Experimental Brain Research

, Volume 236, Issue 8, pp 2245–2253 | Cite as

Structural brain changes versus self-report: machine-learning classification of chronic fatigue syndrome patients

  • Landrew S. Sevel
  • Jeff Boissoneault
  • Janelle E. Letzen
  • Michael E. Robinson
  • Roland StaudEmail author
Research Article


Chronic fatigue syndrome (CFS) is a disorder associated with fatigue, pain, and structural/functional abnormalities seen during magnetic resonance brain imaging (MRI). Therefore, we evaluated the performance of structural MRI (sMRI) abnormalities in the classification of CFS patients versus healthy controls and compared it to machine learning (ML) classification based upon self-report (SR). Participants included 18 CFS patients and 15 healthy controls (HC). All subjects underwent T1-weighted sMRI and provided visual analogue-scale ratings of fatigue, pain intensity, anxiety, depression, anger, and sleep quality. sMRI data were segmented using FreeSurfer and 61 regions based on functional and structural abnormalities previously reported in patients with CFS. Classification was performed in RapidMiner using a linear support vector machine and bootstrap optimism correction. We compared ML classifiers based on (1) 61 a priori sMRI regional estimates and (2) SR ratings. The sMRI model achieved 79.58% classification accuracy. The SR (accuracy = 95.95%) outperformed both sMRI models. Estimates from multiple brain areas related to cognition, emotion, and memory contributed strongly to group classification. This is the first ML-based group classification of CFS. Our findings suggest that sMRI abnormalities are useful for discriminating CFS patients from HC, but SR ratings remain most effective in classification tasks.


Machine learning Gray matter Self-report Chronic fatigue Classification 



This research was supported by NIH (R01 NR014049).

Supplementary material

221_2018_5301_MOESM1_ESM.docx (69 kb)
Supplementary material 1 (DOCX 69 KB)
221_2018_5301_MOESM2_ESM.docx (20 kb)
Supplementary material 2 (DOCX 20 KB)
221_2018_5301_MOESM3_ESM.docx (12 kb)
Supplementary material 3 (DOCX 12 KB)


