An extended depth-of-focus microscopic imaging system incorporating optical tweezers

Abstract

This invention relates to the field of microscopic observation technology, specifically to an extended depth-of-focus microscopic imaging system incorporating optical tweezers functionality. The system includes a beam generation and combining module, a signal acquisition module, a signal separation module, an image reconstruction module, and an optical tweezers analysis module. The system generates a phase-modulated extended depth-of-focus illumination beam and an optical tweezers manipulation beam, which are then spatially coaxially combined to form a combined beam, which is then guided into the microscope objective to act on the sample region. The acquired optical signal is divided into independent detection channels through beam splitting processing. The imaging channel, combined with a depth-of-field extension point spread function model, optimizes the fusion algorithm to reconstruct the three-dimensional structure of the sample. The optical tweezers monitoring channel analyzes the optical path signal, calculates the three-dimensional position of particles, trapping stiffness, and optical force, and visualizes the three-dimensional relative relationship between these two parameters by combining the imaging results. This scheme achieves integrated design of the imaging optical path and the optical tweezers optical path, optimizes image reconstruction quality under large depth of field, and enables synchronous and collaborative operation of microscopic imaging and optical manipulation.

Description

Technical Field

[0001] This invention relates to the field of microscopic observation technology, and in particular to an extended depth-of-focus microscopic imaging system that incorporates optical tweezers. Background Technology

[0002] Currently, microscopic imaging and optical tweezers manipulation equipment generally adopts a separate optical path structure. Extended depth-of-focus microscopic imaging and optical particle capture functions are equipped with dedicated optical paths, and the imaging system and optical tweezers control system are completely independent operating units. Conventional multifocal image fusion algorithms are uniformly based on the standard point spread function model to build the computational framework. The imaging optical path and the optical tweezers monitoring optical path lack integrated collaborative design. The two types of optical signals adopt independent acquisition and transmission modes throughout the process, and there is no unified beam separation and channel detection processing structure.

[0003] A separate optical layout leads to redundancy in the overall system structure, complex optical component matching, and cumbersome optical path calibration and debugging procedures, making it impossible to simultaneously perform sample microscopic imaging and particle optical tweezers capture operations within the same observation field of view. Fixed native depth-of-focus parameters compress the longitudinal coverage of microscopic observations, making it difficult to fully acquire continuous optical information about the multi-layered three-dimensional structure of the sample. The standard point spread function model cannot adapt to the optical transmission laws under the special conditions of extended depth of focus, resulting in insufficient reconstruction of the sample's three-dimensional structure after image fusion processing. The detection data from different optical paths are fragmented and isolated, making it difficult to simultaneously acquire particle motion force parameters and sample microstructure information, and hindering spatial correlation analysis of captured particles and sample structure. Summary of the Invention

[0004] The purpose of this invention is to overcome the shortcomings of existing technologies and propose an extended depth-of-focus microscopy imaging system that combines optical tweezers functionality.

[0005] To achieve the above objectives, the present invention employs the following technical solution: an extended depth-of-focus microscopic imaging system incorporating optical tweezers, comprising: The beam generation and combining module generates an extended depth-of-focus illumination beam with phase modulation and an optical tweezers manipulation beam with a specific spatial shape. The extended depth-of-focus illumination beam and the optical tweezers manipulation beam are spatially coaxially combined to form a combined beam. The signal acquisition module guides the combined light beam into the microscope objective and focuses it on the sample area. And collect the light signals that pass through or are scattered by the sample; The signal separation module performs beam splitting and separation processing on the collected optical signals, separating the imaging optical path signal and the optical tweezers monitoring optical path signal into different detection channels; The image reconstruction module processes the imaging optical path signal in the imaging detection channel using an improved multifocal image fusion algorithm. The improved multifocal image fusion algorithm is optimized based on the point spread function model after depth expansion to reconstruct the three-dimensional structural image of the sample. The optical tweezers analysis module analyzes the optical path signal of the optical tweezers monitoring channel, calculates the three-dimensional position, capture stiffness and optical force of the captured particles, and visualizes the relative relationship between the particles and the sample structure in three-dimensional space based on the reconstructed three-dimensional structural image of the sample and the calculated three-dimensional position of the captured particles.

[0006] As a further aspect of the present invention, generating a phase-modulated extended depth-of-focus illumination beam and a tweezers-manipulated beam with a specific spatial morphology, and spatially coaxially combining the extended depth-of-focus illumination beam and the tweezers-manipulated beam to form a combined beam, includes: The original beam emitted from the light source passes through a polarization control device and forms two linearly polarized beams with orthogonal polarization states. One of the linearly polarized beams is incident on a first spatial light modulator, which is loaded with a computational hologram. The computational hologram is optimized to generate a wavefront with a ring intensity distribution or a Bessel beam shape, thereby generating the optical tweezers manipulated beam. Another beam of linearly polarized light is incident on a second spatial light modulator, which is loaded with a phase modulation pattern. The phase modulation pattern is used to encode the incident wavefront to generate an illumination beam with an extended focal range along the optical axis, i.e., the extended depth-of-focus illumination beam. The optical tweezers-manipulated beam modulated by the first spatial light modulator and the extended depth-of-focus illumination beam modulated by the second spatial light modulator are combined through a polarizing beam splitter, so that the two beams propagate along the same optical path to form the polarization-multiplexed combined beam. The combined beam is expanded and collimated to fill the back aperture of the microscope objective.

[0007] As a further aspect of the present invention, the combined light beam is guided into the microscope objective, focused on the region where the sample is located, and the light signal passing through or scattered by the sample is collected, including: The extended depth-of-focus illumination beam forms an extended uniform illumination focal zone within the sample for large depth-of-field imaging of the sample. At the same time, the optical tweezers manipulate the beam to form a three-dimensional optical potential trap within the illumination focal zone for capturing and manipulating micro- and nano-particles in the sample. The optical signal includes the imaging signal generated by the extended depth-of-focus illumination beam and the scattering signal generated by the interaction between the optical tweezers-manipulated beam and the particles. The extended depth-of-focus illumination beam forms an extended, uniform illumination focal zone within the sample for large depth-of-field imaging. Simultaneously, the optical tweezers manipulate the beam to form a three-dimensional optical potential trap within the illumination focal zone, used to capture and manipulate micro / nano particles in the sample. Specifically, this includes: After the combined beam is focused by the microscope objective, the phase modulation wavefront of the extended depth-of-focus illumination beam generates self-reconstruction characteristics along the optical axis near the focal plane, thereby forming an illumination area in the sample with a size larger than that of traditional point focusing along the optical axis. The illumination area is the extended uniform illumination focal zone. The intensity distribution of the illumination focal area is uniform along the axial direction, so that sample structures at different heights within the axial range of the illumination focal area can obtain consistent illumination intensity, thereby achieving the large depth-of-field imaging. Meanwhile, the optical tweezers manipulate the beam to form a series of intensity maxima on the beam propagation path due to its annular intensity distribution or Bessel beam shape, and these intensity maxima constitute the three-dimensional optical potential trap. The three-dimensional optical potential trap overlaps spatially with the extended uniform illumination focal zone formed by the extended depth-of-focus illumination beam in both axial position and lateral range, such that the captured micro-nano particles are located within the imaging illumination region. By dynamically controlling the computational hologram loaded on the first spatial light modulator, the wavefront of the optical tweezers manipulated beam is changed, thereby moving the spatial position of the three-dimensional optical potential trap within the illumination focal zone, thus realizing the three-dimensional capture and manipulation of the micro-nano particles.

