Clempi Almeida E Silva, Ana Laura, Reis, Victor Hugo Palhares Flávio, Lamoglia, Antonieta Santos Andrade et al. · The Journal of manual & manipulative therapy · 2026 · DOI
This review looked at 43 studies from 2011–2024 to see how artificial intelligence (AI) and machine learning can help diagnose fibromyalgia and other conditions like ME/CFS. Researchers found that AI tools can identify unique patterns in brain scans, genes, eye imaging, and blood markers that differ between sick patients and healthy people. While some AI models showed very high accuracy (up to 100%), most studies tested only small groups of patients, so we need bigger, more rigorous research before these tools can be used reliably in clinics.
This review is important because ME/CFS, like fibromyalgia, lacks specific diagnostic biomarkers and benefits from the same diagnostic challenges. The identification of potential AI-detected biomarkers (such as microRNA signatures) could advance objective testing for ME/CFS if validated in larger studies. Understanding how machine learning can improve diagnosis in related conditions provides a roadmap for developing more accurate, faster diagnostic tools that could reduce diagnostic delays in ME/CFS patients.
This scoping review does not prove that AI/ML tools are ready for clinical use—most studies were small, pilot-based projects without rigorous validation in independent cohorts. The review does not establish whether these biomarkers are mechanistically meaningful or merely statistical correlations. It also does not demonstrate efficacy of AI-guided treatment strategies, only diagnostic potential.
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 →
Spotted an error in this entry? Report it →