Near-infrared deep tissue focusing system based on fast mechanical calibration and method thereof

Abstract

The application discloses a near-infrared deep tissue focusing system based on fast mechanical calibration, which comprises a control device, a laser, a pose adjusting device, a first unpolarized beam splitting cube, a second unpolarized beam splitting cube, a third unpolarized beam splitting cube, a first image sensor and a spatial light modulator, wherein the first image sensor and the spatial light modulator are respectively arranged at symmetrical positions of a reflecting surface of the first unpolarized beam splitting cube, the second unpolarized beam splitting cube is arranged between the first unpolarized beam splitting cube and the spatial light modulator, and the third unpolarized beam splitting cube is arranged between the first unpolarized beam splitting cube and the first image sensor. The application further discloses a near-infrared deep tissue focusing method based on fast mechanical calibration. Through the arrangement of the light path and the setting of the pixel matching system and the focusing system, the symmetry degree of the light modulator and the image sensor is greatly improved, so that the precision and speed of light focusing through the scattering medium are improved.

Description

Technical Field

[0001] This invention relates to the field of optical focusing technology for scattering media, and more specifically, to a near-infrared deep tissue focusing system and method based on rapid mechanical calibration. Background Technology

[0002] Focusing light within a scattering medium has long been a goal in optics, with wide applications in biomedical optics. The depth and precision of focusing directly affect the depth and resolution of imaging, thus impacting the accuracy of medical diagnosis. Time-reversed ultrasound-coded (TRUE) focusing, combining the relative transparency of ultrasound within biological tissues with efficient wavefront shaping based on digital optical phase conjugation (DOPC), has been proposed to address this challenge. By invoking repeated acousto-optic interactions, iterative TRUE (iTRUE) focusing can further overcome the resolution barrier imposed by the acoustic diffraction limit, demonstrating significant potential for deep tissue biomedical applications.

[0003] In the prior art, there is an acousto-optic modulation phase conjugation method for achieving internal focusing in a scattering medium, including: phase extraction and phase conjugation restoration processes; in the phase extraction process, the object light path uses acousto-optic modulation to construct a "point light source" inside the scattering medium, and its outgoing speckle interferes with the modulated reference light that generates a corresponding frequency shift. By adjusting the polarization state of the reference light, the phase difference between the object light and the reference light is changed, and the phase conjugation system obtains a polarization phase-shifted interferogram. A four-step phase-shifting method is used to extract the phase of the light field; in the phase conjugation restoration process, the phase conjugation map of the object light is loaded onto the spatial light modulator, and the reference light directly illuminates it to generate time-reversed light, thereby achieving internal focusing in the scattering medium; the present invention introduces an ultrasonic modulation signal on the basis of a polarization phase-shifting digital optical phase conjugation system, which can construct an acousto-optic interaction point inside the scattering medium, thereby achieving optical focusing inside the scattering medium using time reversal.

[0004] However, the focusing accuracy of the DOPC and iTRUE focusing systems, as well as the iTRUE system based on the phase conjugate system, is still limited by the symmetry of the optical modulator and image sensor, which seriously affects the practical use of DOPC and iTRUE focusing. Summary of the Invention

[0005] The purpose of this invention is to overcome the shortcomings of existing acousto-optic modulation phase conjugation methods, which require strict symmetry between the system and the optical modulator and image sensor when focusing inside the scattering medium, making them difficult to use in practice. This invention provides a near-infrared deep tissue focusing system and method based on rapid mechanical calibration. By arranging the optical path and setting up a pixel matching system and a focusing system, this invention significantly improves the symmetry between the optical modulator and image sensor, thereby improving the accuracy and speed of focusing light through the scattering medium.

[0006] The objective of this invention can be achieved using the following technical solutions:

[0007] A near-infrared deep tissue focusing system based on rapid mechanical calibration includes a control device, a laser, a pose adjustment device, a first unpolarized beam splitter cube, a second unpolarized beam splitter cube, a third unpolarized beam splitter cube, a first image sensor, and a spatial light modulator. The first image sensor and the spatial light modulator are respectively disposed symmetrically on the reflective surface of the first unpolarized beam splitter cube. The second unpolarized beam splitter cube is disposed between the first unpolarized beam splitter cube and the spatial light modulator. The third unpolarized beam splitter cube is disposed between the first unpolarized beam splitter cube and the first image sensor. An optical path adjustment structure is provided between the laser and the first, second, and third unpolarized beam splitter cubes so that the laser beam passes through the first, second, and third unpolarized beam splitter cubes respectively. The spatial light modulator is disposed on the pose adjustment device. The control device is electrically connected to the laser, the pose adjustment device, and the first image sensor.