[0008] As a further aspect of the present invention, the step of performing beam splitting and separation processing on the collected optical signals to separate the imaging optical path signals and the optical tweezers monitoring optical path signals into different detection channels specifically includes: The collected optical signal is guided to a beam splitter, which splits the optical signal into different paths according to the wavelength, polarization state, or light intensity. In the imaging optical path direction, a bandpass filter is provided to selectively transmit light corresponding to the wavelength of the extended depth-of-focus illumination beam, while blocking light of the wavelength of the optical tweezers-controlled beam and other background stray light, thereby forming the imaging optical path signal; In the direction of the optical tweezers monitoring optical path, a four-quadrant detector or a position-sensitive detector is set to receive weak signals related to the optical tweezers control beam that are scattered or refracted by particles in the sample, thus forming the optical tweezers monitoring optical path signal. The imaging optical path signal is imaged onto the photosensitive surface of a scientific array camera using a lens group. The optical tweezers monitor the optical path signal and use a focusing lens to converge it to the center of the sensitive area of ​​the four-quadrant detector or the position-sensitive detector.

[0009] As a further aspect of the present invention, the improved multifocal image fusion algorithm is optimized based on the point spread function model after depth extension, and its working principle includes: The three-dimensional point spread function of the system is pre-calibrated or calculated under the illumination condition of the extended depth-of-field illumination beam, and the point spread function model after the depth of field extension is obtained. During the imaging process, the sample is illuminated by the extended depth-of-focus illumination beam, and the camera acquires a series of two-dimensional images focused at different axial positions, forming an original image stack. For each two-dimensional image in the original image stack, the light intensity distribution of the sample layer corresponding to the two-dimensional image in three-dimensional space is estimated by using the point spread function model after the depth of field expansion and through three-dimensional deconvolution operation. From the preliminary three-dimensional light intensity distribution obtained by three-dimensional deconvolution operation, image information of each axial layer is extracted, and the optimal focus position of each image region is adaptively determined according to the information redundancy and signal-to-noise ratio of adjacent layers. The pixel values ​​of each image layer at its optimal focus position are weighted and fused based on the local image sharpness evaluation function to reconstruct a clear two-dimensional image of the entire sample area. By axial scanning or utilizing the depth-of-field information of the extended depth-of-focus illumination beam, clear two-dimensional images reconstructed from different focus positions are stitched together in three dimensions to ultimately generate a three-dimensional structural image of the sample.

[0010] As a further aspect of the present invention, the step of estimating the light intensity distribution of the sample layer corresponding to the two-dimensional image in three-dimensional space for each two-dimensional image in the original image stack, using the point spread function model after depth expansion, through three-dimensional deconvolution operation, includes: The single two-dimensional image in the original image stack is regarded as a two-dimensional projection of the three-dimensional sample structure after being convolved by the three-dimensional point spread function. An imaging model is established, which correlates the three-dimensional sample structure, the point spread function model after depth expansion, and the acquired two-dimensional image through a convolution equation. The convolution equation is solved using an iterative deconvolution algorithm, which updates the estimate of the three-dimensional sample structure in each iteration. During the iteration process, a total variational regularization constraint is introduced to suppress non-physical oscillations or instabilities caused by noise and the deconvolution process itself, so as to obtain a smoother and more physically reliable three-dimensional light intensity distribution estimate. When the iteration results converge or the preset number of iterations is reached, the iteration stops. The three-dimensional light intensity distribution corresponding to the two-dimensional image at this time is the estimated light intensity distribution of the sample layer in three-dimensional space.

[0011] As a further aspect of the present invention, the step of analyzing the optical tweezers monitoring optical path signal in the optical tweezers monitoring channel and calculating the three-dimensional position, capture stiffness, and optical force experienced by the captured particle includes: The voltage signal related to the particle displacement is read from the four-quadrant detector or the position-sensitive detector, and the change in the voltage signal reflects the amount of deviation of the particle in two perpendicular directions in the detector plane; Based on the pre-calibrated proportional relationship between the voltage signal and the actual displacement of the particle, the voltage signal is converted into two-dimensional position coordinates of the particle in the horizontal plane; By analyzing the potential well gradient characteristics formed by the optical tweezers manipulating the beam along the axial direction, and utilizing the correlation between the detector signal strength and the particle's axial position, the axial position of the particle in the optical axis direction is calculated, and the three-dimensional position of the captured particle is obtained by combining the transverse two-dimensional position coordinates. The Brownian motion trajectory of the particle in the optical trap due to thermal motion is recorded without external disturbance. Power spectrum analysis is performed on the Brownian motion trajectory to obtain the Lorentz curve. The trapping stiffness is calculated from the fitting parameters. Based on the displacement of the three-dimensional position of the captured particle relative to the center of the optical trap, and the trapping stiffness, the components of the optical force on the captured particle in the three directions are calculated according to Hooke's law.

[0012] As a further aspect of the present invention, by analyzing the potential well gradient characteristics formed along the axial direction by the optical tweezers-manipulated beam, and utilizing the correlation between the detector signal intensity and the particle's axial position, the axial position of the particle in the optical axis direction is calculated, including: During the initialization phase of the optical tweezers system, a calibration microsphere of known size is scanned axially, and the change curves of the total light intensity signal or differential signal output by the four-quadrant detector or position-sensitive detector are recorded when the calibration microsphere is in different axial positions, thus establishing an axial position-signal intensity lookup table. During the actual capture and measurement phase, the total light intensity signal or differential signal output by the four-quadrant detector or position-sensitive detector is monitored in real time. The real-time monitored signal strength value is compared with the axial position-signal strength lookup table to find the axial position corresponding to the signal strength value that best matches it. Interpolation is performed on the search results to improve the resolution of axial position positioning; Based on the calculation results of the particle's lateral position, the complete three-dimensional spatial coordinates of the particle are output.

[0013] As a further aspect of the present invention, the visualization of the relative relationship between the particles and the sample structure in three-dimensional space based on the reconstructed three-dimensional structural image of the sample and the calculated three-dimensional position of the captured particles includes: Establish a unified three-dimensional spatial coordinate system and transform the voxel coordinates of the three-dimensional structural image of the sample and the three-dimensional position coordinates of the captured particles to the unified coordinate system; In a 3D rendering environment, the 3D structural image of the sample is loaded, and the 3D morphology of the sample is displayed in a volume rendering or surface rendering manner. In the 3D rendering environment, the real-time 3D position of the captured particles is dynamically marked with highlighted markers of customizable shape and color. Based on user interaction commands, the viewing angle and scaling ratio of the 3D rendering view can be dynamically adjusted, and the cross-sectional display function can be enabled to observe the spatial relationship between particles and the internal structure of the sample. The three-dimensional position of the captured particles is updated in real time, and the position of the marker points is updated synchronously in the three-dimensional rendering environment to form a dynamic visualization of the particles moving within the three-dimensional structure of the sample.

[0014] As a further aspect of the present invention, the establishment of a unified three-dimensional spatial coordinate system includes: The optical axis of the microscope objective is taken as the Z-axis of the unified three-dimensional spatial coordinate system, and the two directions perpendicular to the optical axis and orthogonal to each other are taken as the X-axis and Y-axis. A calibration sample with known spatial characteristics is placed on the sample stage, and an image of the calibration sample is obtained through three-dimensional imaging. Identify the three-dimensional image coordinates of known feature points in the calibration sample image and establish a correspondence between them and the actual physical coordinates of the feature points on the sample stage; By solving the coordinate transformation parameters, the transformation relationship between image voxel coordinates and physical coordinates in a unified three-dimensional spatial coordinate system is established. The three-dimensional position of the particle calculated by the optical tweezers monitoring channel is mapped to the unified three-dimensional spatial coordinate system through a pre-calibrated transformation relationship to ensure the spatial consistency between the imaging coordinates and the manipulation coordinates.

