Self-adaptive optical-assisted three-dimensional fluorescence and light intensity diffraction tomography imaging method

Abstract

The invention discloses a self-adaptive optical assisted three-dimensional fluorescence and light intensity diffraction tomography imaging method. According to the method, the iteration lamination imaging technology is adopted, the three-dimensional refractive index and the wavefront aberration are effectively separated from the intensity image, the wavefront aberration solved in real time is combined, the wavefront aberration is accurately introduced into the point spread function of the imaging system, the three-dimensional fluorescence result is synchronously corrected, and therefore the reconstruction quality in the fluorescence mode is remarkably improved. According to the invention, the contradiction between the coordination of long-term imaging and high temporal-spatial resolution of the traditional three-dimensional bimodal imaging method is solved, and the method has important significance for exploring the structure and the dynamic state of a biological phenomenon on the cell and subcellular level.

Description

Technical Field

[0001] The present invention relates to three-dimensional microscopic imaging technology, in particular to an adaptive optics-assisted three-dimensional fluorescence and light intensity diffraction tomography imaging method. Background Art

[0002] To ensure clear, continuous, high-quality images during long-term observations, researchers use the Olympus IX-ZDC z-axis drift compensator for microscopes. This ensures consistently in-focus images over several days of observation, supporting high-precision and rapid cellular assay analysis. Recently, a technique known as computational adaptive optics has been developed that algorithmically corrects for time-varying aberrations in imaging systems, eliminating the need for additional hardware compensation. Building on this concept, researchers have leveraged computational adaptive optics to directly recover and correct aberrations in Fourier microscopy, leveraging inherent data redundancy to achieve high-quality, long-term quantitative phase imaging. Furthermore, other techniques employ aberration-corrected intensity transfer quantitative phase imaging to eliminate spatially inhomogeneous and temporally varying aberrations, enabling long-term, high-throughput quantitative phase imaging. However, these approaches have been applied to quantitative phase imaging, and there have been no reports applying adaptive optics-assisted techniques to address time-varying aberrations in three-dimensional dual-modality imaging. Summary of the Invention

[0003] The object of the present invention is to provide an adaptive optics-assisted three-dimensional fluorescence and light intensity diffraction tomography imaging method.

[0004] The technical solution for achieving the purpose of the present invention is: an adaptive optics-assisted three-dimensional fluorescence and light intensity diffraction tomography imaging method, the steps of which are as follows:

[0005] Step 1: Illuminate the sample from different angles and collect intensity images;

[0006] Step 2: Use iterative stacking imaging technology to process the intensity image to obtain the three-dimensional spectrum after removing the aberration and the aperture function with aberration;

[0007] Step 3: Perform a 3D inverse Fourier transform on the 3D spectrum after removing the aberration to obtain the 3D refractive index distribution of the reconstructed sample, thus achieving label-free 3D imaging of the measured sample.

[0008] Step 4: Excite the sample for fluorescence and collect a fluorescence image stack;

[0009] Step 5: Introduce the aberration into the point spread function of the imaging system and apply the 3D Richardson-Lucy deconvolution algorithm to correct the 3D fluorescence results.

[0010] Preferably, an adaptive optics-assisted three-dimensional fluorescence and light intensity diffraction tomography imaging system is used to collect intensity images. The adaptive optics-assisted three-dimensional fluorescence and light intensity diffraction tomography imaging system includes a programmable LED, a microscope objective, a fluorescence excitation block, an imaging tube lens, a reflector and a camera. The programmable LED is coaxial with the microscope objective, the fluorescence excitation block and the imaging tube lens, and is placed at a predetermined height above the sample to be measured; the camera is placed on the imaging focal plane; during illumination imaging, the programmable LEDs are lit one by one, and quasi-monochromatic plane waves illuminate the sample from different angles according to the position of the LEDs, and converge onto the camera plane after passing through the microscope objective, the fluorescence excitation block, the imaging tube lens and the reflector, thereby recording a series of original light intensity images on the camera plane.

