Table of contents

A brief review of the state of the art in theoretical and experimental studies of the optical properties of metal particles with dipole and multipole plasmon resonances is presented. Metal spheres, nanorods, spherical and elliptic metal nanoshells are considered. The tuning of plasmon resonances of nanoparticles by varying their size, shape, structure, and dielectric environment is described. A large amount of spectrophotometric data on dimensional characteristics of gold colloidal particles is critically analysed and a new calibration of the dependence of their average size on the extinction plasmon resonance wavelength is proposed. A drastic difference between gold and silver colloids in the region of small deviations of their form from spherical is discussed. An example of the excess over not only the Rayleigh limit for the scattering depolarisation factor for dielectric needles (1/3) but also over the plasmon-resonance limit for metal thin rods (3/4) is presented for the first time. The multipole properties of nanorods and universal linear wavelength scaling of multipole resonances are considered depending on the axial ratio of nanoparticles. The outlook for modern trends in biomedical applications of nanoparticles with plasmon resonances is discussed.

A method is proposed for visualisation of the velocity fields of colloidal plasmon-resonance nanoparticles moving in a laser beam. The method uses the particle image velocimetry for processing ultramicroscopic images. Particles in a thick layer of colloidal solution are illuminated by a slit laser ultramicroscopic source with a large numerical aperture providing a high contrast of particle images and visualisation of the transverse velocity distribution in laser-induced flows with a high spatial resolution.

The spatial temperature distribution in biological tissues upon interaction of laser radiation with plasmon-resonance nanoparticles is simulated numerically. Experimental thermograms of model objects and real biological tissues are obtained in vivo for different localisation depths of nanoparticles in different irradiation regimes. The results obtained in the study can be used for the development of methods of laser photothermolysis of malignant tumours.

Optical coherence tomography (OCT) images of turbid medium with spatially fluctuating optical parameters demonstrate the presence of spatial noise ('shadow noise') caused by shading of the observed layer by variations in scattering and absorption coefficients of superficial layers. In this paper we report the development of the theoretical model of shadow noise in OCT images in which the features for discriminating this type of noise from the noise of a different origin are determined. Shadow noise is discovered in OCT images of biological tissue after specific image processing and the validity of the developed theoretical model is proven. The role of shadow noise as an interfering factor at imaging the variations in the backscattering coefficient, as well as the source of extra information on optical characteristics of the medium is analysed.

Functional imaging, monitoring and quantitative description of glucose diffusion in epithelial and underlying stromal tissues in vivo and controlling of the optical properties of tissues are extremely important for many biomedical applications including the development of noninvasive or minimally invasive glucose sensors as well as for therapy and diagnostics of various diseases, such as cancer, diabetic retinopathy, and glaucoma. Recent progress in the development of a noninvasive molecular diffusion biosensor based on optical coherence tomography (OCT) is described. The diffusion of glucose was studied in several epithelial tissues both in vitro and in vivo. Because OCT provides depth-resolved imaging of tissues with high in-depth resolution, the glucose diffusion is described not only as a function of time but also as a function of depth.

Two new techniques for the analysis of speckle-patterns, formed by laser radiation dispersed on histological sections with malignant and non-malignant growths, are proposed. One of these techniques is based on the calculation of invariant Zernike moments of the spatial distribution of the speckle-field intensity. The second technique is based on the calculation of the fractal dimension of the intensity spatial distribution in the speckle structure. It is shown that both these methods yield the same results, which drastically depend on tissue properties. The possibility of using Zernike moments and fractal dimensions, formed by laser light dispersed on histological sections, in in vitro express-diagnostics of tissues with pathological changes is investigated.

The method of autocorrelation low coherence interferometry is proposed for diagnostics of inhomogeneities and the internal structure of layered technical and biological samples. In this method the low coherence optical field reflected from the layered sample is analysed by using a Michelson interferometer. Because the object is outside the interferometer, the distance between the interferometer and the object under study is not limited and thus the object can move during the measurements. Theoretical substantiation of the autocorrelation method for media with discrete and continuous optical structure modifications is presented.