[0015] Compared with the prior art, the advantages and positive effects of the present invention are as follows: Two beams are spatially coaxially combined to form a unified combined beam, which relies on a single microscope objective to complete optical path transmission and sample focusing. This unified optical path arrangement simplifies the configuration of internal optical system components, streamlines the overall optical path layout, and avoids structural redundancy caused by parallel deployment of multiple independent optical paths. The imaging beam and the optical tweezers control beam maintain a coaxial transmission path, allowing imaging and optical tweezers control operations to be performed simultaneously in the same observation area. This eliminates the problem of inconsistent observation points under a split optical path architecture, unifies the optical focusing path, reduces optical path offset caused by independent operation of multiple optical paths, and allows a single observation area of ​​the sample to simultaneously perform both imaging detection and optical capture functions, achieving integrated operation of optical imaging and optical tweezers control.

[0016] Based on the point spread function model after depth-of-field extension, a targeted optimization of the multifocal image fusion algorithm was completed. The algorithm's operational logic was adjusted to adapt to the optical transmission characteristics of extended depth-of-field imaging, overcoming the limitation of conventional algorithm models in adapting to a single scenario. The fusion processing logic of the imaging optical path signal was optimized to improve the reconstruction effect of the sample's three-dimensional structure. Quantitative analysis was performed on the optical path signal of the optical tweezers monitoring channel to calculate and obtain the three-dimensional position, capture stiffness, and optical force-related parameters of the captured particles. Combining the reconstructed three-dimensional structural information of the sample with particle spatial coordinate data, a visualization of the relative state between particles and sample structure within three-dimensional space was achieved. This established a data link between imaging detection and optical tweezers monitoring, enabling synchronous access and fusion processing of the two types of detection data, and improving the parameter analysis and spatial characterization capabilities during microscopic observation. Attached Figure Description

[0017] Figure 1 This is a timing diagram of an extended depth-of-focus microscopic imaging system that incorporates optical tweezers, as described in this invention. Figure 2 A flowchart for beam generation and combining; Figure 3 This is a flowchart of optical signal splitting and separation processing. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0019] In the description of this invention, it should be understood that the terms "length," "width," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," and "outer," etc., indicating orientation or positional relationships, are based on the orientation or positional relationships shown in the accompanying drawings and are only for the convenience of describing the invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the invention. Furthermore, in the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0020] See Figure 1 This invention provides an extended depth-of-focus microscopy imaging system that combines optical tweezers functionality. The specific system includes: The beam generation and combining module generates a phase-modulated extended depth-of-focus illumination beam and a tweezers manipulation beam with a specific spatial morphology, and spatially coaxially combines the two beams to form a combined beam. The signal acquisition module guides this combined beam into the microscope objective, focusing it on the sample area and collecting light signals passing through or scattered by the sample. The signal separation module performs beam splitting and separation processing on the collected light signals, separating the imaging optical path signal and the optical tweezers monitoring optical path signal into different detection channels. In the imaging detection channel, the image reconstruction module uses an improved multifocal image fusion algorithm to process the imaging optical path signal. This improved multifocal image fusion algorithm is optimized based on the point spread function model after depth-of-field extension to reconstruct a three-dimensional structural image of the sample. In the optical tweezers monitoring channel, the optical tweezers analysis module analyzes the optical tweezers monitoring optical path signal, calculates the three-dimensional position, capture stiffness, and optical force of the captured particles, and visualizes the relative relationship between the particles and the sample structure in three-dimensional space based on the reconstructed three-dimensional structural image of the sample and the calculated three-dimensional position of the particles.

[0021] In one embodiment of the present invention, the method for generating a combined beam using the beam generation and beam combining module is as follows. (See reference...) Figure 2 The original beam emitted from the laser source passes through a polarization control device, forming two linearly polarized beams with orthogonal polarization states. One of these beams is incident on a first spatial light modulator loaded with a computational hologram optimized to generate a wavefront with a ring intensity distribution or a Bessel beam shape, thus creating an optically tweezer-manipulated beam. The other beam is incident on a second spatial light modulator loaded with a phase modulation pattern used to encode the incident wavefront, generating an illumination beam with an extended focal length along the optical axis, i.e., an extended depth-of-focus illumination beam. The optically tweezer-manipulated beam modulated by the first spatial light modulator and the extended depth-of-focus illumination beam modulated by the second spatial light modulator are guided to a polarization beam splitter for combination, allowing the two orthogonally polarized beams to propagate along the same optical path, forming a polarization-multiplexed combined beam. This combined beam then passes through a beam expander and collimator system to adapt its spot size and fill the back aperture of the microscope objective.

[0022] In a specific implementation, the physical realization of the beam generation and combining module begins with a continuous laser with an output wavelength of 1064 nm. The original Gaussian beam emitted from the laser source is separated into two linearly polarized beams with orthogonal polarization states by a polarization control device consisting of a half-wave plate and a polarizing beam splitter. One of the linearly polarized beams, with a horizontal polarization direction, is guided to a reflective pure-phase first spatial light modulator. The first spatial light modulator is loaded with an optimized computational hologram. The phase distribution function of the computational hologram is generated by an iterative Fourier transform algorithm. This function modulates the incident wavefront to produce a wavefront with a ring intensity distribution or a zero-order Bessel beam shape, thereby generating a beam that can be manipulated by optical tweezers. Another beam of linearly polarized light with a perpendicular polarization direction is incident on another reflective pure phase second spatial light modulator of the same type. The second spatial light modulator is loaded with a phase modulation pattern, which can be a spiral phase plate pattern or an axial polarization conversion pattern, to encode the incident wavefront in order to generate an illumination beam with an extended focal range along the optical axis, i.e. an extended depth-of-focus illumination beam.

[0023] In some embodiments, the optimization process for computing the hologram takes into account the discretization effect of the pixels of the first spatial light modulator and the coherence length of the laser source. By introducing an additional aberration compensation term to correct the wavefront error, the generated optical tweezers-manipulated beam has a more ideal annular intensity distribution or Bessel beam shape in the focal region. The optical tweezers-manipulated beam modulated by the first spatial light modulator and the extended depth-of-focus illumination beam modulated by the second spatial light modulator are precisely adjusted in their optical paths so that they are incident at a specific angle onto a polarizing beam splitter. The transmission axis and reflection axis of the polarizing beam splitter correspond to the horizontal and vertical polarization directions, respectively. This allows the two orthogonally polarized optical tweezers-manipulated beams and the extended depth-of-focus illumination beam to be spatially superimposed, so that the two beams propagate along the same physical optical path, forming a polarization-multiplexed combined beam.