  1. Afari N, Buchwald D (2003) Chronic fatigue syndrome: a review. Am J Psychiatry 160:221–236CrossRefPubMedGoogle Scholar
  2. Alpaydin E (2014) Introduction to machine learning. MIT Press, CambridgeGoogle Scholar
  3. Bagarinao E, Johnson KA, Martucci KT et al (2014) Preliminary structural MRI based brain classification of chronic pelvic pain: A MAPP network study. Pain 155:2502–2509. CrossRefPubMedPubMedCentralGoogle Scholar
  4. Baliki MN, Schnitzer TJ, Bauer WR, Apkarian AV (2011) Brain morphological signatures for chronic pain. PLoS One 6:e26010. CrossRefPubMedPubMedCentralGoogle Scholar
  5. Boissoneault J, Letzen J, Lai S, O’Shea A, Craggs J, Robinson ME, Staud R (2016) Abnormal resting state functional connectivity in patients with chronic fatigue syndrome: an arterial spin-labeling fMRI study. Magn Reson Imaging 34:603–608. CrossRefPubMedGoogle Scholar
  6. Boissoneault J, Letzen J, Lai S, Robinson ME, Staud R (2018) Static and dynamic functional connectivity in patients with chronic fatigue syndrome: use of arterial spin labelling fMRI. Clin Physiol Funct Imaging 38:128–137. CrossRefPubMedGoogle Scholar
  7. Cenzer I, Miao Y, Kirby K, Boscardin WJ (2013) Estimating Harrell’s optimism on predictive indices using bootstrap samples. SAS Global Forum 2013Google Scholar
  8. Chen R, Liang FX, Moriya J, Yamakawa J, Sumino H, Kanda T, Takahashi T (2008) Chronic fatigue syndrome and the central nervous system. J Int Med Res 36:867–874. CrossRefPubMedGoogle Scholar
  9. Cohen JF, Korevaar DA, Altman DG, Bruns DE, Gatsonis CA, Hooft L, Irwing L, Levine D, Reitsma JB, de Vet HC, Bossuyt PM (2016) STARD 2015 guidelines for reporting diagnostic accuracy studies: explanation and elaboration. BMJ Open 6:e012799CrossRefPubMedPubMedCentralGoogle Scholar
  10. de Lange FP, Kalkman JS, Bleijenberg G, Hagoort P, van der Meer JW, Toni I (2005) Gray matter volume reduction in the chronic fatigue syndrome. Neuroimage 26:777–781. CrossRefPubMedGoogle Scholar
  11. de Lange FP, Koers A, Kalkman JS, Bleijenberg G, Hagoort P, van der Meer JW, Toni I (2008) Increase in prefrontal cortical volume following cognitive behavioural therapy in patients with chronic fatigue syndrome. Brain 131:2172–2180. CrossRefPubMedGoogle Scholar
  12. Fan R, Chan P, Lin C (2005) LIBSVM: a library for support vector machines. ACM Trans Intell Syst Technol 2:27Google Scholar
  13. Fischl B (2012) FreeSurfer. Neuroimage 62:774–781CrossRefPubMedPubMedCentralGoogle Scholar
  14. Fischl B, Salat DH, Busa E et al (2002) Whole brain segmentation: automated labeling of neuroanatomical structures in the human brain. Neuron 33:341–355CrossRefPubMedGoogle Scholar
  15. Friedman J, Hastie T, Tibshirani R (2009) The elements of statistical learning, 2nd edn. Springer, New YorkGoogle Scholar
  16. Fukuda K, Straus SE, Hickie I, Sharpe MC, Dobbins JG, Komaroff A (1994) The chronic fatigue syndrome: a comprehensive approach to its definition and study. International Chronic Fatigue Syndrome Study Group. Ann Intern Med 121:953–959CrossRefPubMedGoogle Scholar
  17. Gay CW, Robinson ME, Lai S, O’Shea A, Craggs JG, Price DD, Staud R (2016) Abnormal resting-state functional connectivity in patients with chronic fatigue syndrome: results of seed and data-driven analyses. Brain Connect 6:48–56. CrossRefPubMedPubMedCentralGoogle Scholar
  18. Grimes DA, Schulz KF (2005) Refining clinical diagnosis with likelihood ratios. Lancet 365:1500–1505. CrossRefPubMedGoogle Scholar
  19. Gunn WJ, Connell DB, Randall B (1993) Epidemiology of chronic fatigue syndrome: the Centers for Disease Control Study. Ciba Found Symp 173:83–93; (discussion 93–101)PubMedGoogle Scholar
  20. Guyon I, Weston J, Barnhill S, Vapnik V (2002) Gene selection for cancer classification using support vector machines. Mach Learn 46:389–422CrossRefGoogle Scholar
  21. Haufe S, Meinecke F, Gorgen K, Dahne S, Haynes JD, Blankertz B, Biessmann F (2014) On the interpretation of weight vectors of linear models in multivariate neuroimaging. Neuroimage 87:96–110. CrossRefPubMedGoogle Scholar
  22. Hoffstaedter F, Grefkes C, Caspers S et al (2014) The role of anterior midcingulate cortex in cognitive motor control: evidence from functional connectivity analyses. Hum Brain Mapp 35:2741–2753. CrossRefPubMedGoogle Scholar
  23. Hsu C, Chang C, Lin C (2008) A practical guide to support vector classification. BJU Int 101:1396–1400CrossRefGoogle Scholar
  24. Jovicich J, Czanner S, Han X et al (2009) MRI-derived measurements of human subcortical, ventricular and intracranial brain volumes: reliability effects of scan sessions, acquisition sequences, data analyses, scanner upgrade, scanner vendors and field strengths. Neuroimage 46:177–192. CrossRefPubMedPubMedCentralGoogle Scholar
