Rapid high-resolution three-dimensional microscopic imaging method for non-diffraction light field

Abstract

The invention discloses a rapid high-resolution three-dimensional microscopic imaging method for a non-diffraction light field, which belongs to the technical field of optical microscopic imaging, and comprises the following steps of: designing a mixed phase mask, applying the mixed phase mask to an image space frequency domain surface of an inverted fluorescence microscope to modulate the light field, and generating a multi-channel non-diffraction light beam array; three-dimensional coding of object information is achieved, then array information is collected through a camera, and three-dimensional body imaging is achieved through a reconstruction algorithm. According to the invention, high-resolution three-dimensional imaging of single exposure is realized; according to the invention, artifact noise generated in the imaging process is reduced, and the spatial resolution of the imaging system is improved. According to the imaging method, high-speed imaging can be carried out at the speed of 100 frames per second, the imaging depth of 30 micrometers or above can be achieved under the 100-time objective lens, the imaging view field width is larger than 40 micrometers, and the spatial resolution is close to the optical diffraction limit.

Description

TECHNICAL FIELD

[0001] The present application belongs to the field of optical microscopic imaging technology, and relates to a kind of diffraction-free light field fast high-resolution three-dimensional microscopic imaging method. BACKGROUND

[0002] Optical microscope is an important tool for biomedical imaging. Fluorescence microscope can obtain high-resolution and high-contrast images in complex biological tissues by using the selective excitation and emission characteristics of fluorescent probes, and is particularly suitable for applications such as live cell dynamics, tumor heterogeneity analysis and deep tissue imaging. However, conventional fluorescence microscope mainly records single focal plane information, and three-dimensional data acquisition usually requires layer-by-layer scanning along the axial direction, which limits the imaging speed and is accompanied by obvious photobleaching and phototoxicity. The research and development of three-dimensional fluorescence microscopic imaging mainly focuses on three key indicators: resolution, imaging speed and effective depth of field. Existing technologies can be roughly divided into the following three categories: (1) Axial scanning technology, represented by point scanning confocal microscope, rotating disk confocal microscope and multi-photon microscope. The pinhole or nonlinear local excitation effectively suppresses the out-of-focus background, and the resolution is high, which is suitable for thick sample observation. However, the serial scanning mode limits the volume imaging speed, which is not conducive to the fast imaging of large volume samples, and the phototoxicity is relatively high. (2) Light sheet illumination technology, selective plane illumination (SPIM) uses thin light sheet to laterally excite the sample and orthogonally images, which realizes low background, low phototoxicity and large volume fast microscopic imaging. This kind of technology is highly sensitive to optical system alignment, sample transparency and refractive index matching, and the system complexity is high; the performance is limited in samples with uneven refractive index or strong scattering. (3) Parallel coding and light field regulation technology, represented by three-dimensional structured light illumination microscope (3D-SIM), multi-focal plane detection and light field microscope. Three-dimensional structured light illumination microscope not only needs axial scanning, but also needs multiple angle and multiple phase repeated exposure, which limits the imaging speed and increases the phototoxicity; multi-focal plane detection technology can observe multiple planes simultaneously, but it is difficult to cover the entire sample volume, and the field of view, axial range and resolution are mutually restricted; light field microscope encodes the volume information into the detection plane by modulating the fluorescence, and then uses reconstruction algorithm to restore the three-dimensional data, which can realize high-speed volume imaging, but is limited by the depth of field, spatial resolution and signal-to-noise ratio, and is prone to artifacts in the three-dimensional reconstruction process. In summary, the existing three-dimensional fluorescence microscopic imaging technology cannot simultaneously achieve high resolution, fast volume imaging and large depth of field imaging. SUMMARY

[0003] To address the shortcomings of the aforementioned technologies, this invention proposes a rapid, high-resolution three-dimensional microscopic imaging method without diffraction fields. This method solves the problem that existing technologies cannot simultaneously achieve both large depth of field and high resolution. The method performs multi-channel phase modulation of the light field in the frequency domain to generate a diffraction-free beam array for parallel encoding of the sample's three-dimensional information. A two-dimensional image of the encoded sample is captured in a single exposure on the image plane, and a three-dimensional reconstruction algorithm is used to recover the sample's three-dimensional volumetric structure data, thereby achieving rapid, high-resolution three-dimensional microscopic imaging. This method is applicable to fluorescence microscopy and also to microscopic imaging applications of transparent and self-illuminating samples.

