaging to replace pain self-report. It also emphasized that neuroimaging will never replace pain self-report except in individuals with mental conditions or limited levels of consciousness.
Wortolowska K. How neuroimaging can help us to visualise and quantify pain? Eur J Pain Suppl. 2011;5(S2):323–7.
Bushnell MC, Ceko M, Low LA. Cognitive and emotional control of pain and its disruption in chronic pain. Nat Rev Neurosci. 2013;14(7):502–11. doi:10.1038/nrn3516.
Brekelmans MP, Fens N, Brinkman P, Bos LD, Sterk PJ, Tak PP, et al. Smelling the diagnosis: the electronic nose as diagnostic tool in inflammatory arthritis. A case-reference study. PloS One. 2016;11(3):e0151715. doi:10.1371/journal.pone.0151715.
Rizzo G, Raffeiner B, Coran A, Ciprian L, Fiocco U, Botsios C, et al. Pixel-based approach to assess contrast-enhanced ultrasound kinetics parameters for differential diagnosis of rheumatoid arthritis. J Med Imaging (Bellingham). 2015;2(3):034503. doi:10.1117/1.JMI.2.3.034503.
Mackey SC. Central neuroimaging of pain. J Pain: Off J Am Pain Soc. 2013;14(4):328–31. doi:10.1016/j.jpain.2013.01.001.
Borsook D, Becerra L, Hargreaves R. Biomarkers for chronic pain and analgesia. Part 1: the need, reality, challenges, and solutions. Discov Med. 2011;11(58):197–207.
Borsook D, Becerra L, Hargreaves R. Biomarkers for chronic pain and analgesia. Part 2: how, where, and what to look for using functional imaging. Discov Med. 2011;11(58):209–19.
• Callan D, Mills L, Nott C, England R, England S. A tool for classifying individuals with chronic back pain: using multivariate pattern analysis with functional magnetic resonance imaging data. PloS One. 2014;9(6):e98007. doi:10.1371/journal.pone.0098007. This study employed supervised machine learning techniques, specifically sparse logistic regression, to train a classifier of chronic back pain based on contrast images using a leave-one-out cross-validation procedure. It correctly classified 92.3% of the chronic pain group and 92.3% of the normal controls.
Lu HC, Hsieh JC, Lu CL, Niddam DM, Wu YT, Yeh TC, et al. Neuronal correlates in the modulation of placebo analgesia in experimentally-induced esophageal pain: a 3T-fMRI study. Pain. 2010;148(1):75–83. doi:10.1016/j.pain.2009.10.012.
Pustilnik AC. Imaging brains, changing minds: how pain neuroimaging can inform the law. Alabama Law Rev. 2015;66(5).
• Letzen JE, Boissoneault J, Sevel LS, Robinson ME. Test-retest reliability of pain-related functional brain connectivity compared to pain self-report. Pain. 2016;157(3):546–51. doi:10.1097/j.pain.0000000000000356. This study examined the test-retest reliability for functional connectivity MRI (fcMRI) of pain-related brain regions and self-reported pain. Intraclass correlations coefficients for fcMRI values varied widely (range = −.174–.766), whereas intraclass correlations coefficients for VAS scores ranged from .906 to .947. Overall, self-reported pain was more reliable than fcMRI data.
Nielsen CS, Price DD, Vassend O, Stubhaug A, Harris JR. Characterizing individual differences in heat-pain sensitivity. Pain. 2005;119(1–3):65–74. doi:10.1016/j.pain.2005.09.018.
Slesinger D, Archer RP, Duane W. MMPI-2 characteristics in a chronic pain population. Assessment. 2002;9(4):406–14.
Takekawa KS, Goncalves JS, Moriguchi CS, Coury HJ, Sato TO. Can a self-administered questionnaire identify workers with chronic or recurring low back pain? Ind Health. 2015;53(4):340–5. doi:10.2486/indhealth.2014-0241.
