Date of Award
Doctor of Philosophy (PhD)
Capitalizing on the Brownian motion of red blood cells, the distinct optical absorption spectra of oxy-hemoglobin and deoxy-hemoglobin, and the blood flow-induced signal decorrelation, multi-parametric photoacoustic microscopy (PAM) is capable of quantifying concentration of hemoglobin (CHb), blood oxygen saturation (sO2), and hemodynamics at the microscopic level via statistical, spectroscopic, and correlation analyses. Multi-parametric PAM has found broad biomedical applications. In fact, there are still much room to further improve the performance and imaging capability of multi-parametric PAM to meet the ever-increasing demanding requirements of biomedical research. To date, there are two main obstacles that prevent multi-parametric PAM from being more impactful: limited spatiotemporal coverage and imaging contrast. This dissertation focuses on these two properties and explores several potential solutions to further leverage the academic influence of multi-parametric PAM.Improving the imaging speed of multi-parametric PAM is essential to leveraging its temporal coverage. However, to avoid temporal overlap, the A-line rate is limited by the acoustic speed in biological tissues to a few megahertz. Moreover, to achieve high-speed PAM of the sO2, the stimulated Raman scattering effect in optical fibers has been widely used to generate 558 nm from a commercial 532 nm laser for dual-wavelength excitation. However, the fiber length for effective wavelength conversion is typically short, corresponding to a small time delay that leads to a significant overlap of the A-lines acquired at the two wavelengths. Increasing the fiber length extends the time interval but limits the usable pulse energy at 558 nm. To this end, I develop an approach based on a conditional generative adversarial network (cGAN) that enables temporal unmixing of photoacoustic A-line signals with an interval as short as ∼38 ns, breaking the physical limit on the A-line rate. Moreover, this deep learning approach allows the use of multi-spectral laser pulses for PAM excitation, addressing the insufficient energy of monochromatic laser pulses. This technique lays the foundation for ultrahigh-speed imaging and thus leverages the temporal coverage of multi-parametric PAM. Typically, the spatial coverage of multi-parametric PAM is limited by the depth of focus imposed by the Gaussian-beam excitation. As a result, the quantitative measurements become inaccurate when the imaging object is out of focus. To address this problem, I have developed a hardware-software combined approach by integrating Bessel-beam excitation and cGAN-based deep learning. Side-by-side comparison of the new cGAN-powered Bessel-beam multi-parametric PAM against the conventional Gaussian-beam multi-parametric PAM shows that the new system enables high-resolution, quantitative imaging of CHb, sO2, and blood flow over a depth range of ~600 μm in the live mouse brain, with errors ~13–58 times lower than those of the conventional system. Better fulfilling the rigid requirement of light focusing for accurate hemodynamic measurements, this deep learning-powered Bessel-beam multi-parametric PAM successfully improves the spatial coverage and may find applications in large-field functional recording across the uneven brain surface and beyond (e.g., tumor imaging). Pathological aggregation of Aβ peptides results in the deposition of amyloid in the brain parenchyma (senile plaques in Alzheimer’s disease [AD]) and around cerebral microvessels (cerebral amyloid angiopathy [CAA]). Our current understanding of the amyloid-induced microvascular changes has been limited to the structure and hemodynamics— leaving the oxygen-metabolic aspect unattended. To address this issue, we develop a dual-contrast PAM technique, which integrates the molecular contrast of dichroism PAM and the physiological contrast of multiparametric PAM for simultaneous, intravital imaging of amyloid deposition and cerebrovascular function in a mouse model that develops AD and CAA. This technique adds dichroism contrast and opens up new opportunities to study the spatiotemporal interplay between amyloid deposition and vascular-metabolic dysfunction in AD and CAA. Over the past few years, the integration of PAM and two-photon microscopy (TPM) has been implemented by several groups to provide spatiotemporally co-registered dual contrasts of optical absorption and fluorescence in vivo. However, in those implementations, PAM is merely used to capture the structural information of the microvasculature without hemodynamic or oxygen-metabolic insights and is mostly in transmission mode (i.e., optical excitation and ultrasonic detection are on different sides of the imaging target). To tackle this problem, I co-develop a novel integration of multi-parametric PAM and TPM, which enables simultaneous imaging of the microvascular function and fluorescence-labeled molecular contrasts in reflection mode. A parabolic ultrasonic mirror is carefully designed and fabricated to transmit the PAM and TPM excitation through its central opening, while focusing the light-induced ultrasonic waves for detection by a flat transducer positioned sideways. This configuration allows TPM imaging with high numerical aperture and PAM imaging with good sensitivity. I have demonstrated the utility of this technique in the mouse brain by simultaneously imaging the microvascular function and green fluorescent protein-labeled microglia. I also demonstrated simultaneous imaging of the interplay between neural activity and microvascular function during whisker stimulation with frame rate up to ~20 Hz. With the fluorescence contrast, this integrated system provides a new tool for studying neurovascular and neurometabolic coupling in the brain.
Quing Zhu, Jin-Moo Lee, Adam Bauer, Manu Goyal,
Available for download on Monday, August 26, 2024