Method for spatial frequency domain imaging of optical properties of biological tissue based on absorption

Abstract

The application discloses a kind of biological tissue optical characteristic parameter space frequency domain imaging method based on absorption rate, comprising the following steps: setting biological tissue Monte Carlo simulation parameter;Absorption rate-based biological tissue optical characteristic parameter look-up table is constructed by Monte Carlo simulation and one-dimensional zero-order Hankel transformation;For reference phantom projection cosine illumination pattern and collect fringe image;For the fringe image, Fourier transform, frequency domain filtering and inverse Fourier transform are carried out to obtain direct current component and alternating current component modulation degree;Reference phantom is replaced by measured biological tissue body, and the direct current component and alternating current component modulation degree are obtained by repeating the preceding 2 steps;Direct current and alternating current absorption rate measurement value are calculated;The minimum value of absorption rate Euclidean distance is calculated, and the absorption coefficient and reduced scattering coefficient measurement result are obtained by table lookup, to obtain optical characteristic parameter image.The application takes biological tissue absorption rate as the starting point of optical characteristic parameter inversion, significantly improves imaging sensitivity and reduces inversion error.

Description

Technical Field

[0001] This invention relates to the field of optical measurement and imaging technology, specifically to a spatial frequency domain imaging method for biological tissue optical characteristic parameters based on absorption rate. Background Technology

[0002] The characterization of optical properties of biological tissues holds immense potential in biodetection and medical diagnosis, and is increasingly being explored as a technological high ground. Among these advancements, emerging spatial frequency domain imaging techniques can achieve wide-field-of-view and non-destructive quantitative mapping of biological tissue optical images, providing unprecedented application scope for in vivo detection and clinical diagnosis. However, existing spatial frequency domain imaging techniques achieve optical property parameter imaging through inversion based on the measurement of tissue diffuse reflectance. This presents two problems: first, the use of polarizers to eliminate specular reflection leads to low imaging sensitivity and system complexity; second, existing diffuse reflectance measurements struggle to eliminate interface diffuse reflection light generated by the roughness of biological tissue interfaces, while inversion methods rely on back diffuse reflectance (without interface diffuse reflection light) to invert optical property parameters, resulting in inherent inversion errors. This invention addresses these shortcomings by proposing a spatial frequency domain imaging method for biological tissue optical property parameters based on absorptivity. Summary of the Invention

[0003] To improve the imaging effect of spatial frequency domain imaging of biological tissue optical property parameters, this invention discloses a spatial frequency domain imaging method for biological tissue optical property parameters based on absorptivity. Compared with existing spatial frequency domain imaging methods for biological tissue optical property parameters, this invention obtains the absorptivity of biological tissue by measuring reflectivity, including interfacial specular reflection, interfacial diffuse reflection, and backscattered reflection, and then performs optical property parameter inversion based on the obtained absorptivity. The purpose of this invention is to improve the sensitivity of spatial frequency domain imaging of biological tissue optical property parameters and reduce its inversion error, thereby improving the imaging effect.

[0004] The objective of this invention is achieved as follows:

[0005] A spatial frequency domain imaging method for biological tissue optical property parameters based on absorptivity includes the following steps:

[0006] Step a: Based on the properties of biological tissue, set the Monte Carlo simulation parameters for the biological tissue, including the geometry and size of the simulated biological tissue sample, anisotropy coefficient, refractive index, refractive index of the surrounding medium, number and spacing of grid lines, number and wavelength of incident photons, optical characteristic parameters, and spatial frequency of the illumination light. The optical characteristic parameters include the absorption coefficient μ. a Reduced scattering coefficient μ s ′;

[0007] Step b: Using the Monte Carlo simulation parameters of biological tissue set in step a, perform Monte Carlo simulation to obtain the spatial domain shock response function of absorptivity corresponding to the optical property parameters of the simulated sample. Then, perform a one-dimensional zero-order Hankel transform to obtain the simulated absorptivity values ​​T(f=0) and T(f>0) of the simulated sample at zero frequency f=0 and AC spatial frequency f>0, respectively. Construct the mapping relationship between the simulated absorptivity values ​​of the simulated sample and its optical property parameters to form a lookup table of biological tissue optical property parameters based on absorptivity.