•• Robinson ME, O’Shea AM, Craggs JG, Price DD, Letzen JE, Staud R. Comparison of machine classification algorithms for fibromyalgia: neuroimages versus self-report. J Pain: Off J Am Pain Soc. 2015;16(5):472–7. doi:10.1016/j.jpain.2015.02.002. Separate models representing brain volumes, mood ratings, and pain intensity ratings were estimated across several ML algorithms. Structural magnetic resonance imaging data from fibromyalgia patients and healthy controls and self-report data of pain intensity and mood were used. Classification accuracy of brain volumes ranged from 53 to 76%, whereas mood and pain intensity ratings ranged from 79 to 96% and 83 to 96%, respectively. Overall, models derived from self-report data outperformed neuroimaging models.
Friend R, Bennett RM. Distinguishing fibromyalgia from rheumatoid arthritis and systemic lupus in clinical questionnaires: an analysis of the revised Fibromyalgia Impact Questionnaire (FIQR) and its variant, the Symptom Impact Questionnaire (SIQR), along with pain locations. Arthritis Res Ther. 2011;13(2):R58. doi:10.1186/ar3311.
Perrot S, Bouhassira D, Fermanian J, Cercle d’Etude de la Douleur en R. Development and validation of the Fibromyalgia Rapid Screening Tool (FiRST). Pain. 2010;150(2):250–6. doi:10.1016/j.pain.2010.03.034.
Cortes C, Vapnik V. Support-vector networks. Mach Learn. 1995;20(3):273–97.
Pontil M, Verri A. Properties of support vector machines. Neural Comput. 1998;10(4):955–74.
Bagarinao E, Johnson KA, Martucci KT, Ichesco E, Farmer MA, Labus J, et al. Preliminary structural MRI based brain classification of chronic pelvic pain: a MAPP network study. Pain. 2014;155(12):2502–9. doi:10.1016/j.pain.2014.09.002.
Ung H, Brown JE, Johnson KA, Younger J, Hush J, Mackey S. Multivariate classification of structural MRI data detects chronic low back pain. Cereb Cortex. 2014;24(4):1037–44. doi:10.1093/cercor/bhs378.
Labus JS, Van Horn JD, Gupta A, Alaverdyan M, Torgerson C, Ashe-McNalley C, et al. Multivariate morphological brain signatures predict patients with chronic abdominal pain from healthy control subjects. Pain. 2015;156(8):1545–54. doi:10.1097/j.pain.0000000000000196.
Le Cessie S, Van Houwelingen JC. Ridge estimators in logistic regression. Appl Stat. 1992;41(1):191–201.
Langley P, Iba W, Thompson K. An analysis of Bayesian classifiers. AAAI. 1992;90:223–8.
Aha DW, Kibler D, Albert MK. Instance-based learning algorithms. Mach Learn. 1991;6(1):37–66. doi:10.1023/A:1022689900470.
Arora R. Comparative analysis of classification algorithms on different datasets using WEKA. Int J Comput Appl. 2012;54(13):21–5.
Quinlan JR. C4.5 programs for machine learning. San Mateo: Morgan Kaufmann; 1992.
Tibshirani R. Regression shrinkage and selection via the lasso. J Roy Stat Soc B Met. 1996;58(1):267–88.
Kononenko I. Machine learning for medical diagnosis: history, state of the art and perspective. Artif Intell Med. 2001;23(1):89–109.
Nair SS, French RM, Laroche D, Thomas E. The application of machine learning algorithms to the analysis of electromyographic patterns from arthritic patients. IEEE Trans Neural Syst Rehabil Eng. 2010;18(2):174–84. doi:10.1109/TNSRE.2009.2032638.
Mamoshina P, Vieira A, Putin E, Zhavoronkov A. Applications of deep learning in biomedicine. Mol Pharm. 2016;13(5):1445–54. doi:10.1021/acs.molpharmaceut.5b00982.