[0008] It also includes a pixel matching system, which includes a pair of first lenses and a pair of mirrors. The mirrors include a first mirror and a second mirror. The first mirror and the first unpolarized beam splitter cube are located at opposite ends of the optical axis of the pair of lenses, and the reflecting surface of the second mirror is aligned with one side of the second unpolarized beam splitter cube.

[0009] It also includes a focusing system disposed between the first non-polarized beam splitter cube and the optical path adjustment structure. The focusing system includes a second lens, a medium fixing structure, a fourth non-polarized beam splitter cube, and a second image sensor disposed between the first non-polarized beam splitter cube and the optical path adjustment structure. The second lens, the medium fixing structure, and the fourth non-polarized beam splitter cube are arranged in sequence. The optical path adjustment structure is located on the other side of the fourth non-polarized beam splitter cube relative to the medium fixing structure. The second image sensor and the medium fixing structure are respectively located at symmetrical positions on the reflective surface of the fourth non-polarized beam splitter cube.

[0010] Phase conjugation systems involve numerous components, each requiring high pose precision, a contradictory requirement for general optical systems. Furthermore, various components exhibit uncontrollable aberrations. The fundamental principle of phase conjugation utilizes the time symmetry of the wave equation's wave function solution to achieve a time-reversal-like effect in electromagnetic wave propagation. In a digital optical phase conjugation (DOPC) system composed of an image sensor and a spatial light modulator (SLM), the image sensor and SLM are positioned symmetrically on the reflective surface of a non-polarized beam splitter cube (BS). The control device, such as a computer, calculates the complex amplitude distribution (or phase map) from the speckle interferogram recorded by the image sensor using a four-step phase-shifting holography, conjugates it, and immediately loads the processed phase map onto the SLM, thereby achieving reverse propagation of light waves and focusing through the scattering medium. However, since the SLM and image sensor in the DOPC system are on opposite sides of the BS and invisible to each other, the first challenge in calibration is pixel matching between the image sensor and the SLM.

[0011] Therefore, to achieve pixel matching between the image sensor and the SLM, a pose adjustment device is used to adjust the SLM so that the pixels of the first image sensor match the bright and dark grid formed after interference by the SLM. In this process, the present invention incorporates a pixel matching system, in which a pair of first lenses form an optical 4f system. Both the SLM and the first image sensor are positioned on the front focal plane of the first first lens, and the first reflecting mirror is positioned on the rear focal plane of the second first lens. The advantage of the 4f system is that even if the first lens has a certain offset within its plane, due to the principle of optical path reversibility, secondary imaging can correct for small displacement errors. Furthermore, the pitch and tilt angles of the reflecting mirror only affect the angle of the image, not its position, greatly improving the tolerance of optical path adjustment. Thus, the SLM can perfectly image onto the first image sensor, achieving better pixel matching. Simultaneously, a second reflecting mirror is used to reflect the reference light that has not been reflected by the SLM back into the optical path, thereby interfering with the reflected light from the spatial modulator.

[0012] After pixel matching is completed, a scattering medium is placed in the medium-fixed structure. At this point, the second image sensor can only obtain a very blurry focus because pixel matching cannot calibrate the pitch and tilt angles of the SLM to place it in an ideal posture. Therefore, while adjusting each of the six degrees of freedom of the SLM one by one using a pose adjustment device, the intensity of the focus within the second image sensor is recorded. The absolute coordinates of the electric displacement stage and rotary stage when the highest peak intensity is reached are recorded to obtain the optimal pose of the SLM. The pose adjustment device can be a six-axis motor or other devices that can be adjusted in six degrees of freedom of the SLM. Since only simple serial port commands need to be sent to the motor without the need for massive complex amplitude and matrix operations, the entire mechanical calibration process generally only takes 10 minutes. Adaptive stepping accuracy scanning with PID control can also be added, and the mechanical calibration can be shortened to 3-5 minutes, which is much less than the time consumed by other DOPC systems in the world for pure digital six-axis pose compensation.