[0004] The technical solution proposed in this invention is as follows: A fast, high-resolution three-dimensional microscopic imaging method without diffraction field, comprising the following steps:

[0005] S1. Constructing a diffraction-free light field modulation mask: An imaging system is built based on a fluorescence microscope. The imaging system includes a frequency domain phase modulation system, comprising a first lens, a spatial light modulator, and a second lens that are symmetrically assembled with the same focal length. The distances between the original image plane and the first lens, the first lens and the spatial light modulator, the spatial light modulator and the second lens, and the second lens and the camera are all the focal lengths of the first lens. A light field modulation mask containing multiple modulation channels is loaded onto the spatial light modulator. The modulation phase of the mask includes a bending propagation phase, a quadratic central convergence beam modulation phase, a blazed grating, and a one-dimensional slit. After modulation, a diffraction-free beam array is generated, and the three-dimensional information of the sample is encoded in parallel to obtain multi-angle projection images.

[0006] This imaging method is based on the working principle of fluorescence microscopy. When the sample is irradiated by a laser, it emits fluorescence. This fluorescence is received by the microscope objective, passed through a dichroic mirror, and then focused by the tube lens, forming a magnified real image of the sample on the original image plane. The magnified image is then limited in size by an aperture and polarized by a polarizer. It is then passed through a frequency-domain phase modulation system consisting of a first lens, a reflective or transmissive spatial light modulator, and a second lens. The resulting phase mask is loaded onto the spatial light modulator for light field modulation, and the modulated image is received by a camera. The image captured by the camera is processed by a 3D reconstruction algorithm to ultimately recover the 3D information of the sample.

[0007] S2. High-resolution 3D reconstruction: Discretize the 2D coded image captured by a single camera exposure, construct a linear relationship between the projection data and the 3D information of the sample, and transform the 3D reconstruction into an optimization problem with confidence constraints and total variation regularization terms. Iteratively solve the problem using the alternating direction multiplier method, stopping after 50 iterations to suppress noise amplification and restore the 3D volumetric structure data of the sample.

[0008] Furthermore, in step S1, the spatial light modulator is either reflective or transmissive, and the imaging system also includes an objective lens, a dichroic mirror, a tube mirror, an aperture, and a polarizer. The fluorescence emitted by the sample after laser irradiation passes sequentially through the objective lens, dichroic mirror, tube mirror, aperture, and polarizer before entering the frequency domain phase modulation system.

[0009] Furthermore, in step S1, the light field modulation mask consists of 6 modulation channels, and a static modulation pattern is loaded onto the spatial light modulator. It consists of four parts:

[0010] The first part is the phase of bending propagation. It satisfies the following formula:

[0011] in,

[0012] In the formula, In planar coordinates; and The range is determined by the number of pixels and the pixel size of the spatial light modulator; The distance of a single modulation channel from the center of the plane. The value is 1 / 3 of the effective width of the spatial light modulator; The scaling factor for spatial coordinates;

[0013] The second part is the quadratic modulation phase of the center converging beam, which satisfies the following formula:

[0014] (2)

[0015] In the formula, It is a constant;

[0016] The third part is a blazed grating used to separate the unmodulated light field from the modulated light field. During the imaging process, the grating period needs to be adjusted according to the field of view size to make full use of the camera elements.

[0017] The fourth part is a one-dimensional slit that cuts off the cubic phase of the first part in order to reduce the impact of the point spread function sidelobe on the imaging resolution. The slit size needs to be optimized according to the requirements of imaging resolution and light intensity response.