Schnack HG, Kahn RS. Detecting neuroimaging biomarkers for psychiatric disorders: sample size matters. Front Psych. 2016;7:50. doi:10.3389/fpsyt.2016.00050.
Siegel JS, Ramsey LE, Snyder AZ, Metcalf NV, Chacko RV, Weinberger K, et al. Disruptions of network connectivity predict impairment in multiple behavioral domains after stroke. Proc Natl Acad Sci U S A. 2016. doi:10.1073/pnas.1521083113.
Woo CW, Wager TD. Neuroimaging-based biomarker discovery and validation. Pain. 2015;156(8):1379–81. doi:10.1097/j.pain.0000000000000223.
Deo RC. Machine learning in medicine. Circulation. 2015;132(20):1920–30. doi:10.1161/CIRCULATIONAHA.115.001593.
• Robinson M, Boissoneault J, Sevel L, Letzen J, Staud R. The effect of base rate on the predictive value of brain biomarkers. J Pain. 2016;17(6):637–41. doi:10.1016/j.jpain.2016.01.476. Results of this study strongly suggest that many proposed brain biomarkers perform quite poorly when realistic illness prevalence base rates are taken into account.
Lueken U, Zierhut KC, Hahn T, Straube B, Kircher T, Reif A, et al. Neurobiological markers predicting treatment response in anxiety disorders: a systematic review and implications for clinical application. Neurosci Biobehav Rev. 2016;66:143–62. doi:10.1016/j.neubiorev.2016.04.005.
Buckalew N, Haut MW, Aizenstein H, Morrow L, Perera S, Kuwabara H, et al. Differences in brain structure and function in older adults with self-reported disabling and nondisabling chronic low back pain. Pain Med. 2010;11(8):1183–97. doi:10.1111/j.1526-4637.2010.00899.x.
Geha PY, Baliki MN, Harden RN, Bauer WR, Parrish TB, Apkarian AV. The brain in chronic CRPS pain: abnormal gray-white matter interactions in emotional and autonomic regions. Neuron. 2008;60(4):570–81. doi:10.1016/j.neuron.2008.08.022.
Gustin SM, Peck CC, Cheney LB, Macey PM, Murray GM, Henderson LA. Pain and plasticity: is chronic pain always associated with somatosensory cortex activity and reorganization? J Neurosci: Off J Soc Neurosci. 2012;32(43):14874–84. doi:10.1523/JNEUROSCI.1733-12.2012.
Greve DN. An absolute beginner’s guide to surface- and voxel-based morphometric analysis. ISMRM 19th Annual Meeting & Exhibition. 2011;7–13.
Ashburner J, Friston K. Morphometry. In: Ashburner J, editor. Human brain function. 2nd ed. Cambridge: Academic; 2004. p. 707–22.
Fischl B, Liu A, Dale AM. Automated manifold surgery: constructing geometrically accurate and topologically correct models of the human cerebral cortex. IEEE Trans Med Imaging. 2001;20(1):70–80. doi:10.1109/42.906426.
Fischl B, Salat DH, Busa E, Albert M, Dieterich M, Haselgrove C, et al. Whole brain segmentation: automated labeling of neuroanatomical structures in the human brain. Neuron. 2002;33(3):341–55.
Baliki MN, Schnitzer TJ, Bauer WR, Apkarian AV. Brain morphological signatures for chronic pain. PLoS One. 2011;6(10):e26010. doi:10.1371/journal.pone.0026010.
Apkarian AV, Sosa Y, Sonty S, Levy RM, Harden RN, Parrish TB, et al. Chronic back pain is associated with decreased prefrontal and thalamic gray matter density. J Neurosci: Off J Soc Neurosci. 2004;24(46):10410–5. doi:10.1523/JNEUROSCI.2541-04.2004.
Schmidt-Wilcke T, Luerding R, Weigand T, Jurgens T, Schuierer G, Leinisch E, et al. Striatal grey matter increase in patients suffering from fibromyalgia—a voxel-based morphometry study. Pain. 2007;132 Suppl 1:S109–16. doi:10.1016/j.pain.2007.05.010.
