A machine learning approach to identify functional biomarkers in human prefrontal cortex for individuals with traumatic brain injury using functional near‐infrared spectroscopy

Abstract

BackgroundWe have explored the potential prefrontal hemodynamic biomarkers to characterize subjects with Traumatic Brain Injury (TBI) by employing the multivariate machine learning approach and introducing a novel task‐related hemodynamic response detection followed by a heuristic search for optimum set of hemodynamic features. To achieve this goal, the hemodynamic response from a group of 31 healthy controls and 30 chronic TBI subjects were recorded as they performed a complexity task.MethodsTo determine the optimum hemodynamic features, we considered 11 features and their combinations in characterizing TBI subjects. We investigated the significance of the features by utilizing a machine learning classification algorithm to score all the possible combinations of features according to their predictive power.Results and ConclusionsThe identified optimum feature elements resulted in classification accuracy, sensitivity, and specificity of 85%, 85%, and 84%, respectively. Classification improvement was achieved for TBI subject classification through feature combination. It signified the major advantage of the multivariate analysis over the commonly used univariate analysis suggesting that the features that are individually irrelevant in characterizing the data may become relevant when used in combination. We also conducted a spatio‐temporal classification to identify regions within the prefrontal cortex (PFC) that contribute in distinguishing between TBI and healthy subjects. As expected, Brodmann areas (BA) 10 within the PFC were isolated as the region that healthy subjects (unlike subjects with TBI), showed major hemodynamic activity in response to the High Complexity task. Overall, our results indicate that identified temporal and spatio‐temporal features from PFC's hemodynamic activity are promising biomarkers in classifying subjects with TBI.

Near-Infrared Spectroscopy (NIRS) in Traumatic Brain Injury (TBI)

“…The outcome of neurorehabilitation is mainly related to prefrontal cortex features. All the prognosis papers ( n = 18) assessed patients’ neurorehabilitation using NIRS, however, there was just one diagnosis publication where NIRS was used to identify regions within the prefrontal cortex that contributed to distinguishing between TBI and healthy subjects [ 25 ]. Likewise, 18 studies were included that explored NIRS accuracy for hematoma detection after trauma.…”

Current trends in biomarker discovery and analysis tools for traumatic brain injury

“…This approach demonstrates the utility of using predicative algorithms for pediatric TBI biomarker discovery and showcases its power in the probability of detecting which biomarkers are indicative of a more aggressive disease progression later in life. Hemodynamics influenced by injury have also been explored as possible biomarkers of TBI, with predictive classification algorithms revealing significant temporal and spatial activity in the prefrontal cortex as possible diagnostic markers of injury [150].…”

The role of prefrontal cortex in a moral judgment task using functional near‐infrared spectroscopy

“…Here, HbO signals were low passed filtered at 0.1 Hz, then the moving average filter with 1.5 s timing window was applied to smooth the signal. Subsequently, the linear and nonlinear trends were removed by fitting a low order (order of 6) polynomial to the fNIRS signals and subtracting it from the original signal (Karamzadeh et al, 2016;Minati, Visani, Dowell, Medford, & Critchley, 2011;Pfeifer, Scholkmann, & Labruyère, 2017;Zhao, Ji, Li, & Li, 2018).…”