[0024] Optionally, a quarter-wave plate can be inserted in the optical path after the polarizing beam splitter to convert the polarization state of the combined beam into circularly polarized light, thereby eliminating the polarization dependence that may exist in some samples. The combined beam is then guided into a beam expander and collimator system consisting of two lenses: the first lens is a short focal length negative lens, and the second lens is a long focal length positive lens. After passing through the beam expander and collimator system, the beam diameter is magnified and collimated, ultimately filling the back aperture of an oil immersion microscope objective with a numerical aperture of 1.2. It can be understood that the magnification of the beam expander and collimator system is determined by the focal length ratio of the two lenses. Specifically, the calculation relationship for the magnification M is as follows:

[0025] Where f1 represents the focal length of the first negative lens and f2 represents the focal length of the second positive lens. By precisely selecting the focal lengths of the lenses, it is possible to ensure that the size of the combined beam is precisely matched with the size of the back aperture of the microscope objective, thereby maximizing the utilization of the numerical aperture of the microscope objective and achieving a high-quality focused light field. In some embodiments, to further optimize the wavefront, a deformable mirror can be introduced into the beam expander path to dynamically correct static and dynamic aberrations introduced by the spatial light modulator, optical elements, and optical path assembly.

[0026] In one embodiment of the present invention, the implementation of the combined beam acting on the sample and subsequent signal acquisition and separation is as follows. After being focused by the microscope objective, the phase modulation wavefront of the extended depth-of-focus illumination beam generates self-reconstruction characteristics along the optical axis near the focal plane, thereby forming an illumination region within the sample with a size larger than that of conventional point focusing along the optical axis. This illumination region is the extended uniform illumination focal zone, and the intensity distribution of this illumination focal zone is uniform in the axial direction, so that sample structures at different heights within its axial range can obtain consistent illumination intensity, thereby achieving large depth-of-field imaging. The optical tweezers manipulate the beam to form a series of intensity maxima along the beam propagation path due to its annular intensity distribution or Bessel beam shape. These intensity maxima constitute a three-dimensional optical potential trap. This three-dimensional optical potential trap overlaps spatially with the extended uniform illumination focal zone formed by the extended depth-of-focus illumination beam in both axial position and lateral range, so that the captured micro- and nano-particles are located within the imaging illumination region. By dynamically controlling the computational hologram loaded onto the first spatial light modulator, the wavefront of the optical tweezers-manipulated beam can be altered, thereby shifting the spatial position of the three-dimensional optical potential trap within the illumination focal zone, achieving three-dimensional capture and manipulation of micro / nano particles. (See also...) Figure 3 The collected optical signal is guided to a beam splitter, which splits the signal into different paths based on wavelength, polarization state, or intensity. A bandpass filter is placed in the imaging optical path direction to selectively transmit light corresponding to the wavelength of the extended depth-of-focus illumination beam, while blocking light of the optical tweezers-manipulated beam wavelength and other background stray light, forming the imaging optical path signal. In the optical tweezers monitoring optical path direction, a quadrant detector or a position-sensitive detector is placed to receive weak signals related to the optical tweezers-manipulated beam, scattered or refracted by particles in the sample, forming the optical tweezers monitoring optical path signal. The imaging optical path signal is imaged onto the photosensitive surface of the area array scientific camera via lenses, while the optical tweezers monitoring optical path signal is converged at the center of the sensitive area of ​​the quadrant detector or position-sensitive detector via a focusing lens.

[0027] In practice, the combined beam, after passing through a 100x oil immersion microscope objective with a numerical aperture of 1.3, is precisely focused onto the sample area within the sample cell. The phase modulation wavefront of the extended depth-of-focus illumination beam generates self-reconstruction characteristics along the optical axis near the focal plane, thus forming a uniform illumination region extending along the optical axis with a length of approximately 6 micrometers within the sample. This region is the extended uniform illumination focal zone, and the axial intensity distribution uniformity deviation of the illumination focal zone is less than 10%. This ensures that different sample structures within the axial range of the illumination focal zone with height differences within 6 micrometers can obtain consistent illumination intensity, thereby achieving large depth-of-field imaging. The optical tweezers manipulate the beam, generating a ring-shaped intensity distribution due to the computational hologram it is loaded with, forming a series of intensity maxima spaced approximately 1.5 micrometers apart along the beam propagation path. These intensity maxima constitute an axially extended three-dimensional optical potential trap. The three-dimensional optical potential trap completely overlaps with the extended uniform illumination focal zone formed by the extended depth-of-focus illumination beam in both axial position and lateral range, ensuring that the captured micro / nano particles remain within the imaging illumination area. By dynamically controlling the computational hologram loaded on the first spatial light modulator, such as linearly increasing or decreasing the center offset of the phase map within the range of 0-2π, the wavefront tilt angle of the optical tweezers-manipulated beam can be changed, thereby moving the spatial position of the three-dimensional optical potential trap within the illumination focal zone. A movement range of ±50 micrometers can be achieved laterally and ±20 micrometers can be achieved axially, enabling the three-dimensional capture and manipulation of micro- and nano-particles.

[0028] In some embodiments, the light signal passing through the sample and the light signal scattered by particles in the sample are collected by the same microscope objective to form a backscattered light path. This backscattered light is separated from the illumination light path by a dichroic mirror. The collected light signal is guided to a beam splitter consisting of a polarizing beam splitter prism and a bandpass filter. The beam splitter splits the light signal into different paths according to different wavelengths. In the imaging light path direction, a bandpass filter with a center wavelength of 532 nm and a bandwidth of 10 nm is set to selectively transmit light corresponding to the wavelength of the extended depth-of-focus illumination beam, completely blocking light of the 1064 nm optical tweezers manipulation beam wavelength and background stray light with wavelengths below 500 nm and above 600 nm, forming a pure imaging light path signal. In the optical tweezers monitoring light path direction, a narrowband filter with a center wavelength around 1064 nm is set and connected to a four-quadrant detector to receive the weak backscattered signal associated with the optical tweezers manipulation beam scattered or refracted by particles in the sample, forming the optical tweezers monitoring light path signal.

[0029] Optionally, the imaging optical signal passes through a lens group consisting of a tube lens and a relay lens, and is precisely imaged onto the photosensitive surface of a scientific-grade complementary metal-oxide-semiconductor (CMOS) area array camera. The camera's pixel size is 6.5 micrometers, and after matching the magnification of the objective lens, the system's lateral imaging resolution is 130 nanometers. The optical tweezers monitoring optical signal passes through a focusing lens and is converged at the center of the sensitive region of the four-quadrant detector. The diameter of the sensitive region of the four-quadrant detector is 1 millimeter, and the diameter of the focused spot is adjusted to approximately 0.5 millimeters to ensure that the signal falls entirely within the sensitive region. It is understood that the beam splitting device can also employ diffractive optical elements, diffracting light of different wavelengths into different directions according to the incident angle of the beam, thereby achieving signal separation. In some embodiments, to simultaneously monitor the positions of multiple particles, the optical tweezers monitoring channel can use a position-sensitive detector array instead of a single four-quadrant detector, with each detector unit corresponding to an independent optical trap.

[0030] The voltage signal read from the four-quadrant detector has a linear relationship with the lateral displacement of the particle, which can be specifically expressed as:

[0031] Where: symbol This represents the differential voltage signal output by the four-quadrant detector, denoted by [symbol]. This represents the displacement detection sensitivity coefficient of the system, measured in millivolts per nanometer, with the symbol [symbol missing]. This represents the lateral displacement of the particle relative to the center of the optical trap. The imaging optical path signal is integrated and exposed on the camera, while the optical tweezers monitoring optical path signal is sampled in real time on a four-quadrant detector at a rate higher than the characteristic frequency of the particle's Brownian motion. Both signals are recorded synchronously by a synchronous acquisition card for parallel processing by the subsequent image reconstruction module and optical tweezers analysis module.

