E2 ModeratePreliminaryPEM not requiredCross-SectionalPeer-reviewedReviewed
Structural brain changes versus self-report: machine-learning classification of chronic fatigue syndrome patients.
Sevel, Landrew S, Boissoneault, Jeff, Letzen, Janelle E et al. · Experimental brain research · 2018 · DOI
Quick Summary
Researchers compared two methods for identifying ME/CFS patients: looking at structural brain changes on MRI scans versus asking patients to rate their symptoms like fatigue, pain, and sleep quality. While brain imaging showed some useful differences between ME/CFS patients and healthy people, patients' own descriptions of their symptoms were actually better at identifying who had ME/CFS.
Why It Matters
This study demonstrates that while ME/CFS patients show measurable structural brain changes detectable by advanced analysis, subjective symptom reporting remains more reliable for classification. This has important implications for diagnostic development and suggests that validated symptom questionnaires may be more practical clinical tools than expensive neuroimaging for identifying ME/CFS.
Observed Findings
- Structural MRI analysis classified patients with 79.58% accuracy.
- Self-report symptom ratings classified patients with 95.95% accuracy.
- Multiple brain regions related to cognition, emotion, and memory contributed significantly to sMRI-based classification.
- Self-report ratings substantially outperformed structural MRI models in classification performance.
Inferred Conclusions
- Structural MRI abnormalities are useful for discriminating ME/CFS patients from healthy controls but are less effective than self-reported symptoms for classification.
- Self-report measures may be more practical and effective diagnostic indicators than neuroimaging for ME/CFS identification.
- Cognitive, emotional, and memory-related brain regions are implicated in ME/CFS pathology based on structural differences.
Remaining Questions
- Why does self-report substantially outperform structural imaging—is this due to limitations in current sMRI analysis methods or the genuine primacy of subjective symptom experience in ME/CFS?
- Do the observed structural brain changes relate causally to specific ME/CFS symptoms or underlying pathophysiology?
- Would longitudinal imaging reveal progressive structural changes, and how do these correlate with symptom trajectories?
What This Study Does Not Prove
This study does not prove that brain imaging cannot detect ME/CFS—it shows that symptom self-report was more effective in this particular classification task, but this may reflect limitations in the imaging analysis method rather than the absence of brain changes. The cross-sectional design cannot establish causality or progression of structural changes. The small sample size limits generalizability of these findings.
Tags
Symptom:Cognitive DysfunctionPainFatigueUnrefreshing Sleep
Biomarker:Neuroimaging
Method Flag:Weak Case DefinitionSmall SampleExploratory OnlyPEM Not Defined
Metadata
- DOI
- 10.1007/s00221-018-5301-8
- PMID
- 29846797
- Review status
- Editor reviewed
- Evidence level
- Single-study or moderate support from human research
- Last updated
- 12 April 2026
About the PEM badge: “PEM required” means post-exertional malaise was an explicit required diagnostic criterion for participant inclusion in this study — not that PEM was studied, observed, or discussed. Studies using criteria that do not require PEM (e.g. Fukuda, Oxford) are tagged “PEM not required”. How the atlas works →
Contribute
Private, reviewed by a human. Not a public comment thread.