[0013] After the coarse adjustment of pixel matching and the fine adjustment of the six degrees of freedom of the SLM, a very sharp focus has now appeared on the second image sensor.

[0014] Furthermore, it also includes an ultrasonic focusing system, which includes an ultrasonic transducer and a light-blocking structure. The ultrasonic transducer is disposed between the medium fixing structure and the fourth non-polarized beam splitter cube, and the light-blocking structure allows light to pass through or blocks light from passing through between the fourth non-polarized beam splitter cube and the optical path adjustment structure.

[0015] The frequency of use is f US The self-focusing ultrasonic transducer emits an ultrasonic beam into the scattering medium. The focal point of the emitted ultrasonic beam is located between the fourth unpolarized beam splitter cube and the scattering medium. When the initially incident light passes through the ultrasonic focal point, due to the strong acousto-optic effect, some of the reflected light undergoes a frequency shift. This portion of light has a frequency close to that of the reference light and can therefore be detected by heterodyne interferometry. Subsequently, the phase conjugate system only inverts the reflected light that has passed through the ultrasonic focal point, thus focusing the reflected light into the scattering medium. Furthermore, the DOPC system is easily integrated with various guiding targets to further achieve focusing within the scattering medium. In biological tissues, compared to the strong scattering characteristics of light, ultrasound has excellent penetrability due to its longer wavelength, making it a high-performance guiding target.

[0016] Furthermore, the ultrasonic focusing system also includes a transparent water tank, with the fourth non-polarized beam splitter cube designed inside the transparent water tank, the focal point of the ultrasonic transducer located inside the transparent water tank, and a semi-transparent mirror provided between the focal point of the ultrasonic transducer and the fourth non-polarized beam splitter cube.

[0017] Because air has high impedance to ultrasound, the ultrasonic transducer should be immersed in water for better ultrasonic wave transmission. To facilitate observation of the focusing process, the TRUE system typically uses a sandwich structure (scattering medium-water-scattering medium) instead of a large scattering medium, allowing light to propagate through a distance of free space. Conventional iTRUE systems add a fluorescent film at the focal beam location to photograph the scattering medium from the side with a camera. This method is extremely difficult to implement in the invisible light band, has very low energy efficiency, and only provides a cross-sectional view of the beam, failing to visually represent the two-dimensional distribution of the focal point. This invention replaces one of the scattering media pieces with a semi-transparent, semi-reflective mirror, allowing the frequency-shifted light energy to be returned to the second image sensor while also enabling excellent observation of the focal point.

[0018] Furthermore, the optical path adjustment structure includes a mirror that allows the laser emitted by the laser to enter the second unpolarized beam splitter cube, a half-wave plate located on the laser optical path, and a first beam expander located between the half-wave plate and the second unpolarized beam splitter cube.

[0019] The optical path adjustment structure also includes a fifth unpolarized beam splitting cube for splitting the laser emitted by the laser, an acousto-optic modulator (AOM) and a second beam expander arranged sequentially in the reflected light direction of the fifth unpolarized beam splitting cube, and an acousto-optic modulator disposed between the third unpolarized beam splitting cube and the half-wave plate. The fifth unpolarized beam splitting cube is provided with half-wave plates in both the incident and reflection directions.

[0020] The optical path adjustment structure further includes a sixth unpolarized beam splitting cube for splitting the laser emitted by the laser, an acousto-optic modulator disposed between the fourth unpolarized beam splitting cube and the sixth unpolarized beam splitting cube, and an acousto-optic modulator disposed between the sixth unpolarized beam splitting cube and a half-wave plate. The sixth unpolarized beam splitting cube is provided with a half-wave plate in both the incident and reflection directions.

[0021] This solution uses the above-described structure to simply and directly arrange the optical path. The positions and numbers of the reflectors and non-polarizing beam splitters can be changed as needed, allowing the laser beam to enter the first, second, and third non-polarizing beam splitters respectively. The first and second beam expanders can be beam expanders composed of two lenses, or other structures that can change the beam size; these will not affect the implementation of this solution.