[0018] Furthermore, in step S2, in order to recover the three-dimensional information of the sample from the two-dimensional image captured by the camera, the projected images of multiple channels in the image are discretized, and the formula for expressing the projected image generated for each channel is as follows:

[0019] (3)

[0020] In the formula, Projection data recorded by the camera, This is the number of the projection channel; The projection matrix for each channel is composed of non-diffractive beams propagating in different directions; This refers to the three-dimensional information of the sample.

[0021] The so-called three-dimensional information reconstruction is to solve the system of linear equations represented by formula (3). However, because the number of projections is small, the projection matrix is ​​difficult to satisfy the positive semi-definite condition, which means that formula (3) cannot be solved directly. Therefore, the solution of formula (3) is transformed into an optimization problem, described by formula (4):

[0022] (4)

[0023] In the formula, is the weight coefficient for TV regularization, with a value range of (0, 1);

[0024] In the formula, the two terms on the right-hand side are the confidence constraint and the total variation (TV) regularization term, respectively. The confidence constraint is used to ensure the accuracy and reliability of the reconstruction results. The purpose of TV regularization is to introduce continuity constraints and suppress artifacts generated during reconstruction. Equation (4) is solved using the Alternating Direction Multiplier Method (ADMM). The optimal solution is found through an iterative algorithm, and the iterative equation is:

[0025] (5)

[0026] In the formula, This represents the number of iterations. It is a vector in which all elements are equal to 1; The divergence of the vector field is calculated. Using the iterative algorithm represented by formula (5), the calculation result will converge to the vicinity of the optimal solution of equation (4). In order to avoid the generation and amplification of noise, the iteration is stopped in time after 50 iterations, thus obtaining the three-dimensional information of the sample.

[0027] Another objective of this invention is achieved through a diffraction-free, high-resolution three-dimensional microscopic imaging system. The system employs the diffraction-free, high-resolution three-dimensional microscopic imaging method described above, comprising: an imaging optical path module, an optical field modulation module, and a three-dimensional reconstruction module. The imaging optical path module includes a laser source, an objective lens, a dichroic mirror, a tube lens, an aperture, a polarizer, a frequency-domain phase modulation system, and a camera. The frequency-domain phase modulation system consists of a first lens, a spatial light modulator, and a second lens, used to generate a diffraction-free beam array and acquire a two-dimensional coded image of the sample. The optical field modulation module is used to load an optical field modulation mask onto the spatial light modulator to achieve multi-channel phase modulation of fluorescence. The three-dimensional reconstruction module is used to execute a three-dimensional reconstruction algorithm to recover the three-dimensional volumetric structure data of the sample from the two-dimensional coded image.

[0028] Furthermore, the objective lens of the imaging optical path module is a 100x NA 1.45 objective lens, the laser light source wavelength is 647 nanometers, the camera exposure time is set to 10 milliseconds, the three-dimensional imaging rate is 100Hz, the imaging depth is 30 micrometers, and the field of view is not less than 40 micrometers.

[0029] Advantages and benefits of this invention: This invention performs multi-channel phase modulation of the fluorescence emitted by the sample in the frequency domain, generating a diffraction-free beam array with lateral self-acceleration characteristics, thus encoding three-dimensional information into a two-dimensional plane; by acquiring multi-angle projections of the sample in parallel with a single exposure, and combining this with a three-dimensional reconstruction algorithm to recover the three-dimensional volume data of the sample, rapid and high-resolution three-dimensional microscopic imaging is achieved. The imaging system proposed in this invention can be built on a conventional fluorescence microscope platform, exhibiting strong versatility and compatibility. Since a single three-dimensional imaging requires only a single exposure, the imaging speed is limited only by the camera's frame rate; under current commercial camera conditions, the three-dimensional imaging rate can reach 100Hz; benefiting from the stable propagation characteristics of the diffraction-free beam, the imaging depth can reach 30 micrometers. The imaging field of view is limited by the area size of the detector camera, typically not less than 40 micrometers. The three-dimensional reconstruction algorithm ensures that the imaging resolution is close to the optical diffraction limit, typically 300-600 nanometers. Attached Figure Description

