Graduation Year

2022

Document Type

Dissertation

Degree

Ph.D.

Degree Name

Doctor of Philosophy (Ph.D.)

Degree Granting Department

Electrical Engineering

Major Professor

Ashwin B. Parthasarathy, Ph.D.

Committee Member

Arash Takshi, Ph.D.

Committee Member

Stephen E. Saddow, Ph.D.

Committee Member

Andrew Hoff, Ph.D.

Committee Member

W. Scott Burgin, M.D.

Committee Member

John Murray-Bruce, Ph.D.

Keywords

Cerebral blood flow (CBF), Laser speckle contrast imaging, Multi-exposure speckle imaging, Near-infrared spectroscopy (NIRS), Wearable hemodynamic

Abstract

Cerebral blood flow (CBF) is a good indicator of brain health as blood carries necessary nutrients, oxygen, and metabolic byproducts. Quantitative blood flow information can be used in several clinical and therapeutic applications such as stroke detection, measuring autoregulation, evaluating brain injury, or determining neuronal activity. Over the past few decades, light-based deep tissue hemodynamic detection modalities have become popular for non-invasive CBF measurements. In particular, noninvasive Diffuse Correlation Spectroscopy (DCS), has become a tool of choice for research and clinical applications due to its depth sensitivity (>1 cm), portability, validity against other technologies such as Magnetic Resonance Imaging (MRI) or Computed Tomography (CT), and its use of non-ionizing light radiation that permits prolonged and continuous real-time in-vivo measurements. Although the measurement accuracy of DCS is excellent, instrumentational complexity, cost, processing burden, and size constraints limit wide adoption.

In recent years, several alternative implementations of DCS have been proposed addressing its limitations, such as Diffuse Speckle Contrast Analysis (DSCA) and Speckle Contrast Optical Spectroscopy (SCOS). Briefly, in DCS a point detector, i.e., Single Photon Counting Avalanche Photodiode (SPAD), is used to measure temporal fluctuations of light intensity to compute the intensity autocorrelation function and estimate blood flow. DSCA and SCOS measure light intensity fluctuations with a camera, and compute second order speckle statistics to estimate tissue hemodynamics. Use of a camera as photodetector reduces instrumental complexity and data processing burden. These modalities provide a relatively easy measurement framework. However, they are not suitable for portable multichannel operation due to the use of fiber coupled high coherence lasers and camera or APD array as a photodetector. Moreover, realizing a multichannel detection system with this technique would be expensive.

In this dissertation, I address the hardware limitations of DCS and speckle-based blood flow measurement. First, I introduce a novel digital hardware based TTL (Transistor to Transistor Logic) pulse counting technique, Compressed DCS (CDCS), which achieved 87.5% data compression for DCS data acquisition. Aside from excellent data compression, it provides a cost-effective (i.e., >100x cost efficiency) hardware scheme to implement multi-channel hardware acquisition for a DCS instrument. Second, I developed a fiber-less portable low power laser system for DCS – fiber-less DCS (FBDCS). I validated the capability of FBDCS to probe deeper into the tissue surface, up to 3.5 cm. Finally, I have introduced a novel deep tissue blood flow modality, Integrated Diffuse Speckle Contrast Spectroscopy (iDSCS), a simple photodiode based deep tissue blood flow detection technique. iDSCS utilizes an off-the-shelf, low-cost photodetector in photo-voltaic mode to detect temporal speckle fluctuations. A model-based fitting was then performed to compute relative blood flow (rBF).

All three technologies have been validated in phantom and in-vivo measurements. I further investigated noise in these systems thus providing a guide for future implementation. In combination, these techniques could pave the way for the development of a portable, multi-channel deep tissue blood flow detection device.

Download

Share

COinS