Visible and near-infrared spectra collected from the thumbs of patients with chronic fatigue syndrome for diagnosis.

Background: Currently, diagnosis of chronic fatigue syndrome (CFS) is based on clinical symptoms and therefore relies on the experience and skill of the doctors. Here, we have examined the possible diagnosis of CFS based on spectral information and chemometrics analysis, such as principal component analysis (PCA) and soft modeling of class analogy (SIMCA).
Methods: Visible and near-infrared (Vis-NIR) spectroscopy was used to examine possible changes in the region of 600-1100 nm in thumbs and assessed.
Results: The Vis-NIR spectra of thumbs from 57 CFS patients and 74 healthy volunteers were subjected to PCA and SIMCA to develop multivariate models to discriminate between CFS patients and healthy individuals. The model was further assessed by the prediction of 120 determinations (60 in the healthy group and 60 in the CFS patient group). The PCA model predicted a discrimination of the masked samples; specifically the SIMCA model correctly predicted 51 of 60 (83.3%) healthy volunteers and 42 of 60 (70%) CFS patients.
Conclusions: Despite the relatively small number of subjects involved in this trial, who were exclusively Japanese, our results imply that Vis-NIR spectroscopy of the thumb combined with chemometrics analysis may provide a valuable tool for diagnosing CFS.
(Copyright © 2012 Elsevier B.V. All rights reserved.)