[0011] Preferably, in step 2, the intensity image is processed using an iterative stacking imaging technique to obtain a three-dimensional spectrum after removing the aberration and an aperture function with the aberration. The specific steps are as follows:

[0012] Step 2.1, initialize the three-dimensional spectrum and determine the three-dimensional spectrum information according to the incident wave vector and the generalized three-dimensional coherent transfer function;

[0013] Step 2.2: Project the three-dimensional spectrum along the axial frequency coordinate and transfer it to the center of the coordinate to obtain the two-dimensional spectrum.

[0014] Step 2.3, perform aberration compensation on the two-dimensional spectrum to obtain the compensated two-dimensional spectrum

[0015]

[0016] Where P(k T )′ is the aperture function with aberration;

[0017] Step 2.4, calculate the first-order scattering potential based on the compensated two-dimensional spectrum:

[0018]

[0019] Where j is the imaginary unit, k z is the axial wave vector;

[0020] Step 2.5: Combine the intensity image and impose an amplitude constraint on the first-order scattering potential to obtain the constrained scattering potential:

[0021]

[0022] Step 2.6, for the aperture function P(k T )′ to update:

[0023]

[0024] Where β represents the update step size, O is the intensity map I i (k T ) spectrum after Fourier transformation;

[0025] Step 2.7, calculate the two-dimensional spectrum after applying amplitude constraints:

[0026]

[0027] Step 2.8, perform aberration removal on the two-dimensional spectrum:

[0028]

[0029] After de-aberration, the two-dimensional spectrum is mapped to the three-dimensional Ewald shell to achieve synchronous update of the three-dimensional spectrum. After the update, it is determined whether the intensity images taken at all angles have been iterated. If so, proceed to the next step; otherwise, return to step 2.2.

[0030] Preferably, the specific method of performing fluorescence excitation on the sample and collecting the fluorescence image stack is:

[0031] The sample is excited for fluorescence, and the microscope objective lens moves along the z-axis to synchronously trigger the camera (7) to take a picture. The fluorescence passes through the fluorescence excitation block and then reflects through the microscope objective lens to the sample to be tested. The fluorescence is then reflected by the sample to be tested through the microscope objective lens, the fluorescence excitation block, the imaging tube lens, and the reflector to converge on the camera plane, thereby recording a fluorescence image stack on the camera (7) plane.

[0032] Preferably, step 5 introduces the aberration into the point spread function of the imaging system and uses the three-dimensional Richardson-Lucy deconvolution algorithm to correct the three-dimensional fluorescence results. The specific steps are as follows:

[0033] Step 5.1: Compare the optical transfer function of the imaging system with the aperture function P(k T )′, the optical transfer function with aberration is obtained:

[0034] H′(u)=H(u)P(k T )′

[0035] Where H(u) is the optical transfer function of the imaging system;

[0036] Step 5.2, performing a three-dimensional inverse Fourier transform on the optical transfer function H′(u) with aberration to obtain the point spread function h′(u);

[0037] In step 5.3, a three-dimensional Richardson-Lucy deconvolution operation is performed on the fluorescence image stack in combination with the point spread function h′(u) to obtain an optimized fluorescence image stack.

[0038] Preferably, the optimized fluorescence image stack is specifically:

[0039]

[0040] Among them k ′(x) is the optimized fluorescence image stack, I(x) is the captured fluorescence image stack, and o k (x) is the fluorescence image stack after the kth iteration, ε TV is the regularization parameter, div() represents the divergence, Indicates o k The gradient of (x).

[0041] Compared with the prior art, the present invention has the following significant advantages:

[0042] (1) The present invention uniquely introduces adaptive optical assistance into the light intensity diffraction tomography module and derives a time-varying aberration update model for the imaging system, thereby being able to separate the coupled three-dimensional refractive index and aberration from the collected label-free intensity image, significantly improving the accuracy and reliability of imaging.