[0030] Figure 1 Diagram of a reflective imaging system;

[0031] Figure 2 Structure diagram of a transmission imaging system;

[0032] Figure 3 A static modulation pattern loaded onto a spatial light modulator;

[0033] Figure 4 Image of a volumetric fluorescent nanoparticle with a thickness of 30 micrometers;

[0034] Figure 5 The image shows the measurement results of the resolution of the imaging system in the x, y, and z directions.

[0035] Figure 6 Images of mouse renal tubules at different depths with a thickness of 16 micrometers (2 micrometer intervals).

[0036] Figure 7 Tracking imaging of calcium ions in mouse cardiomyocytes (10ms interval).

[0037] Among them, 1. Sample, 2. Objective lens, 3. Light source, 4. Dichroic mirror, 5. Tube lens, 6. Mirror, 7. Aperture, 8. Polarizer, 9. First lens, 10. Camera, 11. Second lens, 12. Spatial light modulator, 13. Mirror. Detailed Implementation

[0038] The invention will be further described below with reference to the accompanying drawings:

[0039] Example 1

[0040] A fast, high-resolution three-dimensional microscopic imaging method without diffraction field is as follows:

[0041] (1) Design of a mask without diffraction field modulation

[0042] The imaging system in this embodiment can be built in two ways, namely... Figure 1 The reflected imaging system structure shown is as follows: Figure 2 The transmissive imaging optical path structure is shown.

[0043] The imaging system works as follows: When the sample is irradiated by a laser, it emits fluorescence. This fluorescence is collected by the microscope objective, filtered by a dichroic mirror, and then focused by a tube lens, forming a magnified real image of the sample on the original image plane. The magnified image is then limited in size by an aperture and polarized by a polarizer. It is then passed through a frequency-domain phase modulation system consisting of a first lens, a reflective or transmissive spatial light modulator, and a second lens. The modulated image is received by a camera. The image captured by the camera is processed by a 3D reconstruction algorithm to ultimately recover the sample's 3D information. In the frequency-domain phase modulation system, the first and second lenses have the same focal length but are symmetrically assembled. The distances between the original image plane and the first lens, the first lens and the spatial light modulator, the spatial light modulator and the second lens, and the second lens and the camera are all equal to the focal length of the first lens. The difference between these two types of imaging systems lies in the spatial light modulator's operating mode: reflective modulation or transmissive modulation.

[0044] This example, aimed at achieving the ultimate goal of three-dimensional microscopic imaging, requires loading a static modulation pattern onto the spatial light modulator. It consists of four parts. The first part is the bending propagation phase, which satisfies the following formula:

[0045] (1)

[0046] In the formula, For planar coordinates, and The range is determined by the number of pixels and the pixel size of the spatial light modulator; The distance of a single modulation channel from the center of the plane. The value is 1 / 3 of the effective width of the spatial light modulator; This is a scaling factor for spatial coordinates, used to adjust the size of the phase pattern.

[0047] The second part is the quadratic modulation phase of the center converging beam, which satisfies the following formula:

[0048] (2)

[0049] In the formula, It is a constant.

[0050] The third part is a blazed grating used to separate the unmodulated light field from the modulated light field, with the grating period adjusted to 5 pixels according to the field size.

[0051] The fourth part is a slit with a spacing of 50-200 pixels that cuts off the cubic phase of the first part.

[0052] The light beams emitted by fluorescent molecules in the sample are modulated by these 6 channels to generate a diffraction-free beam array that can propagate by self-acceleration. The diffraction-free beams can propagate stably over a long depth of field, and their lateral self-acceleration characteristics can encode the three-dimensional information of the sample into a two-dimensional image. Using the diffraction-free beam array generated by the 6 channels, projection images of the sample from 6 different angles can be obtained simultaneously.

