Improved Algorithm For Constructing The Structural Images of Biological Object in OCT


An improved algorithm of structural images of biological object for optical coherence tomography, which allows to increase the depth of coherent sensing and get a better quality picture. Optical coherence tomography (OCT) has emerged in the late eighties, early nineties of the twentieth century. [1] At the beginning of XXI century, it took its place in a number of medical diagnostic equipment. OCT uses optical signal reflected from the surfaces of different optical density, and in many ways is similar to ultrasound (U.S.) diagnosis. The probing depth of dense tissues OCT systems using wavelength ?? = 900 – 1300 nm, is 1-2 mm, which is significantly less than that of ultrasound systems [2, 3] and X-ray devices [4]. Due to the strong scattering of optical radiation in the dense biological tissues, OCT systems are used primarily for the study of the cornea, vitreous and retina. However, the resolution of OCT systems for one, two orders of magnitude higher resolution ultrasound systems for similar research, which is about 1 – 0.1 mm [2]. The aim of this work – to provide an improved algorithm for constructing the structural OCT images of biological object to allowing to increase the depth of coherent sensing and an image with high contrast and informative. Electrical signal received from the detectors OCT included in the balanced circuit, is amplified and digitized by the ADC mean intensity of the radiation reflected from the biological object. Preparation of 2-axis images of the interference signal is reduced to the construction of a spectrogram. The spectrogram is a function of two variables: time and frequency. That is, the interference signal as a function of one variable (time) is converted into a spectrogram is a function of two variables. To construct a spectrogram interference signal is divided into short time segments of equal length. Each of these segments is applied fast Fourier transform (STFT, among LabVIEW). At each of the segments of the spectrum is a complex-valued function of sample number (or time). It is known that a complex-valued function can not be built in one coordinate system in the plane. Therefore, the analysis of the spectrum usually build amplitude and phase spectra of any signal. The amplitude spectrum is a module of the complex spectrum and the phase – his argument. The spectrogram is a combination of the amplitude spectra, calculated on short segments, a function of two variables, or matrix. Similar treatment algorithm is shown in Figure 1. The algorithm can distinguish five fundamentally important stages: “Splitting the signal”, “Fourier transform”, “Isolation of the envelope”, “The logarithm of the envelope”, “Writing data to the matrix” Figure 1 – The processing algorithm of the electrical signal from the detectors optical coherence tomography The next stage of the signal processing is to use the fast Fourier transform to each segment. Since the path difference scanning interferometer arms varies continuously scanning optical delay line, theoretically window Fourier transform, must also move continuously at one point, but it makes the signal processing is quite long, on the order of minutes. Empirically, it has been shown that the signal processing is shifted to the window of 70-80% has the same contrast ratio as well as the continuous shift – to a point. It takes 2 – 5 seconds when using a computer with average parameters (single-core processor 2.4 GHz, 512 MB RAM). Using a powerful computer and specialized software, this time can be reduced to one second. This approach delivers images in real time and visual feedback when using live biomedical facility. Signal processing is shifted windows by 70-80% – is an important feature of our proposed treatment of the electrical signal. [5] The next stage of the signal processing is to separate the spectral envelope of the signal received by the Fourier transform of each segment. An important feature of the received signal is its symmetry with respect to zero,the optical path difference of the waves. This is explained by the fact that the result of the Fourier transform is complex function, the real part of which is symmetric, and the imaginary antisymmetric. Since in real applications is a real part of the signal or its magnitude, the reconstructed signal has a balanced view. In addition to the symmetry of the signal spectral OCT has another important feature – the imposition of a mirror of the complex conjugate signal. If the optical path difference between the reference wave and the wave of the biological object is zero useful signal is superimposed on the autocorrelation component, in which case there will be multiple image artifacts. This can be avoided by placing biological object so that his first boundary was removed from the position of zero path difference of the waves in the interferometer by an amount greater than the optical thickness of the object itself. [5] The next stage will be the logarithm of the envelope interference signal of each segment. It is necessary to correct symmetry. When logarithm removed part located below the zero path difference of the waves. The final step is to combine the processing of amplitude spectra, calculated on short segments in the matrix. The data basis for this matrix imaging. Literature

Leave a Reply

Your email address will not be published.