Guohua Shi

Optical absorption and scattering are critical properties of biological tissues, but strong background light often obscures this information, limiting imaging contrast and the visualization of tissue microstructures. Current methods for enhancing imaging contrast rely on image processing and noise suppression, but often lose critical details under strong background light. To address this issue, we propose a dual-modal imaging technique based on a low-rank and sparse matrix decomposition (LRSD) of light fields, enabling simultaneous high-contrast imaging of absorption and scattering and significantly improving imaging performance. Monte Carlo simulation results demonstrate that the low-rank component of the light field effectively separates background light, while the sparse component accurately captures the absorption and scattering properties of the target. In imaging experiments on skin follicle tissues, this method successfully extracted absorption and scattering information, achieving a twofold improvement in imaging contrast, with the SNR improving by 2.97 dB and significantly enhancing the visualization of tissue microstructures. Compared to traditional image filtering methods, the LRSD technique showed superior performance under strong background light conditions. Furthermore, imaging experiments on different regions of rabbit taste bud slices further validated the broad applicability and potential of this method in biological imaging. The high-contrast dual-modal imaging method proposed in this study demonstrates exceptional capabilities in visualizing the tissue structure, offering an innovative solution for the clinical evaluation of pathological sections.

Traditional cell imaging is limited to 3D surface imaging and has a relatively low resolution. A phase microscopy technique based on transmission matrix has been proposed for high-resolution tomographic imaging within cells. This method requires only a single measurement with the phase microscope to obtain the phase information from the cell and construct a transmission matrix. By analyzing the singular values of the transmission matrix, this method enables the extraction of structural information at different depths. The effectiveness of the technique was validated through Monte Carlo simulations. Experimental results with red blood cells revealed intracellular structures, achieving an axial resolution of 5 nm. Additionally, the application of this technique in distinguishing white blood cell types demonstrates its ability to effectively image and identify unique structural characteristics of cells. Combining the white light diffraction phase microscope with singular value decomposition based on the transmission matrix, lateral resolution can be achieved within 100 nm. Experimental results highlight the potential of this method in distinguishing subtle cellular components and structural variations, making it a valuable tool for cell biology research and medical diagnostics. This innovative technique provides a highly efficient and non-invasive approach for high-resolution cell imaging, with promising applications in medical diagnostics and biological research.

Optical coherence tomography (OCT) provides three-dimensional images of biological tissues based on low-coherence optical interference. Attenuation coefficient estimation is one of the important functional extensions of OCT and has received many research efforts as attenuation coefficient has been found to be associated with histopathological transformation of human tissues. However, attenuation coefficient estimation accuracy is deteriorated by the confocal effect on OCT A-scan. Thus, it is desired to eliminate the confocal effect, which requires accurate determination of the confocal parameters. In this study, we propose what we believe to be a novel method for confocal parameter extraction, called dual NA ratio fitting (DNRF). DNRF requires a repetition of B-scan with varied numerical apertures (NA), altering the Rayleigh length of B-scan but keeping the focal depth fixed. Then, the focal depth and Rayleigh length can be determined by fitting an A-scan ratio function from the two repeated B-scans. The NA tuning was achieved by adding a beam expansion module into the sample arm. The method was evaluated on intralipid samples and multi-layer phantoms. The results demonstrate that our method is capable of determine the confocal parameters with good accuracy. With the extracted confocal parameters, the confocal effect was removed effectively, upon which attenuation coefficient estimation using traditional depth-resolved method appeared to be more accurate than the confocal parameter extraction based on the A-scan ratio function of two repeated B-scans with varied focal depths. Last, our method was tested on human skin in vivo, yielding attenuation coefficients consistent with literature. This indicates good potential of our method to be used for clinical applications.

The reflection matrix optical coherence tomography (RM-OCT) method has made significant progress in extending imaging depth within scattering media. However, the current method of measuring the reflection matrix (RM) through uniform sampling results in relatively long data collection time. This study demonstrates that the RM in scattering media exhibits sparsity. Consequently, a compressed sensing (CS)-based technique for measuring the RM is proposed and applied to RM-OCT images. Experimental results show that this method requires less than 50% of the traditional sampling data to recover target information within scattering media, thereby significantly reducing data acquisition time. These findings not only expand the theory of the RM but also provide a more efficient measurement approach. This advancement opens up broader applications for CS techniques in RM-OCT and holds great potential for improving imaging efficiency in scattering media.