[0032] In one embodiment of the present invention, the improved multifocal image fusion algorithm in the image reconstruction module is implemented as follows. The algorithm's working principle includes pre-calibrating or calculating the three-dimensional point spread function of the system under extended depth-of-field illumination, obtaining a point spread function model with extended depth of field. During imaging, the sample is illuminated using an extended depth-of-field illumination beam, and the camera acquires a series of two-dimensional images focused at different axial positions, forming an original image stack. For each two-dimensional image in the original image stack, the light intensity distribution of the corresponding sample layer in three-dimensional space is estimated using the extended depth-of-field point spread function model through three-dimensional deconvolution operations. Specifically, a single 2D image is considered as a 2D projection of the 3D sample structure after convolution with a 3D point spread function. An imaging model is established, and the 3D sample structure, the point spread function model after depth expansion, and the acquired 2D image are associated through a convolution equation. An iterative deconvolution algorithm is used to solve this convolution equation. The iterative deconvolution algorithm updates the estimate of the 3D sample structure in each iteration. During the iteration process, a total variational regularization constraint is introduced to suppress non-physical oscillations or instabilities caused by noise and the deconvolution process itself, so as to obtain a smoother and physically more reliable 3D light intensity distribution estimate. The iteration stops when the iteration result converges or reaches the preset number of iterations. At this time, the 3D light intensity distribution corresponding to the 2D image is the estimated light intensity distribution of the sample layer in 3D space. Subsequently, image information of each axial layer is extracted from the preliminary three-dimensional light intensity distribution obtained from the three-dimensional deconvolution operation. Based on the information redundancy and signal-to-noise ratio of adjacent layers, the optimal focus position of each image region is adaptively determined. The pixel values ​​of each layer image at its optimal focus position are weighted and fused based on the local image sharpness evaluation function to reconstruct a clear two-dimensional image of the entire sample region. By axial scanning or using the depth information of the extended depth-of-focus illumination beam, the clear two-dimensional images reconstructed at different focus positions are stitched together in three dimensions to generate a three-dimensional structural image of the sample.

[0033] In practical implementation, the working principle of the improved multifocal image fusion algorithm is implemented in a software environment. The first step is to pre-calibrate the three-dimensional point spread function of the system under extended depth-of-field illumination. A 100-nanometer diameter fluorescent microsphere is used as a point source sample, placed on the sample stage. A piezoelectric ceramic displacement stage drives the sample to scan in three-dimensional space with a step accuracy of 50 nanometers. Images of the fluorescent microsphere are acquired at each scanning position, thus obtaining experimental data of the system's three-dimensional point spread function. Three-dimensional Gaussian surface fitting is performed on the experimental data to extract the width, symmetry, and axial extension range parameters of the point spread function, establishing a point spread function model after depth-of-field extension. Model parameters are shown in Table 1. Table 1: Key parameters of the point spread function model after depth of field expansion ; During the imaging process, the biological sample is illuminated using an extended depth-of-focus illumination beam. By controlling the axial position of the microscope objective or sample stage, scanning is performed along the axial depth of the sample in 0.2-micrometer steps. A scientific-grade complementary metal-oxide-semiconductor (CMOS) area array camera acquires one 2D image at each step, for a total of 31 images forming the original image stack. For each 2D image in the original image stack, an improved multifocal image fusion algorithm uses the point spread function model after depth-of-field extension and estimates the light intensity distribution of the corresponding sample layer in 3D space through 3D deconvolution. Specifically, a single 2D image is considered as a 2D projection of the 3D sample structure after convolution with the 3D point spread function. An imaging model is established, and the 3D sample structure, the point spread function model after depth-of-field extension, and the acquired 2D images are correlated through a convolution equation, which is expressed as:

[0034] Where: symbol Represents the acquired two-dimensional image matrix, symbol Represents a three-dimensional point spread function model, symbol Represents a 3D convolution operation, symbol Represents the three-dimensional sample structure to be determined, symbol This represents additive noise in the imaging system.

[0035] In some embodiments, an iterative deconvolution algorithm based on the Richard-Lucy algorithm is used to solve the above convolution equation. The iterative deconvolution algorithm updates the three-dimensional sample structure in each iteration. The estimation is performed by introducing a total variational regularization constraint during the iteration process to suppress non-physical oscillations or instabilities caused by noise and the deconvolution process itself, thereby obtaining a smoother and more physically reliable estimate of the three-dimensional light intensity distribution. The iterative update formula with the regularization term is used to minimize the objective function. The iteration stops when the root mean square error between two adjacent iterations is less than a preset threshold 1e-6, or when the preset number of iterations is reached (200). The three-dimensional light intensity distribution corresponding to the two-dimensional image at this point is the estimated light intensity distribution of the sample layer in three-dimensional space.

[0036] Optionally, image information for each axial layer (corresponding to different objective lens heights) is extracted from the preliminary 3D intensity distribution obtained by 3D deconvolution. The optimal focus position for each image region is adaptively determined based on the information redundancy and signal-to-noise ratio (SNR) of adjacent layers. Information redundancy is evaluated by calculating the mutual information of image patches in adjacent layers, and the SNR is evaluated by calculating the ratio of the local variance to the average intensity of the image patch. Pixel values ​​at their optimal focus positions from each layer are weighted and fused based on a local image sharpness evaluation function. This local image sharpness evaluation function uses the variance of the Laplacian operator based on image gradients to reconstruct a sharp 2D image of the entire sample region. It can be understood that by axial scanning or utilizing the depth-of-field information of the extended depth-of-focus illumination beam, the sharp 2D images reconstructed from different focus positions are stitched together in 3D to generate a 3D structural image of the sample. The 3D stitching process employs a phase-correlation-based image registration algorithm to correct for inter-layer image misalignment that may be caused by scanner nonlinearity. In some embodiments, for the transmission bright-field imaging mode, the three-dimensional point spread function model can be obtained through theoretical calculation, based on scalar diffraction theory and a specific phase pattern loaded on the second spatial light modulator for simulation calculation.

[0037] In one embodiment of the present invention, the optical tweezers analysis module calculates the three-dimensional position and mechanical parameters of a particle as follows: A voltage signal related to the particle's displacement is read from a four-quadrant detector or a position-sensitive detector. Changes in this voltage signal reflect the particle's deviation in two perpendicular directions within the detector plane. Based on a pre-calibrated proportional relationship between the voltage signal and the particle's actual displacement, the voltage signal is converted into two-dimensional position coordinates of the particle in the transverse plane. By analyzing the potential well gradient characteristics formed axially by the optical tweezers-manipulated beam, and utilizing the correlation between the detector signal intensity and the particle's axial position, the axial position of the particle in the optical axis direction is calculated. Combined with the transverse two-dimensional position coordinates, the three-dimensional position of the captured particle is obtained. In the initialization phase of the optical tweezers system, axial position calculation involves scanning a calibration microsphere of known size along its axial direction. The changes in the total light intensity or differential signal output by the four-quadrant detector or position-sensitive detector at different axial positions are recorded, establishing an axial position-signal intensity lookup table. During the actual capture and measurement phase, the total light intensity or differential signal output by the detector is monitored in real time. The monitored signal intensity value is compared with the axial position-signal intensity lookup table to find the axial position corresponding to the best-matching signal intensity value. Interpolation is performed on the lookup results to improve the resolution of axial position localization. Combined with the calculated lateral position of the particle, the complete three-dimensional spatial coordinates of the particle are output. The Brownian motion trajectory of the particle in the optical trap due to thermal motion under undisturbed conditions is recorded. Power spectrum analysis is performed on this Brownian motion trajectory to obtain a Lorentz curve, and the capture stiffness is calculated from the fitted parameters. Based on the displacement of the captured particle's three-dimensional position relative to the center of the optical trap and the capture stiffness, the components of the optical force acting on the captured particle in the three directions are calculated according to Hooke's law.