[0043] (2) The present invention uniquely feeds back the real-time calculated aberration correction results into the point spread function of the system in the fluorescence module, and combines it with the three-dimensional Richardson-Lucy algorithm to obtain three-dimensional fluorescence imaging results after adaptive optics-assisted correction, thereby significantly improving the reconstruction quality of the fluorescence modality and providing strong support for high-precision biological imaging.

[0044] (3) The present invention can effectively overcome focus drift caused by time-varying aberrations and mechanical errors without reducing imaging speed and resolution, thereby enhancing the imaging performance of long-term studies (such as living cell observation) and providing a more stable and efficient imaging tool for biomedical research.

[0045] (4) The present invention combines non-interference diffraction tomography with fluorescence imaging, which can not only fully display the interactions between organelles in living cells in three-dimensional space, but also realize molecular-specific visualization, thereby enabling comprehensive analysis of biological components or interactions that could only be studied individually in the past, providing a new perspective and method for biomedical research. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 This is a flow chart of the adaptive optics-assisted three-dimensional fluorescence and intensity diffraction tomography imaging method.

[0047] Figure 2 This is a schematic diagram of the adaptive optics-assisted three-dimensional fluorescence and light intensity diffraction tomography imaging system.

[0048] Figure 3 It is a schematic flow chart of the adaptive optics-assisted light intensity diffraction tomography module and the three-dimensional fluorescence module.

[0049] Figure 4 This is a long-term rendering comparison of the three-dimensional refractive index distribution of HeLa cells reconstructed using the adaptive optics-assisted before and after light intensity diffraction tomography method.

[0050] Figure 5 This is a comparison of three-dimensional fluorescence imaging of COS-7 cells before and after using adaptive optics. DETAILED DESCRIPTION

[0051] like Figure 1 As shown, an adaptive optics-assisted three-dimensional fluorescence and light intensity diffraction tomography imaging method includes the following five steps:

[0052] Step 1: Light up the LEDs in sequence to illuminate the sample from different angles to obtain intensity images;

[0053] like Figure 2 As shown, the present invention is based on an adaptive optics-assisted three-dimensional fluorescence and light intensity diffraction tomography imaging system, the adaptive optics-assisted three-dimensional fluorescence and light intensity diffraction tomography imaging system comprises a programmable LED (1), a microscope objective lens (3), a fluorescence excitation block (4), an imaging tube lens (5), a reflector (6) and a camera (7), wherein the programmable LED (1) is coaxial with the microscope objective lens (3), the fluorescence excitation block (4) and the imaging tube lens (5) and is placed at a predetermined height above the sample to be measured (2); the camera (7) is placed on the imaging focal plane;

[0054] The specific implementation process is as follows: when illuminating and imaging, the programmable LEDs (1) are lit one by one, and the quasi-monochromatic plane wave illuminates the sample from different angles according to the position of the LEDs, and then converges to the camera (7) plane after passing through the microscope objective (3), the fluorescence excitation block (4), the imaging tube lens (5), and the reflector (6), thereby recording a series of original light intensity images on the camera (7) plane.

[0055] Step 2: Use iterative stacking imaging technology to process the intensity image to obtain the three-dimensional spectrum after removing the aberration and the aperture function with aberration, such as Figure 3 As shown in (a), the process is as follows:

[0056] Step 2.1, initialize the three-dimensional spectrum and determine the three-dimensional spectrum information according to the incident wave vector and the generalized three-dimensional coherent transfer function;

[0057] Step 2.2: Project the three-dimensional spectrum along the axial frequency coordinate and transfer it to the center of the coordinate to obtain the two-dimensional spectrum. Step 2.3, perform aberration compensation on the two-dimensional spectrum to obtain the compensated two-dimensional spectrum (no compensation is required for the first iteration):

[0058]

[0059] Where P(k T )′ is the aperture function with aberration;

[0060] Step 2.4, calculate the first-order scattering potential based on the two-dimensional spectrum:

