Evaluating Brain Flow Index from a temple-worn wearable against depth-resolved time-domain NIRS during head-down tilt and postural transitions in healthy young men
- 1Temple Private Limited, Gurugram, India
Abstract
Significance: Wearable optical sensors could make continuous cerebral hemodynamic monitoring practical in daily life. Because any surface head sensor also samples extracerebral tissue, demonstrating cerebral relevance requires a reference that separates deep from superficial hemodynamics. Aim: We evaluated whether the Temple Brain Flow Index (Temple-BF), derived from a temple-worn photoplethysmographic wearable, covaried more strongly with the deeper (brain-assigned) than the superficial (scalp-assigned) hemodynamics recovered from time-domain near-infrared spectroscopy (TD-NIRS) during controlled postural challenges. Approach: Sixty sessions from 44 healthy young adult men were analyzed across three physiological challenges: 30{degrees} head-down tilt, stand-to-squat, and stand-to-supine transitions. Optical density and mean time of flight from a three-module TD-NIRS system were inverted with a Monte Carlo-derived two-layer model to obtain scalp- and brain-layer hemoglobin time series. Temple-BF was compared with these model-derived signals using lag-adjusted and zero-lag correlations, paired brain-versus-scalp comparisons, heart rate and short-separation-HbO adjusted partial correlations. Results: Median lag-adjusted correlations between Temple-BF and brain-layer {Delta}HbO traces were 0.912, 0.843, and 0.839 for head-down tilt, stand-to-squat, and stand-to-supine transitions, respectively; corresponding scalp-layer medians were 0.494, 0.700, and 0.767. Paired brain-minus-scalp differences were significant for head-down tilt (median 0.333, p < 0.001) and stand-to-squat (0.110, p < 0.001) but not for stand-to-supine (0.036, p = 0.082) transitions. Zero-lag brain-layer medians were 0.826, 0.701, and 0.796, and the association persisted after adjustment for heart rate (partial r = 0.832, 0.850, 0.654) and for short-separation superficial HbO (0.649, 0.570, 0.438). Transition-window correlations had medians of 0.878, 0.861, and 0.876, with directionally concordant transition responses in 59 of 60 sessions. Conclusions: Temple-BF tracked model-derived brain-layer TD-NIRS {Delta}HbO across all three postural challenges, with larger median correlations against the brain-layer than the scalp compartment in all three protocols, and transition responses in the same direction for almost all sessions. These results support Temple-BF as a relative marker of cerebral hemodynamic change during postural perturbations. Establishing cerebral specificity at the temple, and excluding residual systemic and superficial contributions, will require flow-sensitive references and fuller systemic monitoring.
Related
- A Time-Resolved Framework for Quantifying Neuronal Network State Transitions
- Genome-resolved analysis reveals successional dynamics and functional transitions in chicken gut archaea across the broiler growth cycle
- Wearable Estimation of Knee Joint Contact Forces: SiMPLeSENSE, a Simulated-IMU Pipeline Validated Against In Vivo Loads
- Evaluating the efficacy of DEET-based topical repellent products against biting midges using a modified arm-in-cage method
- WITHDRAWN: Evaluating phage-antibiotic synergy in differentiated primary airway epithelial cultures against Pseudomonas aeruginosa
BibTeX
@misc{gulati2026evaluating,
title = {{Evaluating Brain Flow Index from a temple-worn wearable against depth-resolved time-domain NIRS during head-down tilt and postural transitions in healthy young men}},
author = {D. Gulati and A. Rudaeva and A. Dutta and D. Rogers and R. Prajapat and N. Kumar and S. Gupta and D. Goyal and D. A. Boas},
howpublished = {bioRxiv},
year = {2026},
month = sep,
doi = {10.64898/2026.09.17.751664},
url = {https://www.biorxiv.org/content/10.64898/2026.09.17.751664v1},
note = {CC BY-NC-ND},
}
RIS
TY - UNPB
TI - Evaluating Brain Flow Index from a temple-worn wearable against depth-resolved time-domain NIRS during head-down tilt and postural transitions in healthy young men
AU - D. Gulati
AU - A. Rudaeva
AU - A. Dutta
AU - D. Rogers
AU - R. Prajapat
AU - N. Kumar
AU - S. Gupta
AU - D. Goyal
AU - D. A. Boas
PY - 2026
DA - 2026/09/17
PB - bioRxiv
DO - 10.64898/2026.09.17.751664
UR - https://www.biorxiv.org/content/10.64898/2026.09.17.751664v1
AB - Significance: Wearable optical sensors could make continuous cerebral hemodynamic monitoring practical in daily life. Because any surface head sensor also samples extracerebral tissue, demonstrating cerebral relevance requires a reference that separates deep from superficial hemodynamics. Aim: We evaluated whether the Temple Brain Flow Index (Temple-BF), derived from a temple-worn photoplethysmographic wearable, covaried more strongly with the deeper (brain-assigned) than the superficial (scalp-assigned) hemodynamics recovered from time-domain near-infrared spectroscopy (TD-NIRS) during controlled postural challenges. Approach: Sixty sessions from 44 healthy young adult men were analyzed across three physiological challenges: 30{degrees} head-down tilt, stand-to-squat, and stand-to-supine transitions. Optical density and mean time of flight from a three-module TD-NIRS system were inverted with a Monte Carlo-derived two-layer model to obtain scalp- and brain-layer hemoglobin time series. Temple-BF was compared with these model-derived signals using lag-adjusted and zero-lag correlations, paired brain-versus-scalp comparisons, heart rate and short-separation-HbO adjusted partial correlations. Results: Median lag-adjusted correlations between Temple-BF and brain-layer {Delta}HbO traces were 0.912, 0.843, and 0.839 for head-down tilt, stand-to-squat, and stand-to-supine transitions, respectively; corresponding scalp-layer medians were 0.494, 0.700, and 0.767. Paired brain-minus-scalp differences were significant for head-down tilt (median 0.333, p < 0.001) and stand-to-squat (0.110, p < 0.001) but not for stand-to-supine (0.036, p = 0.082) transitions. Zero-lag brain-layer medians were 0.826, 0.701, and 0.796, and the association persisted after adjustment for heart rate (partial r = 0.832, 0.850, 0.654) and for short-separation superficial HbO (0.649, 0.570, 0.438). Transition-window correlations had medians of 0.878, 0.861, and 0.876, with directionally concordant transition responses in 59 of 60 sessions. Conclusions: Temple-BF tracked model-derived brain-layer TD-NIRS {Delta}HbO across all three postural challenges, with larger median correlations against the brain-layer than the scalp compartment in all three protocols, and transition responses in the same direction for almost all sessions. These results support Temple-BF as a relative marker of cerebral hemodynamic change during postural perturbations. Establishing cerebral specificity at the temple, and excluding residual systemic and superficial contributions, will require flow-sensitive references and fuller systemic monitoring.
ER -