Flourishing is an important criterion to assess wellbeing, however, controversies remain, particularly around assessing it with self report measures. Due to this reason, to be able to understand the underlying neural mechanisms of well being, researchers often utilize neuroimaging techniques. However, rather than individual answers, previous neuroimaging studies using statistical approaches provided an answer in average sense. To overcome these problems, we applied machine learning techniques to discriminate 43 highly flourishing from regular flourishing individuals by using a publicly available resting state functional near infrared spectroscopy (rsfNIRS) dataset to get an answer in individual level. We utilized both Pearson correlation (CC) and Dynamic Time Warping (DTW) algorithm to estimate functional connectivity from rs-fNIRS data on temporo-parieto-occipital region as input to nine different machine learning algorithms. Our results revealed that by utilizing oxyhemoglobin concentration change with Pearson correlation (CC HbO) and deoxy hemoglobin concentration change with dynamic time warping (DTW Hb), we could be able to classify flourishing individuals with 90 % accuracy with AUC 0.90 and 0.93 using nearest neighbor and Radial Basis Kernel Support Vector Machine. This finding suggests that temporoparietooccipital regional based resting state connectivity might be a potential biomarker to identify the levels of flourishing and using both connectivity measures might allow us to find different potential biomarkers.
bioRxiv Subject Collection: Neuroscience