[0061]

[0062] Where j is the imaginary unit, k z is the axial wave vector;

[0063] Step 2.5: Combine the intensity image and impose an amplitude constraint on the first-order scattering potential to obtain the constrained scattering potential:

[0064]

[0065] Step 2.6, for the aperture function P(k T )′ to update:

[0066]

[0067] Where β represents the update step size, O is the intensity map I i (k T ) spectrum after Fourier transformation;

[0068] Step 2.7, calculate the two-dimensional spectrum after applying amplitude constraints:

[0069]

[0070] Step 2.8, perform aberration removal on the two-dimensional spectrum:

[0071]

[0072] After de-aberration, the two-dimensional spectrum is mapped to the three-dimensional Ewald shell to achieve synchronous update of the three-dimensional spectrum. After the update, it is determined whether the intensity images taken at all angles have been iterated. If so, proceed to the next step; otherwise, return to step 2.2.

[0073] Step 3: performing a three-dimensional inverse Fourier transform on the three-dimensional spectrum after removing the aberration to obtain a three-dimensional refractive index distribution of the reconstructed sample, thereby achieving label-free three-dimensional imaging of the measured sample;

[0074] Step 4, the sample is excited for fluorescence, the microscope objective lens (3) is moved in the z-axis and the camera (7) is synchronously triggered to take a picture, and a fluorescence image stack is obtained;

[0075] like Figure 2As shown, the fluorescence imaging module is based on the fluorescence excitation block (4) of the IX83 microscope, and cooperates with the motorized z-axis scanning function of the microscope objective (3) to realize the acquisition of fluorescence three-dimensional stacks; the experimental system is programmed in C++, combined with the SDK development kit of the camera and microscope and the serial port communication of the FPGA controller to realize the control of each component of the imaging system and complete the automatic acquisition of dual-modal images;

[0076] The specific implementation process is as follows: the sample is excited for fluorescence, the microscope objective lens (3) moves along the z-axis and is synchronously triggered to take pictures in conjunction with the camera (7), the fluorescence passes through the fluorescence excitation block (4), is reflected through the microscope objective lens (3), and then reaches the sample to be tested (2), and then is reflected by the sample to be tested (2), passes through the microscope objective lens (3), the fluorescence excitation block (4), the imaging tube lens (5), and the reflector (6), and then converges on the camera plane, thereby recording a fluorescence image stack on the camera (7) plane.

[0077] Step 5: The aberration obtained after the iteration of step 2 is introduced into the point spread function of the imaging system, and the three-dimensional Richardson-Lucy deconvolution algorithm is used to correct the three-dimensional fluorescence results, as shown in FIG. Figure 3 As shown in (b), the process is as follows:

[0078] Step 5.1: Compare the optical transfer function of the imaging system with the aperture function P(k T )′, the optical transfer function with aberration is obtained:

[0079] H′(u)=H(u)P(k T )′

[0080] Where H(u) is the optical transfer function of the imaging system;

[0081] Step 5.2, performing a three-dimensional inverse Fourier transform on the optical transfer function H′(u) with aberration to obtain the point spread function h′(u);

[0082] In step 5.3, a three-dimensional Richardson-Lucy deconvolution operation is performed on the fluorescence image stack in combination with the point spread function h′(u) to obtain the optimized fluorescence image stack:

[0083]

[0084] Among them k ′(x) is the optimized fluorescence image stack, I(x) is the captured fluorescence image stack, and o k (x) is the fluorescence image stack after the kth iteration, ε TV is the regularization parameter, div() represents the divergence, Indicates o k The gradient of (x).

[0085] Figure 4 This is a long-term rendering comparison of the three-dimensional refractive index distribution of HeLa cells reconstructed using the adaptive optics-assisted light intensity diffraction tomography method before and after. In the reconstruction results of the traditional method without the assistance of adaptive optics, the time-varying defocus aberration causes focus drift, and the cell information represented on the same z-plane has significant differences at different time points. The method of the present invention has a significantly better reconstruction effect, effectively removes the influence of aberrations, and can obtain clear cell structures and cell contours in long-term imaging, verifying the stability and high-speed and high-resolution performance of the method of the present invention in light intensity diffraction tomography imaging.

