In recent years, development of tissue purification, which renders biological sample transparent, has proposed a solution to solve the penetration depth problem. Limitation of these studies was not able to provide the analysis of tissue property changes induced by tissue clearing and to compare tissue clearing characteristics. Here we present optical coherence tomography (OCT) and magnetic resonance imaging (MRI) to quantitatively assess the tissue clearing technique.
Because MR signal is based on atomic properties, we can physically investigate the basic principle of tissue cleaning by monitoring the change of tissue atomic properties. Through this study, we can investigate the different tissue clearing characteristics and compare the existing clearing technique.
Introduction
Optical Imaging for Neuroanatomy Research
- Optical Imaging Modalities
- Limitation of Optical Imaging
- Techniques to Overcome the Light Penetration Issue
- Tissue Clearing by Reducing Refractive Index Gap
- Tissue Clearing by Removing Lipid
Therefore, many researchers developed techniques to compensate for the defect of light penetration, such as serial block view (SBF) imaging and tissue clearing. By reducing scatter, clearing the tissue increases light transmission and enables deep tissue imaging. Tissue clearing is non-invasive and reversible, so it can provide intact deep brain imaging.
The principle of tissue clearing is to reduce scattering and the way is divided into two. First, to clear tissue by reducing the refractive index aperture, the following equation is used.
New Approach for Quantification of Tissue Clearing
- Needs on Quantitative Analysis of Tissue Clearing
It means that tissue clearing increases the mean free path where single scattering is controlled. From the previous discussion, we use the mean free path map to check the efficiency of tissue clearing. To assess the efficiency of clearing other tissue, we compared the mean free path of existing tissue removal.
Through the mean free path analysis, we can know that tissue clearance increases the mean free path. To compare the regional difference of tissue clearance efficiency, we measure the mean free path change due to ClearT by brain region.
Experimental Material and Methods.…
Optical Coherence Tomography
- Characteristics of Optical Coherence Tomography
- Brain Imaging with Optical Coherence Tomography
Optical coherence tomography performs label-free, non-invasive and cross-sectional tissue morphology based on Michelson interferometer. In spectral domain optical coherence tomography (SD-OCT)[45], light is low coherent, it means that the bandwidth of light is wide. Because incoherent light such as white light causes interference that is not detectable, on the other hand, high coherent light performs a low axial resolution.
Where n is the refractive index of the sample, 𝜆0 is the center wavelength of the light source and ∆𝜆 is the bandwidth of the light source. For high axial resolution, the bandwidth must be wide and the center wavelength must be short. For high lateral resolution, the center wavelength and focal length must be short, the beam diameter must be too large.
In conclusion, SD-OCT must use a broadband light source for proper interference, because interference. The system used a superluminescent diode (EXS210046-02, Exalos) operating at a central wavelength of 1310 nm with a bandwidth of 70 nm, providing an axial resolution of 10.7 µm in tissue. Objective lenses with a focal length of 40 mm and a lateral resolution of 5.7 µm were used in each path.
The interference signal was acquired from the camera and processed with custom-built software written in LabVIEW that supports standard SD-OCT signal processing such as wavenumber linearization, dispersion compensation, and inverse Fourier transform. To achieve wide-field imaging while fixing the focal plane on the upper surface of the tissue, we specially designed a sample holder mounted on two-axis linear motorized stages as shown in the figure. It allowed the tissue to be flattened and immersed in various solutions such as PBS and cleaning solutions during imaging.
Magnetic Resonance Imaging
- Characteristics of Magnetic Resonance Imaging.…
- Brain Imaging with Magnetic Resonance Imaging
- Multi-Slice Multi-Echo(MSME) sequence
MRI contrast between different tissue types is generated by proton density, T1 recovery, and T2 decay[47]. T1-weighted imaging, T2-weighted imaging, proton density imaging are commonly used for MRI brain imaging. To understand the T1-weighted image, T2-weighted image, proton density image, T1 recovery meaning, T2 decay should be preceded.
T1 recycling is caused by the exchange of energy from nuclei to their surrounding environment or lattice by RF pulse. T1 recovery is a measure of how quickly the net magnetization vector (NMV) recovers to its ground state in the direction of B0. As the nuclei dissipate their energy, their magnetic moments relax or return to B0. T1 recovery time is defined as the time it takes for 63% of the longitudinal magnetization to recover in that tissue.
On the other hand, the T1 recovery time of water is long because water is inefficient in receiving energy from nuclei. The T2 decay time is defined as the time required for 37% of the transverse magnetization to be lost due to dephasing. In the proton density picture, high proton density is hyperintense and low proton density is hypointense.
Unlike conventional single-echo pulse sequences, it monitors a sequence. Figure 2-4 T1 recovery and T2 decay. Thus, it is ideal for fitting T2 decay curves along multiple TEs, as shown in the following equation. SO is the initial signal amplitude, which is proportional to the proton density and water content of individual voxels.
Tissue Preparation and Clearing
- Perfusion and Dissection
- Tissue Clearing
- Image Processing for OCT Analysis
- Image Processing for MRI Analysis
For hydrogel monomer infusion, harvested brain is immersed in the monomer solution for 3 days at 4℃. Finally, we measured the spacing between two intersection points on the same straight line, and defined the spacing as the shift in the boundaries. Then we define the attenuation coefficient as the slope of the first-order fitting curve on OCT signal[49].
Finally, we define the mean free path as the inverse of the corresponding value of the attenuation coefficient. Using MATLAB (Mathworks, Natick, MA), we calculated the proton density S0 and the spin-spin interaction parameter T2 from the MRI data obtained with the MSME sequence. The S0 proton density signal is the amplitude of the MRI signal at TE = 0 ms, which is extrapolated from the T2 decay curve and is proportional to the water content of individual voxels.