[0053] (2) High-resolution 3D reconstruction algorithm

[0054] To recover the three-dimensional information of a sample from a two-dimensional image captured by a camera, the projected images of multiple channels in the image are first discretized. The projected image generated by each channel can be described by the following formula:

[0055] (3)

[0056] In the formula, Projection data recorded by the camera, This is the number of the projection channel; The projection matrix of each channel is composed of non-diffractive beams propagating in different directions, and can be obtained through theoretical calculation or experimental measurement. The three-dimensional information of the sample. The so-called three-dimensional information reconstruction is to solve the linear equation system represented by formula (3). However, because the number of projections is small, the projection matrix is ​​difficult to satisfy the positive semi-definite condition, which means that formula (3) cannot be solved directly. Therefore, the solution of formula (3) is transformed into an optimization problem, described by formula (4):

[0057] (4)

[0058] In the formula, the two terms on the right-hand side are the confidence constraint and the total variation (TV) regularization term, respectively. The weight coefficient for TV regularization is denoted by , and its value ranges from (0,1). The confidence constraint ensures the accuracy and reliability of the reconstruction results, while TV regularization introduces a continuity constraint to suppress artifacts generated during the reconstruction process. Equation (5) is solved using the Alternating Direction Multiplier Method (ADMM), and its optimal solution is found through an iterative algorithm. The iterative equation is:

[0059] (5)

[0060] In the formula, This represents the number of iterations. It is a vector in which all elements are equal to 1; The divergence of the vector field is calculated. Using the iterative algorithm represented by formula (5), the calculation result will quickly converge to the vicinity of the optimal solution of equation (4). In order to avoid the generation and amplification of noise, it is necessary to stop the iteration in time after 50 iterations to obtain the three-dimensional information of the sample.

[0061] Example 2

[0062] This embodiment uses Figure 1 The reflective imaging system used a 647 nm laser for illumination and a 100x objective lens with an NA of 1.45 to perform three-dimensional imaging of a 30 μm thick volumetric fluorescent nanoparticle, a 16 μm thick mouse renal tubule sample, and stained mouse cardiomyocytes. The phase modulation pattern was calculated according to formula (1) as follows: Figure 3 As shown, The value was set to 1.28 mm, the camera exposure time was set to 10 milliseconds, and a single exposure yielded six projected images of the sample from different angles. These projected images were processed using a 3D reconstruction algorithm, resulting in the reconstructed images shown below. Figures 4-6As shown. The system's spatial resolution is 349 nm in the x-direction, 343 nm in the y-direction, and 513 nm in the z-direction, achieving rapid, high-resolution three-dimensional microscopic imaging. Signal transduction of calcium ions in mouse cardiomyocytes was observed at 10 ms intervals, as shown... Figure 7 As shown.

Claims

1. A method for rapid, high-resolution three-dimensional microscopic imaging without diffraction field, characterized in that, Includes the following steps: S1. Constructing a diffraction-free light field modulation mask: An imaging system is built based on a fluorescence microscope. The imaging system includes a frequency domain phase modulation system, comprising a first lens, a spatial light modulator, and a second lens that are symmetrically assembled with the same focal length. The distances between the original image plane and the first lens, the first lens and the spatial light modulator, the spatial light modulator and the second lens, and the second lens and the camera are all the focal lengths of the first lens. A light field modulation mask containing multiple modulation channels is loaded onto the spatial light modulator. The modulation phase of the mask includes a bending propagation phase, a quadratic central convergence beam modulation phase, a blazed grating, and a one-dimensional slit. After modulation, a diffraction-free beam array is generated, and the three-dimensional information of the sample is encoded in parallel to obtain multi-angle projection images. S2. High-resolution 3D reconstruction: Discretize the 2D coded image captured by a single camera exposure, construct a linear relationship between the projection data and the 3D information of the sample, and transform the 3D reconstruction into an optimization problem with confidence constraints and total variation regularization terms. Iteratively solve the problem using the alternating direction multiplier method, stopping after 50 iterations to suppress noise amplification and restore the 3D volumetric structure data of the sample.