[0086] Figure 5 This is a comparison of three-dimensional fluorescence imaging of COS-7 cells before and after the use of adaptive optics assistance. Without the use of adaptive optics assistance, the image quality is significantly reduced and accompanied by focus drift. However, after the introduction of adaptive optics assistance, the quality of the reconstruction results is significantly improved. Both the healthy tubular mitochondrial structure and the spherical mitochondrial changes caused by phototoxicity are clearly visible under the assistance of adaptive optics, verifying the accuracy and high resolution of the method of the present invention.

[0087] This paper proposes a novel adaptive optics-assisted three-dimensional fluorescence and intensity diffraction tomography imaging method and constructs a corresponding experimental setup. The method combines adaptive optics-assisted aberration correction with the intensity diffraction tomography modality and uniquely employs an iterative stacking imaging approach to effectively separate the coupled three-dimensional refractive index and aberration from the captured label-free intensity image. Simultaneously, the method feeds back the real-time calculated aberration correction results into the system's point spread function to synchronously correct the three-dimensional fluorescence imaging results, significantly improving the reconstruction quality of the fluorescence modality. The method effectively overcomes focus drift caused by time-varying aberrations and mechanical errors without compromising imaging speed and resolution, thereby enhancing imaging performance for long-term studies such as live cell observation. Furthermore, non-interferometric diffraction tomography enables comprehensive visualization of three-dimensional interactions between organelles within living cells without the need for additional labeling or complex experimental setup. Combined with fluorescence imaging modality, the standard for molecular-specific visualization, the method stands out as an effective imaging solution, enabling the integrated and synergistic analysis of biological components and interactions previously studied only in isolation, and capable of imaging the morphology and properties of subcellular organelles such as mitochondria with high spatiotemporal resolution. This invention applies adaptive optics technology to three-dimensional dual-modal imaging for the first time, achieving long-term, high-temporal and high-spatial resolution three-dimensional dual-modal cell imaging results.

Claims

1. An adaptive optics-assisted three-dimensional fluorescence and light intensity diffraction tomography imaging method, characterized in that: Here are the steps: Step 1: Illuminate the sample from different angles and collect intensity images; Step 2: Use iterative stacking imaging technology to process the intensity image to obtain the three-dimensional spectrum after removing the aberration and the aperture function with aberration; Step 3: Perform a 3D inverse Fourier transform on the 3D spectrum after removing the aberration to obtain the 3D refractive index distribution of the reconstructed sample, thus achieving label-free 3D imaging of the measured sample. Step 4: Excite the sample for fluorescence and collect a fluorescence image stack; Step 5: Introduce the aberration into the point spread function of the imaging system and apply the 3D Richardson-Lucy deconvolution algorithm to correct the 3D fluorescence results.

2. The adaptive optics-assisted three-dimensional fluorescence and light intensity diffraction tomography imaging method according to claim 1, characterized in that: An adaptive optics-assisted three-dimensional fluorescence and light intensity diffraction tomography imaging system is used to collect intensity images. The adaptive optics-assisted three-dimensional fluorescence and light intensity diffraction tomography imaging system comprises a programmable LED (1), a microscope objective lens (3), a fluorescence excitation block (4), an imaging tube lens (5), a reflector (6), and a camera (7). The programmable LED (1) is coaxial with the microscope objective lens (3), the fluorescence excitation block (4), and the imaging tube lens (5) and is placed at a predetermined height above a sample to be measured (2); the camera (7) is placed on an imaging focal plane; during illumination imaging, the programmable LEDs (1) are lit one by one, and quasi-monochromatic plane waves illuminate the sample from different angles according to the positions of the LEDs, and converge onto the camera (7) plane after passing through the microscope objective lens (3), the fluorescence excitation block (4), the imaging tube lens (5), and the reflector (6), thereby recording a series of original light intensity images on the camera (7) plane.

