Abstract:
A major goal in the management of traumatic brain injury (TBI) is to optimize cerebral blood flow (CBF) over the
days to weeks following injury. Yet, there is currently no well-established, non-invasive method to continuously
monitor CBF in adults. Though diffuse optical flowmetry (DOF) methods for monitoring CBF based on dynamic
scattering of near-infrared (NIR) light are used in research, these methods suffer from fundamental limitations.
First, they require costly photon counting, which strictly constrains achievable speed, brain specificity, and brain
coverage. Second, they lack depth discrimination, leading to contamination from blood flow in superficial tissues.
Third, they require assumptions about optical properties, which can vary between individuals or across brain
regions. Fourth, they are sensitive to noise from ambient light.
To address these limitations, we introduce interferometry to human diffuse optics, creating a new class of NIR
light-based monitoring tools, called interferometric Diffuse Optical Spectroscopy (iDOS). First, we show that with
interferometry, a CMOS sensor can replace photon counting and parallelize measurements of weak diffuse light
fluctuations that reveal CBF. This advance improves light throughput-to-cost ratio by ~100x. We can thus take
more measurements (improving brain coverage) with larger source collector separations (improving brain
specificity). Then, we show that by rapidly tuning the light source wavelength, we also achieve time-of-flight
(TOF) resolution. This extra TOF dimension better distinguishes brain from superficial tissue, and also provides
estimates of optical properties, improving quantification. Finally, interferometric methods are essentially
unaffected by ambient light.
Building on our promising results in adult humans, we will develop, optimize, and validate iDOS for quantitative,
rapid, and robust CBF monitoring. We will identify the advantages and weaknesses of iDOS relative to
conventional methods. Finally, we will perform observational CBF monitoring in severe TBI patients in the
neurointensive care unit (neuro-ICU), testing the ability of our non-invasive CBF measurements to predict periods
of hypoxia.