Phantom and Model-Based Near Infrared Spectroscopy Measurements of Intracranial Hematoma From Infants to Adults

Wang, Lei; İzzetoğlu, Meltem; Du, Jie; Ayaz, Hasan

doi:10.1109/tim.2022.3157343

Cited by 6 publications

(1 citation statement)

References 45 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Owing to its main features, such as affordability, portability, and versatility, fNIRS holds great potential to be applied to pediatric and neonatal (Peng & Hou, 2021;Tang et al, 2024), neurodegenerative (Blum et al, 2022;Kuruvilla et al, 2013;Keles et al, 2022), and critically brain-injured (Abdalmalak et al, 2020;Bicciato et al, 2022;Wang et al, 2022;Shu et al, 2023) populations, as well as naturalistic paradigms more broadly (see Pinti et al, 2018 for a review).…”

Section: Introductionmentioning

confidence: 99%

Assessing the consistency and sensitivity of the neural correlates of narrative stimuli using functional near-infrared spectroscopy

Kolisnyk¹,

Novi²,

Abdalmalak³

et al. 2023

Preprint

View full text Add to dashboard Cite

The neural correlates of narrative stimuli have been established in clinical contexts as an effective means for detecting covert consciousness in behaviourally unresponsive patients using both fMRI and EEG. In particular, synchronization in frontoparietal regions between behaviourally unresponsive patients and controls can indicate whether these patients are processing the movie in a similar fashion as controls. However, it is unclear whether functional near-infrared spectroscopy (fNIRS) is sufficiently sensitive and consistent to detect these neural correlates at the level required for effective clinical diagnoses. In this study, whole brain inter-subject synchronization was observed using fNIRS in healthy controls when they watched part of the movie Bang! You’re Dead and listened to an audio clip from the movie Taken. Importantly, synchronization was largely reduced or absent when participants viewed scrambled versions of those clips. Moreover, neural activity during intact clips could predict independently acquired ratings of suspense. To assess the prospect of using synchronization as a diagnostic tool in behaviourally unresponsive patients, single-participant reproducibility was estimated by measuring the consistency and sensitivity of the results. Consistency was quantified by estimating each participant’s dissimilarity in their synchronization using Euclidean distance, which determined that the synchronization was statistically similar in 14/15 participants in Bang! You are Dead and 15/15 participants in Taken. Moreover, machine learning algorithms were able to decode between both intact conditions and their scrambled counterparts using synchronization, resulting in an 83% accuracy in Bang! You are Dead and 80% accuracy in Taken. Overall, the neural correlates of narrative stimuli, as assessed by fNIRS, are highly reproducible across participants and hold great potential for identifying covert consciousness in behaviourally unresponsive patients.

show abstract