Burgmer M, Gaubitz M, Konrad C, Wrenger M, Hilgart S, Heuft G, et al. Decreased gray matter volumes in the cingulo-frontal cortex and the amygdala in patients with fibromyalgia. Psychosom Med. 2009;71(5):566–73. doi:10.1097/PSY.0b013e3181a32da0.
Puri BK, Agour M, Gunatilake KD, Fernando KA, Gurusinghe AI, Treasaden IH. Reduction in left supplementary motor area grey matter in adult female fibromyalgia sufferers with marked fatigue and without affective disorder: a pilot controlled 3-T magnetic resonance imaging voxel-based morphometry study. J Int Med Res. 2010;38(4):1468–72.
Robinson ME, Craggs JG, Price DD, Perlstein WM, Staud R. Gray matter volumes of pain-related brain areas are decreased in fibromyalgia syndrome. J Pain: Off J Am Pain Soc. 2011;12(4):436–43. doi:10.1016/j.jpain.2010.10.003.
Sundermann B, Burgmer M, Pogatzki-Zahn E, Gaubitz M, Stuber C, Wessolleck E, et al. Diagnostic classification based on functional connectivity in chronic pain: model optimization in fibromyalgia and rheumatoid arthritis. Acad Radiol. 2014;21(3):369–77. doi:10.1016/j.acra.2013.12.003.
•• López-Solà M, Woo CW, Pujol J, Deus J, Harrison BJ, Monfort J, Wager TD. Towards a neurophysiological signature for fibromyalgia. Pain. 2016. A ‘Multisensory’ classifier trained on non-painful sensory stimulation revealed a brain signature for FM with high sensitivity and specificity. This study characterized individual FM patients based on symptom-related brain features.
Detre JA, Wang J. Technical aspects and utility of fMRI using BOLD and ASL. Clin Neurophysiol. 2002;113(5):621–34.
Howseman AM, Bowtell RW. Functional magnetic resonance imaging: imaging techniques and contrast mechanisms. Philos Trans R Soc Lond B Biol Sci. 1999;354(1387):1179–94. doi:10.1098/rstb.1999.0473.
Chen JJ, Jann K, Wang DJ. Characterizing resting-state brain function using arterial spin labeling. Brain Connect. 2015;5(9):527–42. doi:10.1089/brain.2015.0344.
Smith SM, Jenkinson M, Woolrich MW, Beckmann CF, Behrens TE, Johansen-Berg H, et al. Advances in functional and structural MR image analysis and implementation as FSL. NeuroImage. 2004;23 Suppl 1:S208–19. doi:10.1016/j.neuroimage.2004.07.051.
Giesecke T, Gracely RH, Grant MA, Nachemson A, Petzke F, Williams DA, et al. Evidence of augmented central pain processing in idiopathic chronic low back pain. Arthritis Rheum. 2004;50(2):613–23. doi:10.1002/art.20063.
Wager TD, Atlas LY, Lindquist MA, Roy M, Woo CW, Kross E. An fMRI-based neurologic signature of physical pain. N Engl J Med. 2013;368(15):1388–97.
Acknowledgements
This work was supported by NIH grant R01 NR014049-01
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflict of Interest
The authors declare that they have no conflict of interest.
Human and Animal Rights and Informed Consent
This article does not contain any studies with human or animal subjects performed by any of the authors.
Additional information
This article is part of the Topical Collection on Chronic Pain
Rights and permissions
About this article
Cite this article
Boissoneault, J., Sevel, L., Letzen, J. et al. Biomarkers for Musculoskeletal Pain Conditions: Use of Brain Imaging and Machine Learning. Curr Rheumatol Rep 19, 5 (2017). https://doi.org/10.1007/s11926-017-0629-9
Published:
DOI: https://doi.org/10.1007/s11926-017-0629-9