Comparing the performance potential of speckle contrast optical spectroscopy and diffuse correlation spectroscopy for cerebral blood flow monitoring using Monte Carlo simulations in realistic head geometries

Neurophotonics, Journal Year: 2024, Volume and Issue: 11(01)

Published: Jan. 27, 2024

SignificanceThe non-invasive measurement of cerebral blood flow based on diffuse optical techniques has seen increased interest as a research tool for perfusion monitoring in critical care and functional brain imaging. Diffuse correlation spectroscopy (DCS) speckle contrast (SCOS) are two such that measure complementary aspects the fluctuating intensity signal, with DCS quantifying temporal fluctuations signal SCOS spatial blurring pattern. With increasing use these techniques, thorough comparison would inform new adopters benefits each technique.AimWe systematically evaluate performance flow.ApproachMonte Carlo simulations dynamic light scattering an MRI-derived head model were performed. For both SCOS, estimates sensitivity to changes, coefficient variation measured flow, contrast-to-noise ratio calculated. By varying data collection between methods, we investigated different strategies, including altering number modes per detector, integration time/fitting time measurement, laser source delivery strategy.ResultsThrough across metrics simulated detectors having realistic noise properties, determine several guiding principles optimization report over range properties tissue geometries. We find outperforms terms ideal case here but note requires careful experimental calibrations ensure accurate measurements flow.ConclusionWe provide design by which development systems their flow.

Language: Английский

---

Quantification of blood flow index in diffuse correlation spectroscopy using a robust deep learning method

Journal of Biomedical Optics, Journal Year: 2024, Volume and Issue: 29(01)

Published: Jan. 27, 2024

SignificanceDiffuse correlation spectroscopy (DCS) is a powerful, noninvasive optical technique for measuring blood flow. Traditionally the flow index (BFi) derived through nonlinear least-square fitting measured intensity autocorrelation function (ACF). However, process computationally intensive, susceptible to measurement noise, and easily influenced by properties (absorption coefficient μa reduced scattering μs′) scalp skull thicknesses.AimWe aim develop data-driven method that enables rapid robust analysis of multiple-scattered light’s temporal ACFs. Moreover, proposed can be applied range source–detector distances instead being limited specific distance.ApproachWe present deep learning architecture with one-dimensional convolution neural networks, called DCS network (DCS-NET), BFi coherent factor (β) estimation. This DCS-NET was performed using simulated data based on three-layer brain model. We quantified impact from physiologically relevant property variations, layer thicknesses, realistic noise levels, multiple (5, 10, 15, 20, 25, 30 mm) β estimations among DCS-NET, semi-infinite, models.ResultsDCS-NET shows much faster speed, around 17,000-fold 32-fold than traditional semi-infinite models, respectively. It offers higher intrinsic sensitivity tissues compared methods. excellent anti-noise features less sensitive variations μs′ at separation mm. Also, we have demonstrated relative (rBFi) extracted lower error 8.35%. By contrast, models result in significant errors rBFi 43.76% 19.66%, respectively.ConclusionsDCS-NET robustly quantify measurements considerable distances, corresponding deeper biological tissues. has potential hardware implementation, promising continuous real-time measurements.

Language: Английский

---

Using depth-enhanced diffuse correlation spectroscopy and near-infrared spectroscopy to isolate cerebral hemodynamics during transient hypotension

Neurophotonics, Journal Year: 2023, Volume and Issue: 10(02)

Published: June 5, 2023

Combining diffuse correlation spectroscopy (DCS) and near-infrared (NIRS) permits simultaneous monitoring of multiple cerebral hemodynamic parameters related to autoregulation; however, interpreting these optical measurements can be confounded by signal contamination from extracerebral tissue.We aimed evaluate in NIRS/DCS data acquired during transient hypotension assess suitable means separating scalp brain signals.A hybrid time-resolved NIRS/multidistance DCS system was used simultaneously acquire oxygenation blood flow orthostatic induced rapid-onset lower body negative pressure (LBNP) nine young, healthy adults. Changes microvascular were verified against changes middle artery velocity (MCAv) measured transcranial Doppler ultrasound.LBNP significantly decreased arterial (-18%±14%), (>30%), tissue (all p≤0.04 versus baseline). However, implementing depth-sensitive techniques for both NIRS indicated that LBNP did not alter relative their baseline values p≥0.14). In agreement, there no significant reduction MCAv (8%±16%; p=0.09).Transient caused larger the compared brain. We demonstrate importance accounting within measures hemodynamics physiological paradigms designed test autoregulation.