2. The method for rapid, high-resolution three-dimensional microscopic imaging without diffraction field according to claim 1, characterized in that, In step S1, the spatial light modulator is either reflective or transmissive, and the imaging system also includes an objective lens, a dichroic mirror, a tube mirror, an aperture, and a polarizer. The fluorescence emitted by the sample after laser irradiation passes sequentially through the objective lens, dichroic mirror, tube mirror, aperture, and polarizer before entering the frequency domain phase modulation system.

3. The method for rapid, high-resolution three-dimensional microscopic imaging without diffraction field according to claim 2, characterized in that, In step S1, the light field modulation mask consists of 6 modulation channels, and a static modulation pattern is loaded onto the spatial light modulator. It consists of four parts: The first part is the phase of bending propagation. It satisfies the following formula: in, In the formula, In planar coordinates; and The range is determined by the number of pixels and the pixel size of the spatial light modulator; The distance of a single modulation channel from the center of the plane. The value is 1 / 3 of the effective width of the spatial light modulator; The scaling factor for spatial coordinates; The second part is the quadratic modulation phase of the center converging beam, which satisfies the following formula: （2） In the formula, It is a constant; The third part is a blazed grating used to separate the unmodulated light field from the modulated light field. During the imaging process, the grating period needs to be adjusted according to the field of view size to make full use of the camera elements. The fourth part is a one-dimensional slit that cuts off the cubic phase of the first part in order to reduce the impact of the point spread function sidelobes on the imaging resolution.

4. The method for rapid, high-resolution three-dimensional microscopic imaging without diffraction field according to claim 3, characterized in that, In step S2, the projected images of multiple channels in the image are discretized, and the formula for expressing the projected image of each channel is as follows: （3） In the formula, Projection data recorded by the camera, This is the number of the projection channel; The projection matrix for each channel is composed of non-diffractive beams propagating in different directions; This refers to the three-dimensional information of the sample. Solving equation (3) is transformed into an optimization problem, described by equation (4): （4） In the formula, is the weight coefficient for TV regularization, with a value range of (0, 1); Equation (4) is solved using the alternating direction multiplier method, and its optimal solution is found through an iterative algorithm. The iterative equation is: （5） In the formula, This represents the number of iterations. It is a vector in which all elements are equal to 1; The divergence of the vector field is calculated. Using the iterative algorithm represented by formula (5), the calculation result will converge to the vicinity of the optimal solution of equation (4). After 50 iterations, the iteration is stopped in time, and the three-dimensional information of the sample is obtained.

5. A rapid, high-resolution three-dimensional microscopic imaging system without diffraction field, characterized in that, The method employs a diffraction-free, high-resolution three-dimensional microscopic imaging technique as described in any one of claims 1-4, comprising: an imaging optical path module, an optical field modulation module, and a three-dimensional reconstruction module. The imaging optical path module includes a laser source, an objective lens, a dichroic mirror, a tube lens, an aperture, a polarizer, a frequency-domain phase modulation system, and a camera. The frequency-domain phase modulation system consists of a first lens, a spatial light modulator, and a second lens, used to generate a diffraction-free beam array and acquire a two-dimensional coded image of the sample. The optical field modulation module is used to load an optical field modulation mask onto the spatial light modulator to achieve multi-channel phase modulation of fluorescence. The three-dimensional reconstruction module is used to execute a three-dimensional reconstruction algorithm to recover the three-dimensional volumetric structure data of the sample from the two-dimensional coded image.

6. The diffraction-free, high-resolution three-dimensional microscopic imaging system according to claim 5, characterized in that, The imaging optical path module uses a 100x objective lens with an NA of 1.45, a laser light source wavelength of 647 nanometers, a camera exposure time of 10 milliseconds, a three-dimensional imaging rate of 100 Hz, an imaging depth of 30 micrometers, and a field of view of not less than 40 micrometers.

Images