[0038] In practice, the optical tweezers analysis module reads voltage signals related to particle displacement from the four-quadrant detector. These voltage signals include differential voltage signals in the X and Y directions, and changes in these signals reflect the particle's deviation in two perpendicular directions within the detector plane. Based on a pre-calibrated ratio between the voltage signals and the actual particle displacement, the voltage signals are converted into two-dimensional position coordinates of the particle in the transverse plane. The displacement sensitivity coefficient is calibrated by moving a fixed scatterer a known distance using a piezoelectric displacement stage; a typical displacement sensitivity coefficient is 0.25 mV / nm. By analyzing the potential well gradient characteristics formed along the axial direction by the optical tweezers-manipulated beam, and utilizing the correlation between the total light intensity signal of the four-quadrant detector and the particle's axial position, the axial position of the particle in the optical axis direction is calculated. Combined with the transverse two-dimensional position coordinates, the three-dimensional position of the captured particle is obtained. During the initialization phase of the optical tweezers system, a 1.0-micrometer diameter silica calibration microsphere was axially scanned with a step accuracy of 50 nanometers using a piezoelectric ceramic objective lens positioner. The variation curves of the total voltage signal after conversion by the transimpedance amplifier from the four current signals output by the four-quadrant detector at different axial positions of the calibration microsphere were recorded. An axial position-signal intensity lookup table was established. The key parameters of the calibration process are shown in Table 2. Table 2: Parameter Table for Generating the X-axis Position-Signal Intensity Lookup Table ; In some embodiments, during the actual capture and measurement phase, the total light intensity signal output by the four-quadrant detector, calculated by a summing circuit, is monitored in real time. The monitored total light intensity signal value is compared with the data stored in the axial position-signal intensity lookup table, and the axial position corresponding to the best-matching signal intensity value is found using nearest neighbor interpolation. Cubic spline interpolation is performed on the lookup result to improve the axial position positioning resolution to approximately 10 nanometers. Combined with the calculated particle lateral position, the complete three-dimensional spatial coordinates of the particle are output. The Brownian motion trajectory of the particle in the optical trap due to thermal motion is recorded without external disturbance. The trajectory sampling frequency is 200 kHz, and the recording duration is 10 seconds. Power spectrum analysis is performed on the Brownian motion trajectory. A fast Fourier transform is applied to the position-time data in each direction (X, Y) to calculate the one-sided power spectral density. A Lorentz curve is fitted, and the capture stiffness and power spectral density are calculated from the fitted parameters. With capture stiffness The relationship is:

[0039] Where: symbol The displacement power spectral density of the particle in the X direction is represented by the symbol. Indicates frequency, symbol Represents the Boltzmann constant, symbol Represents absolute temperature, symbol Represents the viscous damping coefficient, symbol Indicates corner frequency, capture stiffness .

[0040] Optional, viscous damping coefficient It is calculated using Stokes' theorem, the formula is as follows: , where the symbol Indicates the viscosity of the medium, symbol This represents the hydrodynamic radius of the particle. Based on the displacement of the trapped particle's three-dimensional position relative to the center of the optical trap, and the trapping stiffness, the components of the optical force acting on the trapped particle in the three directions are calculated according to Hooke's law. The calculation formula is as follows: , where the symbol Represents the optical force in the X direction, symbol Indicates the capture stiffness in the X direction, symbol This represents displacement in the X direction. It is understood that the original position data needs to be detrended before power spectrum analysis to eliminate possible baseline drift. In some embodiments, the capture stiffness in the axial direction can be fitted and calculated using a similar method by analyzing the power spectrum of the total light intensity signal fluctuations.

[0041] In one embodiment of the present invention, the method for visualizing the relative relationship between particles and sample structures in three-dimensional space is as follows: A unified three-dimensional spatial coordinate system is established, with the optical axis of the microscope objective as the Z-axis of the unified three-dimensional spatial coordinate system, and two directions perpendicular to the optical axis and orthogonal to each other as the X-axis and Y-axis. A calibration sample with known spatial characteristics is placed on the sample stage. An image of the calibration sample is obtained through three-dimensional imaging. The three-dimensional image coordinates of known feature points in the calibration sample image are identified and a correspondence is established with the actual physical coordinates of the feature points on the sample stage. By solving the coordinate transformation parameters, a transformation relationship from image voxel coordinates to physical coordinates in the unified three-dimensional spatial coordinate system is established. The original coordinates of the particle three-dimensional position calculated by the optical tweezers monitoring channel are mapped to the unified three-dimensional spatial coordinate system through a pre-calibrated transformation relationship to ensure spatial consistency between imaging coordinates and manipulation coordinates. In a three-dimensional rendering environment, a three-dimensional structural image of the sample is loaded, and the three-dimensional morphology of the sample is displayed in a volumetric or surface rendering manner. In this three-dimensional rendering environment, the real-time three-dimensional position of the captured particles is dynamically calibrated with highlighted markers of customizable shape and color. Based on user interaction commands, the viewing angle and scaling ratio of the 3D rendering view can be dynamically adjusted, and the cross-sectional display function can be enabled to observe the spatial relationship between particles and the internal structure of the sample. The system updates the 3D position of the captured particles in real time and updates the position of the marker points synchronously in the 3D rendering environment, forming a dynamic visualization of the movement of particles within the 3D structure of the sample.

[0042] In practice, a unified three-dimensional spatial coordinate system is established, with the optical axis of the microscope objective as the Z-axis, and two directions perpendicular to the optical axis and orthogonal to each other as the X and Y axes. The directions of the X and Y axes are aligned with the movement direction of the scanning galvanometer or sample stage. A calibration sample with known spatial characteristics is placed on the sample stage. The calibration sample is a chromium-layer calibration plate with a precise three-dimensional mesh structure, fabricated on a glass substrate using photolithography. The mesh period is 10 micrometers in both the X and Y directions, and it has a stepped structure with a height of 5 micrometers in the Z direction. The image stack of the calibration sample in multiple focal planes is obtained through three-dimensional imaging. The three-dimensional image coordinates of known feature points in the calibration sample image are identified and a correspondence is established with the actual physical coordinates of the feature points on the sample stage. The actual physical coordinates of the feature points are given by the manufacturing tolerance certificate of the calibration plate, with an accuracy better than 50 nanometers.

[0043] In some embodiments, by solving for coordinate transformation parameters, a transformation relationship is established from image voxel coordinates to physical coordinates in a unified three-dimensional spatial coordinate system. This transformation relationship is achieved through an affine transformation matrix that includes rotation, scaling, and translation. The original coordinates of the particle's three-dimensional position calculated by the optical tweezers monitoring channel are obtained based on the voltage-displacement relationship of the four-quadrant detector and an axial lookup table. These original coordinates are then mapped to a unified three-dimensional spatial coordinate system using a pre-calibrated transformation relationship. The transformation formula is as follows:

[0044] Where: symbol Represents the physical coordinates in the unified world coordinate system after transformation, with the symbol... Represents the lateral coordinate calculated from the signals of the four-quadrant detector, with the sign... This represents the axial coordinate obtained from the axial position-signal strength lookup table, with the symbol... This represents a 4x4 affine transformation matrix, which is calculated by fitting known calibration point pairs using the least squares method, ensuring that the spatial consistency between imaging coordinates and manipulation coordinates is better than 100 nanometers in three-dimensional space.