3. The adaptive optics-assisted three-dimensional fluorescence and light intensity diffraction tomography imaging method according to claim 1, characterized in that: In step 2, the intensity image is processed using iterative stacking technology to obtain the three-dimensional spectrum after removing the aberration and the aperture function with the aberration. The specific steps are as follows: Step 2.1, initialize the three-dimensional spectrum and determine the three-dimensional spectrum information according to the incident wave vector and the generalized three-dimensional coherent transfer function; Step 2.2: Project the three-dimensional spectrum along the axial frequency coordinate and transfer it to the center of the coordinate to obtain the two-dimensional spectrum. Step 2.3, perform aberration compensation on the two-dimensional spectrum to obtain the compensated two-dimensional spectrum Where P(k T )′ is the aperture function with aberration; Step 2.4, calculate the first-order scattering potential based on the compensated two-dimensional spectrum: Where j is the imaginary unit, k z is the axial wave vector; Step 2.5: Combine the intensity image and impose an amplitude constraint on the first-order scattering potential to obtain the constrained scattering potential: Step 2.6, for the aperture function P(k T )′ to update: Where β represents the update step size, O is the intensity map I i (k T ) spectrum after Fourier transformation; Step 2.7, calculate the two-dimensional spectrum after applying amplitude constraints: Step 2.8, perform aberration removal on the two-dimensional spectrum: After de-aberration, the two-dimensional spectrum is mapped to the three-dimensional Ewald shell to achieve synchronous update of the three-dimensional spectrum. After the update, it is determined whether the intensity images taken at all angles have been iterated. If so, proceed to the next step; otherwise, return to step 2.

2.

4. The adaptive optics-assisted three-dimensional fluorescence and light intensity diffraction tomography imaging method according to claim 2, characterized in that: The specific method for fluorescence excitation of the sample and acquisition of fluorescence image stack is as follows: The sample is excited for fluorescence, and the microscope objective lens (3) moves along the z-axis and is synchronously triggered to take pictures in conjunction with the camera (7). The fluorescence passes through the fluorescence excitation block (4) and is reflected through the microscope objective lens (3) to the sample to be tested (2). The fluorescence is then reflected by the sample to be tested (2) and passes through the microscope objective lens (3), the fluorescence excitation block (4), the imaging tube lens (5), and the reflector (6) to converge on the camera plane, thereby recording a fluorescence image stack on the camera (7) plane.

5. The adaptive optics-assisted three-dimensional fluorescence and light intensity diffraction tomography imaging method according to claim 3, characterized in that: Step 5 introduces the aberration into the point spread function of the imaging system and uses the 3D Richardson-Lucy deconvolution algorithm to correct the 3D fluorescence results. The specific steps are as follows: Step 5.1: Compare the optical transfer function of the imaging system with the aperture function P(k T )′, the optical transfer function with aberration is obtained: H′(u)=H(u)P(k T )′ Where H(u) is the optical transfer function of the imaging system; Step 5.2, performing a three-dimensional inverse Fourier transform on the optical transfer function H′(u) with aberration to obtain the point spread function h′(u); In step 5.3, a three-dimensional Richardson-Lucy deconvolution operation is performed on the fluorescence image stack in combination with the point spread function h′(u) to obtain an optimized fluorescence image stack.

6. The adaptive optics-assisted three-dimensional fluorescence and light intensity diffraction tomography imaging method according to claim 1, characterized in that: The optimized fluorescence image stack is as follows: Among them k ′(x) is the optimized fluorescence image stack, I(x) is the captured fluorescence image stack, and o k (x) is the fluorescence image stack after the kth iteration, ε TV is the regularization parameter, div() represents the divergence, Indicates o k The gradient of (x).

Images