[0008] Step c: For the reflectivity R at spatial frequencies f = 0 and f > 0 ref (f=0) and R ref Given a known reference phantom (f > 0), a DMD projector projects a cosine illumination pattern with a spatial frequency of f > 0 onto the surface of the reference phantom under computer control. A CCD camera then acquires the cosine fringe image I of the reference phantom surface. ref (x,y), where x and y are the pixel coordinates of the CCD camera image plane, I ref (x,y) represents the pixel grayscale value;

[0009] Step d: For the cosine fringe image I ref Perform a two-dimensional Fourier transform on (x,y) to obtain the cosine fringe image I. ref The spectrum of (x,y) is obtained, and then two-dimensional frequency domain filtering is performed on this spectrum to separate and extract its zero-frequency spectral component and first-order spectral component. Then, inverse Fourier transform is performed on the zero-frequency spectral component and the first-order spectral component respectively to obtain the DC component image. and the image of the component modulation system

[0010] Step e: Replace the reference phantom with the tested biological tissue, and repeat steps c and d to obtain the DC component image I of the cosine fringe image I(x,y) on the surface of the tested biological tissue. DC (x,y) and AC component modulation plot I AC (x,y);

[0011] Step f: For each pixel (x, y), use the following formula

[0012]

[0013] and

[0014]

[0015] Calculations were performed to obtain the absorbance values ​​T of the tested biological tissue at spatial frequencies f = 0 and f > 0. DC (f=0) and T AC (f > 0);

[0016] Step g, according to the following formula

[0017]

[0018] Calculate the Euclidean distance L between the measured absorbance value and the simulated absorbance value of each sample in the biological tissue optical property parameter lookup table based on absorbance in step b, and select the absorption coefficient μ corresponding to the simulated sample with the smallest Euclidean distance. a Reduced scattering coefficient μ s The measurement results of the optical properties parameters of the biological tissue are used to obtain the optical property parameter image of the tested biological tissue.

[0019] The aforementioned spatial-frequency domain imaging method for biological tissue optical characteristic parameters based on absorptivity is implemented on a spatial-frequency domain imaging system for biological tissue optical characteristic parameters based on absorptivity. This system includes a computer, a laser, a diffuser, a collimating lens, a DMD projector, a cosine illumination pattern, a reference phantom, a filter, a CCD camera, a base, a precision lifting stage and its control system, and the biological tissue being measured. In steps c and d, the computer is used to generate the cosine fringe pattern and control the laser, DMD projector, CCD camera, base, precision lifting stage and its control system. The laser is used as a single-wavelength laser. A long illumination source, diffuser, and collimating lens are used to diffuse and collimate to form a parallel illumination beam. A DMD projector is used to project a cosine light illumination pattern. A filter and a CCD camera are used to receive single-wavelength reflected light for imaging. A reference phantom and the biological tissue under test are used to reflect the cosine light illumination pattern. The reference phantom and the biological tissue under test can be interchangeably fixed horizontally on the base. The base, the precision lifting platform, and its control system can be adjusted for translation, lifting, and pitch angles so that the upper surfaces of the reference phantom and the biological tissue under test are at the same horizontal height and orientation. The DMD projector and the CCD camera are placed face down, with a small angle between their principal optical axes and overlapping fields of view.

[0020] Beneficial effects: The method of this invention obtains the absorptivity by measuring the reflectivity of biological tissue, thus eliminating the need for polarizers and avoiding the low imaging sensitivity caused by polarizers. Furthermore, since the optical characteristic parameters are inverted based on the absorptivity, the inversion error inherent in the principle of optical characteristic parameter inversion based on diffuse reflectivity is avoided. This effectively overcomes the shortcomings of existing spatial frequency domain imaging methods for biological tissue optical characteristic parameters, which suffer from low imaging sensitivity and inherent inversion errors due to the use of polarizers and inversion based on diffuse reflectivity. Attached Figure Description

[0021] Figure 1 This is a schematic diagram of the system composition and structure;

[0022] Figure 2 This is a flowchart of the method of the present invention;

[0023] In the figure, 1—computer, 2—laser, 3—diffuser, 4—collimating lens, 5—DMD projector, 6—cosine illumination pattern, 7—reference phantom, 8—filter, 9—CCD camera, 10—base and precision lifting platform and its control system, 11—biological tissue under test. Detailed Implementation

[0024] The specific embodiments of the present invention will be described in further detail below with reference to the accompanying drawings.