[0045] Optionally, in the 3D rendering environment, a 3D structural image of the sample generated by the image reconstruction module is loaded, and the 3D morphology of the sample is displayed in a volumetric rendering manner. The volumetric rendering employs a ray casting-based algorithm and applies a transfer function to map voxels of different grayscale values ​​to different colors and transparency. In the 3D rendering environment, the real-time 3D position of the captured particles is dynamically marked with highlighted markers of customizable shape and color. The markers are displayed by default as red spheres with a virtual diameter of 200 nanometers. It can be understood that, based on user interaction commands, the viewpoint and scaling ratio of the 3D rendering view can be dynamically adjusted, and a cross-sectional display function can be enabled to observe the spatial relationship between the particles and the internal structure of the sample. The cross-sectional display is achieved by defining an interactive clipping plane to cut the volume data in real time. In some embodiments, the system updates the 3D position of the captured particles obtained from the optical tweezers analysis module in real time, and synchronously updates the position of the markers in the 3D rendering environment through a callback function of the graphics interface. The update frequency of the marker positions is consistent with the output frequency of the optical tweezers position data, reaching up to 10 kHz, thereby forming a dynamic visualization of the particle movement within the 3D structure of the sample. This dynamic image can be recorded as a video file for subsequent analysis.

[0046] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention in any other way. Any person skilled in the art may make changes or modifications to the above-disclosed technical content to create equivalent embodiments that can be applied to other fields. However, any simple modifications, equivalent changes, and modifications made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the protection scope of the present invention.

Claims

1. An extended depth-of-focus microscopic imaging system incorporating optical tweezers, characterized in that, The system includes: The beam generation and combining module generates an extended depth-of-focus illumination beam with phase modulation and an optical tweezers manipulation beam with a specific spatial shape. The extended depth-of-focus illumination beam and the optical tweezers manipulation beam are spatially coaxially combined to form a combined beam. The signal acquisition module guides the combined light beam into the microscope objective and focuses it on the sample area. And collect the light signals that pass through or are scattered by the sample; The signal separation module performs beam splitting and separation processing on the collected optical signals, separating the imaging optical path signal and the optical tweezers monitoring optical path signal into different detection channels; The image reconstruction module processes the imaging optical path signal in the imaging detection channel using an improved multifocal image fusion algorithm. The improved multifocal image fusion algorithm is optimized based on the point spread function model after depth expansion to reconstruct the three-dimensional structural image of the sample. The optical tweezers analysis module analyzes the optical path signal of the optical tweezers monitoring channel, calculates the three-dimensional position, capture stiffness and optical force of the captured particles, and visualizes the relative relationship between the particles and the sample structure in three-dimensional space based on the reconstructed three-dimensional structural image of the sample and the calculated three-dimensional position of the captured particles.

2. The extended depth-of-focus microscopic imaging system combining optical tweezers function according to claim 1, characterized in that, The process involves generating a phase-modulated extended depth-of-focus illumination beam and a tweezers-manipulated beam with a specific spatial morphology, and then spatially coaxially combining the extended depth-of-focus illumination beam and the tweezers-manipulated beam to form a combined beam, including: The original beam emitted from the light source passes through a polarization control device and forms two linearly polarized beams with orthogonal polarization states. One of the linearly polarized beams is incident on a first spatial light modulator, which is loaded with a computational hologram. The computational hologram is optimized to generate a wavefront with a ring intensity distribution or a Bessel beam shape, thereby generating the optical tweezers manipulated beam. Another beam of linearly polarized light is incident on a second spatial light modulator, which is loaded with a phase modulation pattern. The phase modulation pattern is used to encode the incident wavefront to generate an illumination beam with an extended focal range along the optical axis, i.e., the extended depth-of-focus illumination beam. The optical tweezers-manipulated beam modulated by the first spatial light modulator and the extended depth-of-focus illumination beam modulated by the second spatial light modulator are combined through a polarizing beam splitter, so that the two beams propagate along the same optical path to form the polarization-multiplexed combined beam. The combined beam is expanded and collimated to fill the back aperture of the microscope objective.

3. The extended depth-of-focus microscopic imaging system combining optical tweezers function according to claim 2, characterized in that, The combined light beam is guided into the microscope objective, focused on the region where the sample is located, and the light signal passing through or scattered by the sample is collected, including: The extended depth-of-focus illumination beam forms an extended uniform illumination focal zone within the sample for large depth-of-field imaging of the sample. At the same time, the optical tweezers manipulate the beam to form a three-dimensional optical potential trap within the illumination focal zone for capturing and manipulating micro- and nano-particles in the sample. The optical signal includes the imaging signal generated by the extended depth-of-focus illumination beam and the scattering signal generated by the interaction between the optical tweezers-manipulated beam and the particles. The extended depth-of-focus illumination beam forms an extended, uniform illumination focal zone within the sample for large depth-of-field imaging. Simultaneously, the optical tweezers manipulate the beam to form a three-dimensional optical potential trap within the illumination focal zone, used to capture and manipulate micro / nano particles in the sample. Specifically, this includes: After the combined beam is focused by the microscope objective, the phase modulation wavefront of the extended depth-of-focus illumination beam generates self-reconstruction characteristics along the optical axis near the focal plane, thereby forming an illumination area in the sample with a size larger than that of traditional point focusing along the optical axis. The illumination area is the extended uniform illumination focal zone. The intensity distribution of the illumination focal area is uniform along the axial direction, so that sample structures at different heights within the axial range of the illumination focal area can obtain consistent illumination intensity, thereby achieving the large depth-of-field imaging. Meanwhile, the optical tweezers manipulate the beam to form a series of intensity maxima on the beam propagation path due to its annular intensity distribution or Bessel beam shape, and these intensity maxima constitute the three-dimensional optical potential trap. The three-dimensional optical potential trap overlaps spatially with the extended uniform illumination focal zone formed by the extended depth-of-focus illumination beam in both axial position and lateral range, such that the captured micro-nano particles are located within the imaging illumination region. By dynamically controlling the computational hologram loaded on the first spatial light modulator, the wavefront of the optical tweezers manipulated beam is changed, thereby moving the spatial position of the three-dimensional optical potential trap within the illumination focal zone, thus realizing the three-dimensional capture and manipulation of the micro-nano particles.

4. The extended depth-of-focus microscopic imaging system combining optical tweezers function according to claim 1, characterized in that, The process of splitting and separating the collected optical signals to separate the imaging optical path signal and the optical tweezers monitoring optical path signal into different detection channels specifically includes: The collected optical signal is guided to a beam splitter, which splits the optical signal into different paths according to the wavelength, polarization state, or light intensity. In the imaging optical path direction, a bandpass filter is provided to selectively transmit light corresponding to the wavelength of the extended depth-of-focus illumination beam, while blocking light of the wavelength of the optical tweezers-controlled beam and other background stray light, thereby forming the imaging optical path signal; In the direction of the optical tweezers monitoring optical path, a four-quadrant detector or a position-sensitive detector is set to receive weak signals related to the optical tweezers control beam that are scattered or refracted by particles in the sample, thus forming the optical tweezers monitoring optical path signal. The imaging optical path signal is imaged onto the photosensitive surface of a scientific array camera using a lens group. The optical tweezers monitor the optical path signal and use a focusing lens to converge it to the center of the sensitive area of ​​the four-quadrant detector or the position-sensitive detector.