Language: Английский

---

Optimizing a two-layer method for hybrid diffuse correlation spectroscopy and frequency-domain diffuse optical spectroscopy cerebral measurements in adults

Neurophotonics, Journal Year: 2023, Volume and Issue: 10(02)

Published: May 23, 2023

SignificanceThe sensitivity to extracerebral tissues is a well-known confounder of diffuse optics. Two-layer (2L) head models can separate cerebral signals from artifacts, but they also carry the risk crosstalk between fitting parameters.AimWe aim implement constrained 2L model for hybrid correlation spectroscopy (DCS) and frequency-domain optical (FD-DOS) data characterize errors in blood flow tissue absorption with proposed model.ApproachThe algorithm uses analytical solution cylinder an priori layer thickness fit multidistance FD-DOS (0.8 4 cm) DCS 2.5 data, assuming homogeneous reduced scattering. We characterized algorithm's accuracy simulated noise generated using slab realistic adult geometries vitro phantom data.ResultsOur recovered index 6.3 [2.8, 13.2]% 34 [30, 42]% (median absolute percent error [interquartile range]) geometries, respectively. Corresponding coefficient were 5.0 [3.0, 7.9]% 4.6 [2.4, 7.2]% 8 [5, 12]% our experiment. Our results minimally sensitive second-layer scattering changes robust cross-talk parameters.ConclusionsIn adults, promises improve FD-DOS/DCS compared conventional semi-infinite approach.

Language: Английский

---

Deep-learning-based separation of shallow and deep layer blood flow rates in diffuse correlation spectroscopy

Biomedical Optics Express, Journal Year: 2023, Volume and Issue: 14(10), P. 5358 - 5358

Published: Sept. 13, 2023

Diffuse correlation spectroscopy faces challenges concerning the contamination of cutaneous and deep tissue blood flow. We propose a long short-term memory network to directly quantify flow rates shallow deep-layer tissues. By exploiting different contributions auto-correlation functions, we accurately predict (RMSE = 0.047 0.034 ml/min/100 g simulated tissue, R2 0.99 0.99, respectively) in two-layer phantom experiment. This approach is useful evaluating responses active muscles, where both deep-muscle increase with exercise.

Language: Английский

---

Influence of oversimplifying the head anatomy on cerebral blood flow measurements with diffuse correlation spectroscopy

Neurophotonics, Journal Year: 2023, Volume and Issue: 10(01)

Published: March 30, 2023

SignificanceDiffuse correlation spectroscopy (DCS) is an emerging optical modality for non-invasive assessment of index regional cerebral blood flow. By the nature this noninvasive measurement, light must pass through extracerebral layers (i.e., skull, scalp, and spinal fluid) before detection at tissue surface. To minimize contribution these to measured signal, analytical model has been developed that treats head as a series three parallel infinitely extending slabs (mimicking brain). The three-layer shown provide significant improvement in flow estimation over typically used bulk homogenous medium. However, still gross oversimplification geometry ignores curvature, presence cerebrospinal fluid (CSF), heterogeneity layer thickness.AimDetermine influence oversimplifying on estimated with model.ApproachData were simulated Monte Carlo four-layer slab medium sphere isolate CSF respectively. Additionally, simulations performed magnetic resonance imaging (MRI) templates spanning wide-range ages. Simulated data fit both CBF. Finally, mitigate errors potential CBF due difficulty defining thickness, we investigated approach identify equivalent, "optimized" thickness via pressure modulation.ResultsBoth curvature failing account lead effect relative changes minimal. Further, found was underestimated all MRI-templates, although magnitude underestimations highly influenced by small variations source detector optode positioning. optimized obtained from modulation did not improve accuracy CBF, it significantly CBF.ConclusionsIn sum, findings suggest holds promise improving flow; however, estimations absolute should be viewed caution given difficult appreciable sources error, such CSF.