[0025] The device composition and arrangement are shown in the diagram below. Figure 1 As shown, the device consists of a computer 1, a laser 2, a diffuser 3, a collimating lens 4, a DMD projector 5, a cosine illumination pattern 6, a reference phantom 7, a filter 8, a CCD camera 9, a base and a precision lifting platform and its control system 10, and the biological tissue to be tested 11. The DMD projector 5 is positioned downwards with its main optical axis pointing vertically downwards, and its lens is 60 cm from the surface of the reference phantom. The CCD camera 9 is also positioned downwards, with its main optical axis forming a 15° angle with the main optical axis of the DMD projector 5. The fields of view of both the DMD projector 5 and the CCD camera 9 coincide on the surface of the reference phantom 7. The reference phantom 7 and the biological tissue to be tested 11 are horizontally placed and fixed on the base, and their surface shapes are identical. Under the control of the computer 1, the base and the precision lifting platform and its control system 10 perform translation, lifting, and pitch angle adjustments to ensure that the upper surfaces of the reference phantom 7 and the biological tissue to be tested are at the same horizontal level and orientation.

[0026] according to Figure 1 As shown, a 633nm narrow-linewidth laser 2 is used as a single-wavelength illumination source. Its emitted light is expanded by a diffuser 3 and then collimated by a collimating lens 4 to form parallel light, which serves as the illumination source for the DMD projector 5. Under the control of a computer 1, the DMD projector 5 forms a cosine illumination pattern 6 and illuminates the surface of a reference phantom 7 or the biological tissue 11 being tested. The reflected light is filtered by a filter 8 to remove stray light and then enters a CCD camera 9 to form a striped image. The image is then sent to the computer 1 for processing and calculation to obtain an image of the optical characteristic parameters of the biological tissue.

[0027] After the system is built and adjusted, the specific process of the biological tissue optical property parameter imaging method based on absorption rate of this invention is as follows: Figure 2 As shown, it includes the following steps:

[0028] Step a: Set the shape of the biological tissue simulation sample to an infinitely large plate, the thickness to d = 5 cm, the anisotropy coefficient to g = 0.9, the refractive index to n2 = 1.35, the refractive index of the surrounding medium to n1 = 1, and the total number of grid lines in the depth z direction to n. z =500, the total number of grid lines in the direction of radius r is set to n l =600, grid line spacing (resolution) set to Δ = 0.01cm, incident photon count set to m = 10 7 The wavelength is set to λ = 633 nm and the absorption coefficient is set to μ. a ∈{0.04cm -1 0.10cm -1 0.30cm -1 1.00cm -1 3.00cm -1 The reduced scattering coefficient is set to μ. s ′∈{4.00cm -1 7.00cm -1 10.00cm -1 20.00cm -1 30.00cm -1 40.00cm -1 The spatial frequency of the illumination light is set to f = 0 cm. -1 and f = 5cm -1 ;

[0029] Step b: Using the Monte Carlo simulation parameters set in step a, perform Monte Carlo simulation to obtain the spatial domain impact response function of the absorptivity of 30 simulated biological tissue samples, formed by the combination of 5 absorption coefficients and 6 reduced scattering coefficients. Then, at a spatial frequency f = 0 cm⁻¹... -1 and f = 5cm -1 A one-dimensional zero-order Hankel transform is performed to obtain the simulated absorbance value T(0cm) for each simulated sample. -1 ) and T(5cm -1 ), construct the simulated absorption rate value T(0cm) of the simulated sample. -1 ) and T(5cm -1 Its optical characteristic parameter μ a and μ s The mapping relationship between the ′ is used to form a lookup table of the optical properties of biological tissues based on absorption rate, as shown in Table 1;

[0030] Table 1. Lookup table for optical property parameters of biological tissues based on absorption rate.