5. The extended depth-of-focus microscopic imaging system combining optical tweezers function according to claim 4, characterized in that, The improved multifocal image fusion algorithm is optimized based on the point spread function model after depth extension, and its working principle includes: The three-dimensional point spread function of the system is pre-calibrated or calculated under the illumination condition of the extended depth-of-field illumination beam, and the point spread function model after the depth of field extension is obtained. During the imaging process, the sample is illuminated by the extended depth-of-focus illumination beam, and the camera acquires a series of two-dimensional images focused at different axial positions, forming an original image stack. For each two-dimensional image in the original image stack, the light intensity distribution of the sample layer corresponding to the two-dimensional image in three-dimensional space is estimated by using the point spread function model after the depth of field expansion and through three-dimensional deconvolution operation. From the preliminary three-dimensional light intensity distribution obtained by three-dimensional deconvolution operation, image information of each axial layer is extracted, and the optimal focus position of each image region is adaptively determined according to the information redundancy and signal-to-noise ratio of adjacent layers. The pixel values ​​of each image layer at its optimal focus position are weighted and fused based on the local image sharpness evaluation function to reconstruct a clear two-dimensional image of the entire sample area. By axial scanning or utilizing the depth-of-field information of the extended depth-of-focus illumination beam, clear two-dimensional images reconstructed from different focus positions are stitched together in three dimensions to ultimately generate a three-dimensional structural image of the sample.

6. The extended depth-of-focus microscopic imaging system combining optical tweezers function according to claim 5, characterized in that, For each two-dimensional image in the original image stack, the point spread function model after depth expansion is used to estimate the light intensity distribution of the corresponding sample layer in three-dimensional space through three-dimensional deconvolution operation, including: The single two-dimensional image in the original image stack is regarded as a two-dimensional projection of the three-dimensional sample structure after being convolved by the three-dimensional point spread function. An imaging model is established, which correlates the three-dimensional sample structure, the point spread function model after depth expansion, and the acquired two-dimensional image through a convolution equation. The convolution equation is solved using an iterative deconvolution algorithm, which updates the estimate of the three-dimensional sample structure in each iteration. During the iteration process, a total variational regularization constraint is introduced to suppress non-physical oscillations or instabilities caused by noise and the deconvolution process itself, so as to obtain a smoother and more physically reliable three-dimensional light intensity distribution estimate. When the iteration results converge or the preset number of iterations is reached, the iteration stops. The three-dimensional light intensity distribution corresponding to the two-dimensional image at this time is the estimated light intensity distribution of the sample layer in three-dimensional space.

7. The extended depth-of-focus microscopic imaging system combining optical tweezers function according to claim 5, characterized in that, In the optical tweezers monitoring channel, the optical tweezers monitoring optical path signal is analyzed to calculate the three-dimensional position, capture stiffness, and optical force acting on the captured particle, including: The voltage signal related to the particle displacement is read from the four-quadrant detector or the position-sensitive detector, and the change in the voltage signal reflects the amount of deviation of the particle in two perpendicular directions in the detector plane; Based on the pre-calibrated proportional relationship between the voltage signal and the actual displacement of the particle, the voltage signal is converted into two-dimensional position coordinates of the particle in the horizontal plane; By analyzing the potential well gradient characteristics formed by the optical tweezers manipulating the beam along the axial direction, and utilizing the correlation between the detector signal strength and the particle's axial position, the axial position of the particle in the optical axis direction is calculated, and the three-dimensional position of the captured particle is obtained by combining the transverse two-dimensional position coordinates. The Brownian motion trajectory of the particle in the optical trap due to thermal motion is recorded without external disturbance. Power spectrum analysis is performed on the Brownian motion trajectory to obtain the Lorentz curve. The trapping stiffness is calculated from the fitting parameters. Based on the displacement of the three-dimensional position of the captured particle relative to the center of the optical trap, and the trapping stiffness, the components of the optical force on the captured particle in the three directions are calculated according to Hooke's law.

8. The extended depth-of-focus microscopic imaging system combining optical tweezers function according to claim 7, characterized in that, By analyzing the potential well gradient characteristics formed along the axial direction by the optical tweezers-manipulated beam, and utilizing the correlation between the detector signal intensity and the particle's axial position, the axial position of the particle in the optical axis direction is calculated, including: During the initialization phase of the optical tweezers system, a calibration microsphere of known size is scanned axially, and the change curves of the total light intensity signal or differential signal output by the four-quadrant detector or position-sensitive detector are recorded when the calibration microsphere is in different axial positions, thus establishing an axial position-signal intensity lookup table. During the actual capture and measurement phase, the total light intensity signal or differential signal output by the four-quadrant detector or position-sensitive detector is monitored in real time. The real-time monitored signal strength value is compared with the axial position-signal strength lookup table to find the axial position corresponding to the signal strength value that best matches it. Interpolation is performed on the search results to improve the resolution of axial position positioning; Based on the calculation results of the particle's lateral position, the complete three-dimensional spatial coordinates of the particle are output.

9. The extended depth-of-focus microscopic imaging system combining optical tweezers function according to claim 1, characterized in that, The visualization of the relative relationship between the particles and the sample structure in three-dimensional space, based on the reconstructed three-dimensional structural image of the sample and the calculated three-dimensional position of the captured particles, includes: Establish a unified three-dimensional spatial coordinate system and transform the voxel coordinates of the three-dimensional structural image of the sample and the three-dimensional position coordinates of the captured particles to the unified coordinate system; In a 3D rendering environment, the 3D structural image of the sample is loaded, and the 3D morphology of the sample is displayed in a volume rendering or surface rendering manner. In the 3D rendering environment, the real-time 3D position of the captured particles is dynamically marked with highlighted markers of customizable shape and color. Based on user interaction commands, the viewing angle and scaling ratio of the 3D rendering view can be dynamically adjusted, and the cross-sectional display function can be enabled to observe the spatial relationship between particles and the internal structure of the sample. The three-dimensional position of the captured particles is updated in real time, and the position of the marker points is updated synchronously in the three-dimensional rendering environment to form a dynamic visualization of the particles moving within the three-dimensional structure of the sample.

10. The extended depth-of-focus microscopic imaging system combining optical tweezers function according to claim 9, characterized in that, The establishment of a unified three-dimensional spatial coordinate system includes: The optical axis of the microscope objective is taken as the Z-axis of the unified three-dimensional spatial coordinate system, and the two directions perpendicular to the optical axis and orthogonal to each other are taken as the X-axis and Y-axis. A calibration sample with known spatial characteristics is placed on the sample stage, and an image of the calibration sample is obtained through three-dimensional imaging. Identify the three-dimensional image coordinates of known feature points in the calibration sample image and establish a correspondence between them and the actual physical coordinates of the feature points on the sample stage; By solving the coordinate transformation parameters, the transformation relationship between image voxel coordinates and physical coordinates in a unified three-dimensional spatial coordinate system is established. The three-dimensional position of the particle calculated by the optical tweezers monitoring channel is mapped to the unified three-dimensional spatial coordinate system through a pre-calibrated transformation relationship to ensure the spatial consistency between the imaging coordinates and the manipulation coordinates.

Images