fNIRS-derived neurocognitive ratio as a biomarker for neuropsychiatric diseases

Abstract

. Significance: Clinical use of fNIRS-derived features has always suffered low sensitivity and specificity due to signal contamination from background systemic physiological fluctuations. We provide an algorithm to extract cognition-related features by eliminating the effect of background signal contamination, hence improving the classification accuracy. Aim: The aim in this study is to investigate the classification accuracy of an fNIRS-derived biomarker based on global efficiency (GE). To this end, fNIRS data were collected during a computerized Stroop task from healthy controls and patients with migraine, obsessive compulsive disorder, and schizophrenia. Approach: Functional connectivity (FC) maps were computed from [HbO] time series data for neutral (N), congruent (C), and incongruent (I) stimuli using the partial correlation approach. Reconstruction of FC matrices with optimal choice of principal components yielded two independent networks: cognitive mode network (CM) and default mode network (DM). Results: GE values computed for each FC matrix after applying principal component analysis (PCA) yielded strong statistical significance leading to a higher specificity and accuracy. A new index, neurocognitive ratio (NCR), was computed by multiplying the cognitive quotients (CQ) and ratio of GE of CM to GE of DM. When mean values of NCR ( ) over all stimuli were computed, they showed high sensitivity (100%), specificity (95.5%), and accuracy (96.3%) for all subjects groups. Conclusions: can reliable be used as a biomarker to improve the classification of healthy to neuropsychiatric patients.

Four-Class Classification of Neuropsychiatric Disorders by Use of Functional Near-Infrared Spectroscopy Derived Biomarkers

“…All experiments were conducted according to the latest Declaration of Helsinki. Parts of these datasets have been utilized in previous works performed by our group and coworkers [ 23 , 24 , 25 , 26 , 27 , 28 , 29 ].…”

“…Such a procedure is necessary to isolate the extent of correlation caused by only neuronally induced hemodynamic effects, since common physiological effects of non-neuronal origin could inflate the correlation between signals of channel pairs. A partial correlation approach was adapted from the works of Akin [ 28 ] and Akin [ 29 ] to reduce the impact of common systemic interference to Pearson’s correlation coefficients calculated between HBO signals of each channel pair. Similar to these works, HBO signals of all channels were initially high-pass-filtered with a cut-off frequency of 0.009 Hz using an 8th order Butterworth filter.…”

“…Similar to these works, HBO signals of all channels were initially high-pass-filtered with a cut-off frequency of 0.009 Hz using an 8th order Butterworth filter. The high-pass-filtered HBO signals were then averaged to have a single global signal regressor, which was utilized as the partial regressor for modeling and removing the impact of common systemic noise from the correlations between each channel pair in the subsequent step of the analysis [ 22 , 24 , 25 , 26 , 27 , 28 , 29 ].…”

“…Similarly, the partial correlation regressor was obtained by truncating and concatenating the time segments belonging to all task blocks in the global signal regressor. Then, 16-by-16 partial correlation (PC)-corrected functional connectivity (FC) matrices for each subject were generated after removing the impact of this partial correlation regressor [ 22 , 29 ].…”

“…Regarding the hemodynamic features, a relatively novel functional connectivity metric called global efficiency (GE) was obtained from the 16-by-16 partial correlation-corrected FC matrices obtained for each subject. The GE metric was obtained from a graph theoretical network analysis approach, and its efficacy in demonstrating the degree of connectedness and information transfer between cortical regions during various cognitive tasks has been shown in previous studies [ 22 , 27 , 28 , 29 ].…”