[0031]

[0032]

[0033] Step c: Select its reflectivity R ref (0cm -1 ) and R ref (5cm -1 Using a known diffuse reflector as a reference phantom 7, the DMD projector 5, under the control of computer 1, projects a spatial frequency f = 5 cm onto the surface of the reference phantom 7. -1 The cosine illumination pattern 6 is obtained, and then the cosine fringe image I on the surface of the reference phantom 7 is acquired using a CCD camera 9. ref (x,y), where x and y are the pixel coordinates of the 9-plane image of the CCD camera, I ref (x,y) represents the pixel grayscale value;

[0034] Step d: For the cosine fringe image I ref Perform a two-dimensional Fourier transform on (x,y) to obtain the cosine fringe image I. ref The spectrum of (x,y) is obtained, and then two-dimensional frequency domain filtering is performed on this spectrum to separate and extract its zero-frequency spectral component and first-order spectral component. Then, inverse Fourier transform is performed on the zero-frequency spectral component and the first-order spectral component respectively to obtain the DC component image. and the image of the component modulation system

[0035] Step e: Replace the reference phantom 7 with the flat-surfaced biological tissue 11 being tested, and repeat steps c and d to obtain the DC component image I of the cosine fringe image I(x,y) of the biological tissue 11 being tested. DC (x,y) and AC component modulation plot I AC (x,y);

[0036] Step f: For each pixel (x, y), use the following formula

[0037]

[0038] and

[0039]

[0040] Calculations were performed to obtain the spatial frequency f = 0 cm⁻¹ of the tested biological tissue 11. -1 and f = 5cm -1 Absorption rate measurement T DC (0cm -1 ) and T AC (5cm -1 );

[0041] Step g, according to the following formula

[0042]

[0043] The Euclidean distance L between the measured absorbance value and the simulated absorbance value of each sample in Table 1 was calculated, resulting in 30 Euclidean distances. The absorption coefficient μ corresponding to the simulated sample with the smallest Euclidean distance was selected. a Reduced scattering coefficient μ s The measurement results of the optical properties parameters of the biological tissue are used to obtain the optical property parameter image of the tested biological tissue.

[0044] It should be noted that the technical field corresponding to the technical solution of this invention is the field of optical measurement and imaging technology. For those skilled in the art, the devices involved in the method of this invention can be selected and applied by those skilled in the art based on their professional knowledge. There is no need to provide specific explanations of the principles and descriptions of the instruments and equipment in this invention. The Monte Carlo simulation and one-dimensional zero-order Hankel transform in step b, the two-dimensional Fourier transform and two-dimensional frequency domain filtering and the inverse Fourier transform in step d are all well-known knowledge and technologies in the field. Those skilled in the art can fully implement them independently. This invention has been fully disclosed.

[0045] The following analysis illustrates that the spatial frequency domain imaging method for biological tissue optical property parameters based on absorption rate of the present invention has the technical advantages of high sensitivity and small inversion error compared with existing spatial frequency domain imaging methods for biological tissue optical property parameters.

[0046] Existing spatial frequency domain imaging methods for biological tissue optical properties first measure the diffuse reflectance of the biological tissue and then deduce the optical properties based on the diffuse reflectance. To accurately measure diffuse reflectance, a polarizer is placed between the light source and the tissue to generate polarized illumination, but this reduces the illumination energy by 50%. A polarizer is also placed between the tissue and the camera to eliminate specular reflection, but this reduces the diffuse reflection light energy by 50%. Furthermore, the measured diffuse reflectance includes the influence of diffuse reflection caused by the roughness of the biological tissue interface, while current optical property parameter inversion methods utilize backscattered light generated after entering the biological tissue, thus introducing an inherent inversion error.

[0047] The method proposed in this invention first measures the absorptivity of biological tissue, and then derives optical characteristic parameters based on the absorptivity. The absorptivity can be measured by directly illuminating the biological tissue with a light source, and then directly measuring all reflected light, including specular reflection, interfacial diffuse reflection, and backscattered reflection, to obtain the absorptivity. This avoids the energy reduction problem of illumination and backscattered light by eliminating the need for polarized light, and the inversion using absorptivity is theoretically error-free. In comparison, the method proposed in this invention has the advantages of high sensitivity and the absence of inversion error in principle.