Optical coherence tomography (OCT) images of model human skin samples are obtained by using Monte Carlo simulations. The contributions of least and multiple scattering, diffusion and nondiffusion components and of separate scattering orders are studied by using a multilayer skin model based on experimental images. The model images are obtained by neglecting speckles or taking them into account. It is shown that least scattering forms the image of the upper skin layers, while the contribution of multiple scattering can be characterised as a blurred full image with a lower contrast. Repeated scattering mainly contributes to the OCT image at depths up to 1 mm. The diffusion component contributes to the image beginning from the epidermal basal layer. The partial image produced by this component is more blurred compared to the partial image produced by to multiple scattering. The nondiffusion component forms the OCT skin image at depths up to ≈1.3 mm.

The feasibility of using an artificial neural network (ANN), which is the standard Matlab tool, for non-invasive (based on the data of backscattering) diagnostics of macro-inhomogeneities, localised at subsurface layers of the turbid strongly scattering medium was shown. The spatial and angle distribution of the backscattered optical radiation was calculated by using the Monte-Carlo method combining the modelling of effective optical paths and the use of statistical weights. It was shown that application of the backscattering method together with the ANN allows solving inverse problems for determining the average radius of the scattering particles and for reconstructing the images of structural elements within the medium with a high accuracy.

The possibility of selective laser photothermolysis improvement for the removal of tattoo pigments due to the optical clearing of human skin is investigated. It is shown experimentally that the optical skin clearing increases the tattoo image contrast. Computer Monte Carlo simulations show that by decreasing the laser beam scattering in upper skin layers, it is possible to reduce the radiation power required for tattoo removal by 30%—40% and, therefore, to increase the the photothermolysis efficiency.

The photon average trajectory method is considered, which is used as an approximate method of diffuse optical tomography and is based on the solution of the Radon-like trajectory integral equation. A system of linear algebraic equations describing a discrete model of object reconstruction is once inverted by using a modified multiplicative algebraic technique. The blurring of diffusion tomograms is eliminated by using space-varying restoration and methods of nonlinear colour interpretation of data. The optical models of the breast tissue in the form of rectangular scattering objects with circular absorbing inhomogeneities are reconstructed within the framework of the numerical experiment from optical projections simulated for time-domain measurement technique. It is shown that the quality of diffusion tomograms reconstructed by this method is close to that of tomograms reconstructed by using Newton-like multistep algorithms, while the computational time is much shorter.

Results of investigation of cutaneous benign and malignant pigmented lesions by laser-induced autofluorescence spectroscopy (LIAFS) and diffuse reflectance spectroscopy (DRS) are presented. The autofluorescence of human skin was excited by a 337-nm nitrogen laser. A broadband halogen lamp (400—900 nm) was used for diffuse reflectance measurements. A microspectrometer detected in vivo the fluorescence and reflectance signals from human skin. The main spectral features of benign (dermal nevi, compound nevi, dysplastic nevi) and malignant (melanoma) lesions are discussed. The combined usage of the fluorescence and reflectance spectral methods to determine the type of the lesion, which increases the total diagnostic accuracy, is compared with the usage of LIAFS or DRS only. We also applied colorimetric transformation of the reflectance spectra detected and received additional evaluation criteria for determination of type of the lesion under study. Spectra from healthy skin areas near the lesion were detected and changes between healthy and lesion skin spectra were revealed. The influence of the main skin pigments on the detected spectra is discussed and evaluation of possibilities for differentiation between malignant and benign lesions is performed based on their spectral properties. This research shows that the non-invasive and high-sensitive in vivo detection by means of appropriate light sources and detectors should be possible, related to the real-time determination of existing pathological conditions.

A theoretical model is developed and an algorithm is proposed for calculating far-field light scattering by a transparent dielectric particle significantly larger than a wavelength. The accuracy of this algorithm is close to that of the discrete dipole approximation. The calculation time for this algorithm in the case of particles with the size parameter higher than 50 is much lower than that for the discrete dipole approximation. Scattering diagrams for spheroidal particles of different sizes, orientations and refractive indices are calculated. The proposed algorithm has a great potential for quick calculations of parameters of light scattering by large biological particles such as erythrocytes and their aggregates, bacteria, etc.