We normalized the S0 proton density in the brain to the S0 proton density in the area of the 1.5% agarose gel for comparison in several samples. The S0 proton density in the agarose gel was estimated from the manually drawn region of interest (ROI) for the gel area in each sample, and the S0 proton density in the agarose gel area is generally uniform. In the present study, S0 proton density was used as a quantitative parameter of the hydration/dehydration effect for each type of tissue clearing method.
T2 decay is faster (smaller T2) in the region where the large molecules such as fat are abundant because the protons are densely packed and easily interact with each other. We visualized the change in proton density S0 and spin-spin interaction parameter T2 in the brain before and after each type of tissue clearance. In addition, we manually drew ROIs for the cerebral cortex and the corpus callosum regions in each sample to compare MRI parameter changes in gray matter and white matter regions in the brain.
Results
Investigation of Tissue Clearing Effect with OCT
- Confirmation of Morphological Change
- Quantification of Tissue Size Change
- Visualization of Imaging Depth Enhancement
- Quantification of Imaging Depth Enhancement
- Reflectivity and Attenuation Coefficient
- Reflectivity and Attenuation Coefficient Varying with Tissue Clearing
- Mean Free Path for Quantification of Tissue Clearing
- Mean Free Path Varying with Clear T
- Mapping of Reflectivity, Attenuation Coefficient and Mean Free Path
- Mapping Mean Free Path with Clear T
- Comparing with Tissue Clearing Efficiency…
Investigation of Tissue Clearing Principle with MRI
- Tissue Clearing Analysis with S0 Map
- ROI Analysis of S0 Map
- Tissue Clearing Analysis with T2 Map
- ROI Analysis of T2 Map
- Comparison of MRI Data with OCT Data
In PACT, the rate of increase of the S0 value in the corpus callosum is ∼48% while the rate of increase of the S0 value in the cortex is ∼13%. This means that the S0 value in the corpus callosum is more affected by PACT clearance than the S0 value in the cortex. This is because the corpus callosum includes more lipids than the cortex and the corpus callosum is more affected by hyperhydration.
Because lipids are more affected by Scale, the rate of increase of the S0 value in the corpus callosum is ~34%, on the other hand, the rate of increase of the S0 value in the cortex is ~15%. The reduction rate of the S0 value in the cortex is ~77% while the reduction rate of the S0 value in the corpus callosum is ~70%. That is, the rate of decrease of the S0 value in the cortex is greater than that of the corpus callosum.
In PACT, the degree of T2 enhancement in the corpus callosum is ~80%, while the degree of T2 enhancement in the cortex is ~28%. Although the final S0 value in the cortex and corpus callosum is similar, the final T2 value in the cortex (93.19) and corpus callosum (104.05) is different. Because the corpus callosum has more lipids, the T2 gain rate in the corpus callosum is ~66%.
In particular, the T2 reduction rate in the cortex is ~70%, while the T2 reduction rate in the corpus callosum is ~66%. A remarkable fact is that the rate of decrease in S0 value in the cortex is lower than in the corpus callosum, while the rate of decrease in the T2 value in the corpus callosum is higher than in the corpus callosum. The T2 reduction rate in the cortex is ~51%, on the other hand, the T2 reduction rate in the corpus callosum is ~37%.
Discussion
This mean free path value does not take into account the change in optical path length due to refractive index. This mean free path value takes into account the change in optical path length due to refractive index. In the case of the MRI analysis, we confirmed that the MRI matched the data point well, except for ClearT.
As shown in Figure 4-2, 4-3, the previous ClearT fitting does not match the data point, so we modify the fitted MRI line. Because lipid is a dispersive factor, it is important to analyze the change in lipids due to tissue clearance. To monitor lipid changes due to tissue clearance, we plan to perform magnetic resonance spectroscopy (MRS) analysis[51-54] or fat suppression[55-57].
Lipid analysis by MRS or lipid suppression will contribute to lipid change due to tissue purification, which will be useful for understanding the principle of tissue purification with lipids. Other MRI adjustment lines match the data point well, but MRI adjustment with ClearT does not account for the data point.
Conclusion
That is, with MRI analysis we can predict the efficiency of tissue clearance depending on the brain region. Li, A., et al., Micro-optical section tomography to obtain a high-resolution atlas of the mouse brain. Dodt, H.-U., et al., Ultramicroscopy: three-dimensional visualization of neuronal networks in the whole mouse brain.
Marchesini, R., et al., Extinction and absorption coefficients and scattering phase functions of human tissue in vitro. Renier, N., et al., iDISCO: a simple, rapid method to immunolabel large tissue samples for volume imaging. Ke, M.-T., et al., Super-Resolution Mapping of Neuronal Circuitry With an Index-Optimized Clearing Agent.
Kuwajima, T., et al., ClearT: a detergent- and solvent-free cleaning method for neuronal and non-neuronal tissues. Srinivasan, V.J., et al., Optical coherence microscopy for deep tissue imaging of cerebral cortex with intrinsic contrast. Arous, J.B., et al., Label-free imaging of single myelin fibers in living rodents by deep optical coherence microscopy.
Larin, K.V., et al., Optical cleaning for OCT image enhancement and in-depth monitoring of molecular diffusion. Kwong, K.K., et al., Dynamic magnetic resonance imaging of human brain activity during primary sensory stimulation. Ogawa, S., et al., Intrinsic signal changes accompanying sensory stimulation: functional magnetic resonance imaging of the brain.
