In the present study, the possible difference in visible and near-infrared (Vis-NIR) spectra of plasma in psychiatric patients compared to healthy control subjects was investigated. Mean Vis-NIR spectra of 60 psychiatric patients showed several specific peaks in the 630- to 930-nm region. Principal component analysis (PCA) and soft independent modeling of class analogy (SIMCA) were applied to manifest the spectral characteristics of each group. The PCA and SIMCA models demonstrated successful discrimination of the two groups using a score plot of the third versus fourth principal component and a Coomans plot, respectively. The PCA loadings were generally consistent with the discriminating power of the SIMCA model, indicating a specific pattern of Vis-NIR spectra in the plasma of psychiatric patients. The specific pattern possibly indicates altered superoxide, amine, and primary and secondary amides in plasma and may provide a possible indicator for the diagnosis of heterogeneous psychiatric disorders. This is the first step toward a clinical tool which has the potential to be applied in the diagnosis of psychiatric disease. Yet, there remain obstacles to its practical usage.

Related Articles