[0022] Furthermore, the laser is a near-infrared light generator.

[0023] Deep biological tissue focusing and imaging has long been hampered by the massive absorption and excessive scattering of photons by various particles, hindering its application. Wavefront shaping techniques can only overcome the scattering of light by the scattering medium, but cannot eliminate the effects of absorption. Most biological tissues absorb visible light strongly, causing unnecessary energy loss. Furthermore, although phase-conjugated systems possess powerful performance, they are limited by the size of spatial light modulators (SLMs), remaining only partially spatially controlled systems. This means they cannot control scattered light spilling beyond the optical path. Near-infrared light, with its longer wavelength than visible light and a size comparable to many scatterers within the body, exhibits weaker Mie scattering and can penetrate deeper tissues. Simultaneously, the superior properties of near-infrared light allow it to penetrate thicker muscle tissue and thinner bone, making this technology promising for applications such as subcutaneous laser therapy, non-invasive high-resolution brain microscopy, and optogenetics research. However, in the invisible light band, the optical path construction and error control of phase-conjugated systems are particularly difficult. This invention provides a focusing system with rapid calibration and high focusing accuracy, significantly improving the usability of focusing light through scattering media using infrared light sources.

[0024] A near-infrared deep tissue focusing method based on rapid mechanical calibration includes the following steps:

[0025] S1: The reference light formed by the light emitted by the laser after passing through the optical path adjustment structure causes interference between the reference light and the reflected light of the spatial light modulator, thus giving the image generated by the spatial light modulator a specific interference pattern.

[0026] The reference light, formed by the acousto-optic modulator (AOM), enters the third unpolarized beam splitter cube. The laser, after entering the second unpolarized beam splitter cube through the optical path adjustment structure, enters the spatial light modulator (SLM) to form the reflected light of the spatial light modulator.

[0027] S2: Adjust the pose adjustment device to control the Δx, Δy, Δz, and Δθ of the spatial light modulator. z The four degrees of freedom are coarsely adjusted so that the first image sensor senses the image generated by the spatial light modulator and the image generated by the spatial light modulator corresponds to each pixel of the first image sensor.

[0028] S3: A scattering medium is placed in a medium-fixed structure so that the reference light and the sample light passing through the scattering medium form heterodyne interference. The control device continuously records the multi-frame interference diagrams identified by the first image sensor and calculates the phase diagram. Then, the calculated phase diagram is loaded onto the spatial light modulator so that the light beam penetrates the scattering medium in reverse, re-forms a parallel light beam that enters the sample, and focuses it on the second image sensor.

[0029] After the laser enters the fourth unpolarized beam-splitting cube through the optical path adjustment structure, it enters the scattering medium, forming sample light that passes through the scattering medium. The sample light enters from one side of the scattering medium, and the newly formed parallel beam incident on the sample re-enters the scattering medium from the other side. The first sensor records at least two frames of interferograms.

[0030] S4: The control device controls the pose adjustment device to adjust the Δx, Δy, Δz, and Δθ of the spatial light modulator. x , Δθ y , Δθ z The six degrees of freedom are adjusted one by one, and at the same time the control device records the peak intensity of the focal point in the image identified by the second image sensor and the corresponding coordinate position and angle of the spatial light modulator to obtain the optimal pose of the spatial light modulator.

[0031] Furthermore, step S3 also includes the following steps:

[0032] Adjust the frequency of the acousto-optic modulator (AOM) to f US And make the beat frequency f of both M The frame rate is set to 10Hz, allowing heterodyne interference between the sample light and the reference light generated by the AOM. The first image sensor is a scientific-grade CMOS camera (sCMOS), and the second image sensor is an industrial-grade CMOS camera. The sCMOS frame rate is set to 4fps. M After recording four consecutive interferograms, the conjugate of the calculated phase diagram is loaded onto the spatial light modulator to make the beam penetrate the scattering medium in reverse, re-form a parallel beam that enters the sample, and focus it on the CMOS.

[0033] Furthermore, it also includes the following steps:

[0034] S5: Take a multi-order Zernike polynomial, scan the coefficients of each order from low to high order, and obtain the optimal coefficients of each order by feedback from the focal peak intensity of the second image sensor. Then, superimpose the final Zernike compensated phase map onto the conjugate phase map in step S3.