  25. Labus JS, Van Horn JD, Gupta A et al (2015) Multivariate morphological brain signatures predict patients with chronic abdominal pain from healthy control subjects. Pain 156:1545–1554. CrossRefPubMedPubMedCentralGoogle Scholar
  26. Lopez Puga J, Krzywinski M, Altman N (2015) Points of significance: Bayes’ theorem. Nat Methods 12:277–278CrossRefPubMedGoogle Scholar
  27. Mansson KN, Frick A, Boraxbekk CJ et al (2015) Predicting long-term outcome of Internet-delivered cognitive behavior therapy for social anxiety disorder using fMRI and support vector machine learning. Transl Psychiatry 5:e530. CrossRefPubMedPubMedCentralGoogle Scholar
  28. Miller AH, Jones JF, Drake DF, Tian H, Unger ER, Pagnoni G (2014) Decreased basal ganglia activation in subjects with chronic fatigue syndrome: association with symptoms of fatigue. PLoS One 9:e98156. CrossRefPubMedPubMedCentralGoogle Scholar
  29. Nouretdinov I, Costafreda SG, Gammerman A, Chervonenkis A, Vovk V, Vapnik V, Fu CH (2011) Machine learning classification with confidence: application of transductive conformal predictors to MRI-based diagnostic and prognostic markers in depression. Neuroimage 56:809–813. CrossRefPubMedGoogle Scholar
  30. Okada T, Tanaka M, Kuratsune H, Watanabe Y, Sadato N (2004) Mechanisms underlying fatigue: a voxel-based morphometric study of chronic fatigue syndrome. BMC Neurol 4:14CrossRefPubMedPubMedCentralGoogle Scholar
  31. Pereira F, Mitchell T, Botvinick M (2009) Machine learning classifiers and fMRI: a tutorial overview. Neuroimage 45:S199-209. CrossRefPubMedGoogle Scholar
  32. Price DD, Harkins SW (1987) Combined use of visual analogue scales and experimental pain in providing standardized measurement of clinical pain. Clin J Pain 3:1–8CrossRefGoogle Scholar
  33. Puri BK, Counsell SJ, Zaman R, Main J, Collins AG, Hajnal JV, Davey NJ (2002) Relative increase in choline in the occipital cortex in chronic fatigue syndrome. Acta Psychiatr Scand 106:224–226CrossRefPubMedGoogle Scholar
  34. Puri BK, Jakeman PM, Agour M et al (2012) Regional grey and white matter volumetric changes in myalgic encephalomyelitis (chronic fatigue syndrome): a voxel-based morphometry 3 T MRI study. Br J Radiol 85:e270-273. CrossRefGoogle Scholar
  35. Radloff LS (1977) The CES-D Scale: A self-report depression scale for research in the general population. Appl Psychol Meas 1:385–401CrossRefGoogle Scholar
  36. Robinson ME, O’Shea AM, Craggs JG, Price DD, Letzen JE, Staud R (2015) Comparison of machine classification algorithms for fibromyalgia: neuroimages versus self-report. J Pain 16:472–477. CrossRefPubMedPubMedCentralGoogle Scholar
  37. Robinson M, Boissoneault J, Sevel L, Letzen J, Staud R (2016) The effect of base rate on the predictive value of brain biomarkers. J Pain 17:637–641. CrossRefPubMedPubMedCentralGoogle Scholar
  38. Smith GC, Seaman SR, Wood AM, Royston P, White IR (2014) Correcting for optimistic prediction in small data sets. Am J Epidemiol 180:318–324. CrossRefPubMedPubMedCentralGoogle Scholar
  39. Staud R, Mokthech M, Price DD, Robinson ME (2015) Evidence for sensitized fatigue pathways in patients with chronic fatigue syndrome. Pain 156:750–759. CrossRefPubMedPubMedCentralGoogle Scholar
  40. Ung H, Brown JE, Johnson KA, Younger J, Hush J, Mackey S (2014) Multivariate classification of structural MRI data detects chronic low back pain. Cereb Cortex 24:1037–1044. CrossRefPubMedGoogle Scholar
  41. Vincent A, Brimmer DJ, Whipple MO et al (2012) Prevalence, incidence, and classification of chronic fatigue syndrome in Olmsted County, Minnesota, as estimated using the Rochester Epidemiology Project. Mayo Clin Proc 87:1145–1152. CrossRefPubMedPubMedCentralGoogle Scholar
  42. Walsh CG, Ribeiro JD, Franklin JC (2017) Predicting risk of suicide attempts of time through machine learning. Clin Psychol Sci 5:457–469CrossRefGoogle Scholar
  43. Wortinger LA, Endestad T, Melinder AM, Oie MG, Sevenius A, Bruun Wyller V (2016) aberrant resting-state functional connectivity in the salience network of adolescent chronic fatigue syndrome. PLoS One 11:e0159351. CrossRefPubMedPubMedCentralGoogle Scholar

Copyright information

© Springer-Verlag GmbH Germany, part of Springer Nature 2018

Authors and Affiliations

  1. 1.Department of Clinical and Health PsychologyUniversity of FloridaGainesvilleUSA
  2. 2.Department of Medicine, College of MedicineUniversity of FloridaGainesvilleUSA

Personalised recommendations