Claims

1. A spatial frequency domain imaging method for biological tissue optical property parameters based on absorptivity, characterized in that, Includes the following steps: Step a: Based on the properties of biological tissue, set the Monte Carlo simulation parameters for the biological tissue, including the geometry and size of the simulated biological tissue sample, anisotropy coefficient, refractive index, refractive index of the surrounding medium, number and spacing of grid lines, number and wavelength of incident photons, optical characteristic parameters, and spatial frequency of the illumination light; the optical characteristic parameters include the absorption coefficient. Reduced scattering coefficient ; Step b: Using the Monte Carlo simulation parameters for biological tissue set in step a, perform Monte Carlo simulation to obtain the spatial domain impact response function of the absorptivity corresponding to the optical property parameters of the simulated sample. Then, perform a one-dimensional zero-order Hankel transform to obtain the zero-frequency response. and communication space frequency Simulated absorbance values ​​of the simulated sample and A mapping relationship between the simulated absorptivity values ​​of the simulated sample and its optical property parameters was constructed, and a lookup table of biological tissue optical property parameters based on absorptivity was formed. Step c, for spatial frequencies of and Reflectivity and Given a known reference phantom, a DMD projector, under computer control, projects a spatial frequency onto the surface of the reference phantom. The cosine illumination pattern is obtained, and then the cosine fringe image of the reference phantom surface is acquired using a CCD camera. ,in and For CCD camera image plane pixel coordinates, The pixel grayscale value; Step d: For the cosine fringe image Perform a two-dimensional Fourier transform to obtain a cosine fringe image. The spectrum is then analyzed, and two-dimensional frequency domain filtering is performed on this spectrum to separate and extract its zero-frequency spectral component and first-order spectral component. Finally, inverse Fourier transforms are performed on the zero-frequency spectral component and the first-order spectral component respectively to obtain the DC component image. and the image of the component modulation system ; Step e: Replace the reference phantom with the biological tissue to be tested, and repeat steps c and d to obtain the DC component image and AC component modulation image of the cosine fringe image on the surface of the biological tissue to be tested. Step f: For each pixel, use the following formula and Calculations were performed to obtain the spatial frequency of the tested biological tissue. and Absorption rate measurement values ​​below and ; Step g, according to the following formula Calculate the Euclidean distance between the measured absorptivity and the simulated absorptivity value for each sample in the absorptivity-based biological tissue optical property parameter lookup table in step b. L The absorption coefficient corresponding to the simulation sample with the smallest Euclidean distance is selected. Reduced scattering coefficient The measurement results of the optical properties parameters of biological tissues result in an image of the optical properties parameters of the tested biological tissue.

2. The spatial frequency domain imaging method for biological tissue optical characteristic parameters based on absorptivity according to claim 1, characterized in that, This study utilizes a spatial frequency domain imaging system for biological tissue optical characteristic parameters based on absorptivity. The system comprises a computer, a 633nm narrow-linewidth laser, a diffuser, a collimating lens, a DMD projector, a reference phantom, filters, a CCD camera, a base, a precision lifting platform and its control system, and the biological tissue to be measured. The 633nm narrow-linewidth laser serves as a single-wavelength illumination source. Its emitted light is expanded by the diffuser and then collimated by the collimating lens to form parallel light, which serves as the illumination source for the DMD projector. Under computer control, the DMD projector forms a cosine illumination pattern and illuminates the surface of the reference phantom or the biological tissue to be measured. The reflected light is filtered by filters to remove stray light before entering the CCD camera to form a striped image, which is then processed by the computer to obtain the biological tissue optical characteristic parameters. Image; In steps c and d, a computer is used to generate a cosine fringe pattern and control a laser, a DMD projector, a CCD camera, a base, a precision lifting platform, and their control systems. The laser is used as a single-wavelength illumination source, a diffuser and a collimating lens are used to diffuse and collimate to form a parallel illumination beam, the DMD projector is used to project a cosine light illumination pattern, the filter and the CCD camera are used to receive single-wavelength reflected light for imaging, and the reference phantom and the tested biological tissue are used to reflect the cosine light illumination pattern. The reference phantom and the tested biological tissue can be interchangeably fixed horizontally on the base. The base and the precision lifting platform and their control systems can be translated, lifted, and tilted to ensure that the upper surfaces of the reference phantom and the tested biological tissue are at the same horizontal height and orientation. The DMD projector and the CCD camera are placed face down, with a small angle between their principal optical axes and overlapping fields of view.

Images