Language: Английский

---

Pulsatile microvascular cerebral blood flow waveforms change with intracranial compliance and age

Neurophotonics, Journal Year: 2024, Volume and Issue: 11(01)

Published: Jan. 20, 2024

SignificanceDiffuse correlation spectroscopy (DCS) is an optical method to measure relative changes in cerebral blood flow (rCBF) the microvasculature. Each heartbeat generates a pulsatile signal with distinct morphological features that we hypothesized be related intracranial compliance (ICC).AimWe aim study how three of rCBF waveforms: augmentation index (AIx), pulsatility index, and area under curve, change respect ICC. We describe ICC as combination vascular extravascular compliance.ApproachSince patients Chiari malformations (CM) (n=30) have been shown altered compliance, compare morphology waveforms CM age-matched healthy control (n=30).ResultsAIx measured supine position was significantly less compared controls (p<0.05). Since physiologic aging also leads vessel stiffness intravascular evaluate waveform age find AIx feature strongly correlated (Rhealthy subjects=−0.63, Rpreoperative patient=−0.70, Rpostoperative patients=−0.62, p<0.01).ConclusionsThese results suggest microvasculature using DCS may

Language: Английский

---

Study of Time-Resolved Dynamics in Turbid Medium Using a Single-Cavity Dual-Comb Laser

ACS Photonics, Journal Year: 2024, Volume and Issue: unknown

Published: June 5, 2024

In measuring cerebral blood flow (CBF) noninvasively using optical techniques, diffusing-wave spectroscopy is often combined with near-infrared to obtain a reliable index. Measuring the index at determined depth remains ultimate goal. this study, we present simple approach dual-comb lasers where simultaneously measure absorption coefficient (μ

Language: Английский

---

ATLAS: a large array, on-chip compute SPAD camera for multispeckle diffuse correlation spectroscopy

Biomedical Optics Express, Journal Year: 2024, Volume and Issue: 15(11), P. 6499 - 6499

Published: Sept. 2, 2024

We present ATLAS, a 512 × single-photon avalanche diode (SPAD) array with embedded autocorrelation computation, implemented in 3D-stacked CMOS technology, suitable for correlation spectroscopy applications, including diffuse (DCS). The shared per-macropixel SRAM architecture provides 128 macropixel resolution, parallel minimum lag-time of 1 µs. demonstrate the direct, on-chip computation function sensor, and its capability to resolve changes decorrelation times typical body tissue real time, at long source-detector separations similar those achieved by current leading optical modalities cerebral blood flow monitoring. Finally, we suitability in-vivo measurements through cuff-occlusion forehead cardiac signal measurements.

Language: Английский

---

Pathlength-selective, interferometric diffuse correlation spectroscopy (PaLS-iDCS)

bioRxiv (Cold Spring Harbor Laboratory), Journal Year: 2024, Volume and Issue: unknown

Published: June 27, 2024

Diffuse correlation spectroscopy (DCS) is an optical method that offers non-invasive assessment of blood flow in tissue through the analysis intensity fluctuations diffusely backscattered coherent light. The nature technique has enabled several clinical applications for deep measurements, including cerebral monitoring as well tumor mapping. While a promising technique, measurement configurations targeting hemodynamics, standard DCS implementations suffer from insufficient signal-to-noise ratio (SNR), depth sensitivity, and sampling rate, limiting their utility. In this work, we present enhanced called pathlength-selective, interferometric (PaLS-iDCS), which improves upon both sensitivity to hemodynamics SNR using pathlength-specific gain. Through detection, PaLS-iDCS can provide time-of-flight (ToF) specific information without use expensive time-tagging electronics low-jitter detectors. new compared time-domain (TD-DCS), another able resolve photon ToF tissue, Monte Carlo simulation, phantom experiments, human subject measurements. consistently demonstrates improvements (>2x) similar conditions (same ToF), allow measurements at extended ToFs, have increased (~50% increase). Further, like TD-DCS, allows direct estimation properties sampled distribution need separate spectroscopic measurement. This relatively straightforward way systems make robust with greatly enabling further technology.

Language: Английский

---