[0035] After step S4, a very sharp focus appears on the second image sensor, but it still falls short of the DOPC system's performance limits. One reason is that the surface curvature of the SLM is not ideal; current wafer manufacturing processes still have certain defects, and the backplane of a single LCoS (liquid crystal on silicon) wafer may have some curvature. Although the height difference from the center to the edge of this deformation is generally less than 10 μm, having almost no impact on geometric optics imaging systems, it degrades the modulation wavefront in phase-modulated DOPC systems. Secondly, limited by the manufacturing processes of various optical components in the optical path, such as insufficient flatness of mirrors and the inability to fully correct higher-order aberrations of lenses, slight wavefront aberrations inevitably appear in all beams. These defects are completely unrelated to the SLM's pose and can only be digitally compensated by loading the corresponding phase map onto the SLM.

[0036] For typical DOPC devices, the SLM at the system's end functions as both optical field control and error compensation. Specifically, during system debugging, the accumulated phase distortion of the system needs to be calculated using a specific feedback algorithm. This phase map is then preloaded onto the SLM, and the wavefront of the scattered light from the sample is phase-conjugated to achieve high-quality focusing through the scattering medium. However, this digital compensation method consumes a significant amount of time in the initial calculations, has limited accuracy and a small adjustment range, and cannot compensate for larger errors in invisible light systems.

[0037] Therefore, this invention performs digital compensation based on the mechanical calibration of the focusing system. Characterized by Zernike polynomials, these polynomials, through linear superposition, can generate almost arbitrary phase diagrams. Solving for the coefficients of each order of the Zernike polynomials is similar to the 6-DOF correction of the SLM pose. Therefore, this invention uses Zernike polynomials for digital compensation, thereby obtaining a sharper focus. Step S4 of this invention performs fine-tuning by adjusting the SLM displacement. Because it is mechanical calibration, the speed is very fast. Then, in step S5, digital compensation further improves the calibration accuracy. This invention combines mechanical calibration and digital compensation, and its calibration speed and final focusing accuracy are higher than those of typical visible light DOPC systems.

[0038] Furthermore, in step S5, an 8th-order Zernike polynomial is used; the scanning range of each order coefficient is [-2,2], and the step precision is 0.01.

[0039] The phase map is multiplied by different coefficients to detect changes in focal point brightness in real time, and the coefficient corresponding to the strongest focal point is recorded. Focusing accuracy is significantly improved when performing scans of orders 1 to 3, but higher-order Zernikal aberrations exhibit greater shape changes and increase scanning time. In this scheme, an 8th-order Zernikal polynomial is used.

[0040] Furthermore, it also includes the following steps:

[0041] S6: The ultrasonic transducer emits an ultrasonic beam into the scattering medium, and the incident light in the scattering medium passes through the ultrasonic intersection point, causing a frequency shift in some of the reflected light. Then, the reference light generated by the laser is heterodyne-interfered with this part of the reflected light to detect the reflected light. Finally, the control device inverts the reflected light that has passed through the ultrasonic focal point, and can then focus it into the scattering medium.

[0042] Furthermore, it also includes the following steps:

[0043] S7: Iterate the inverted light as the incident light in the scattering medium multiple times;

[0044] In step S6, a regularly focused beam is first incident into the medium. A weak signal light (marker light) is reflected from the point where the ultrasound focal point is located. This reflected light is then inverted using phase conjugation. It is observed that the full width at half maximum (FWHM) of the initially formed focal point is close to the FWHM of the ultrasound focal point. Because the signal-to-noise ratio (SNR) of the first reflected light is very low, the quality of the inverted focal point within the medium is poor. However, this time, more light hits near the ultrasound focal point, so the SNR is significantly improved when it is reflected again. The reflected marker light is then conjugated again, focusing it within the scattering medium. These steps are repeated continuously, and the focal point within the medium becomes increasingly stronger and smaller.

