Measurements Of Optical Properties And Imaging Techniques For Tissue Using Multi-scale Analysis

This thesis is concerned about the hot issue on biological tissue photonics, and develops methods for measuring tissue optical properties and imaging techniques combining multi-scale analysis with different scaling imaging modes (phase-contrast microscopy, speckle, optical coherence tomography and photoacoustic imaging). Innovations in the main research include three aspects.A method combining mathematical morphology with Mie theory to simulate optical scattering in biological tissues is proposed:Multi-scale morphological granulometry is used to analyze the phase-contrast images of biological tissue and to estimate scatterer size distribution of the tissue samples. Using the fractal features of the size distribution, the optical parameters associated with light scattering in tissue are quantitatively estimated using Mie theory. In addition, the method could be extended to describe the local optical parameters. Furthermore, our results suggest that this unique method can be used to characterize biological tissues for disease diagnosis. Finally, the improved MC simulation algorithm, using the discrete phase function can be a more accurate picture of the light propagation in tissue.Some indirect methods are developed for measuring tissue optical scattering properties, absorption properties and the Brownian dynamics based on mathematical morphology and continous wavelet transform:an approximately U-quadratic theoretical model is proposed for simultaneous estimating the optical scattering coefficient and the Brownian diffusion coefficient of the superficial layer of tissue; a theoretical model is presented for average granulometric size of speckle in terms of scattering coefficient based on spatial correlation function of the backscattered intensity; a theoretical model for analyzing of speckle contrast of optical coherence tomography shows contrast ratio is a linear function of depth and the slope of this linear dependence is proportional to the scattering coefficient; a method using focusing photoacoustic imaging to quantify the target optical absorption coefficient is proposed. In this method, wavelet transform is used to extract the photoacoustic signals at specific acoustic frequencies, and correct them. This method is particularly useful to provide accurate absorption coefficient for predicting the outcomes of photothermal interaction for cancer treatment with absorption enhancement.The methods using mathematical morphology and wavelet multi-scale filtering to improve image quality:since the shape of speckle is airy, we employ a morphological opening operation with a disc-shaped structuring element to suppress the speckle coarse grains, improving the quality of polarization-difference images; wavelet ridge transform is used to reduce the speckle of OCT signal; the wavelet analysis could improve PA signal-to-noise ratio (SNR) through choosing suitable scale, and make photoacoustic imaging system using longitudinal single sampling quickly obtain information of sample. The photoacoustic imaging is applied to identify the myocardium of rat in vivo.