[0045] Compared with the prior art, the beneficial effects of the present invention are:

[0046] (1) By setting the first image sensor and the spatial light modulator at a symmetrical position on the reflective surface of the first non-polarized beam splitter cube, setting the second non-polarized beam splitter cube and the pixel matching system to achieve pixel matching between the image generated by the spatial light modulator and the first image sensor, and then quickly focusing through the focusing system, high-quality light focusing through the scattering medium is achieved.

[0047] (2) Based on mechanical calibration, digital compensation is performed using Zernike polynomials to further improve the accuracy of the focus while reducing calibration time.

[0048] (3) The wavefront shaping technique has been extended to the invisible light band and achieved better performance, which has further advanced the application of wavefront shaping in deep tissue bioimaging and also provided some technical support for other long-distance adaptive optics imaging. Attached Figure Description

[0049] Figure 1 This is a schematic diagram of the DOPC focusing system structure of the present invention;

[0050] Figure 2 This is a schematic diagram of the iTRUE focusing system structure of the present invention;

[0051] Figure 3This is a flowchart of the method of the present invention;

[0052] Figure 4 This is a schematic diagram illustrating the improvement of the focusing brightness through the scattering medium during the SLM mechanical calibration process of this invention;

[0053] Figure 5 This is a schematic diagram illustrating the focusing effect through the scattering medium during system calibration of the present invention;

[0054] Figure 6 This is a schematic diagram of the focal point of the present invention, which is focused and iterated within a scattering medium.

[0055] The markings in the diagram are explained below:

[0056] 1-Laser, 2-Pose adjustment device, 31-First unpolarized beam splitter cube, 32-Second unpolarized beam splitter cube, 33-Third unpolarized beam splitter cube, 4-First image sensor, 5-Spatial light modulator, 6-Pixel matching system, 61-First lens, 62-First reflector, 63-Second reflector, 7-Focusing system, 71-Second lens, 72-Fourth unpolarized beam splitter cube, 73-Second image sensor, 8-Ultrasonic focusing system, 81-Ultrasonic transducer, 82-Light blocking structure, 83-Transparent water tank, 84-Semi-transparent reflector, 9-Optical path adjustment structure, 91-Reflector, 92-Half-wave plate, 93-First beam expander structure, 94-Fifth unpolarized beam splitter cube, 95-Acousto-optic modulator, 96-Second beam expander structure, 97-Sixth unpolarized beam splitter cube. Detailed Implementation

[0057] The present invention will be further described below with reference to specific embodiments. The accompanying drawings are for illustrative purposes only, representing schematic diagrams rather than actual physical objects, and should not be construed as limiting the scope of this patent. To better illustrate the embodiments of the present invention, some components in the drawings may be omitted, enlarged, or reduced, and do not represent the actual dimensions of the product. It is understandable to those skilled in the art that some well-known structures and their descriptions may be omitted in the drawings.

[0058] In the accompanying drawings of the embodiments of the present invention, the same or similar reference numerals correspond to the same or similar components. In the description of the present invention, it should be understood that if terms such as "upper," "lower," "left," "right," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, they are only for the convenience of describing the present 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. Therefore, the terms used to describe positional relationships in the drawings are only for illustrative purposes and should not be construed as limiting the present patent. For those skilled in the art, the specific meaning of the above terms can be understood according to the specific circumstances.

[0059] Example 1

[0060] like Figure 1 and Figure 2 As shown, a near-infrared deep tissue focusing system based on rapid mechanical calibration includes a control device, a laser 1, a pose adjustment device 2, a first unpolarized beam splitter cube 31, a second unpolarized beam splitter cube 32, a third unpolarized beam splitter cube 33, a first image sensor 4, and a spatial light modulator 5. The first image sensor 4 and the spatial light modulator 5 are respectively positioned symmetrically on the reflective surface of the first unpolarized beam splitter cube 31. The second unpolarized beam splitter cube 32 is positioned between the first unpolarized beam splitter cube 31 and the spatial light modulator 5. The third unpolarized beam splitter cube 33... A beam splitter cube 33 is disposed between the first unpolarized beam splitter cube 31 and the first image sensor 4. An optical path adjustment structure 9 is provided between the laser 1 and the first unpolarized beam splitter cube 31, the second unpolarized beam splitter cube 32 and the third unpolarized beam splitter cube 33 so that the laser beam passes through the first unpolarized beam splitter cube 31, the second unpolarized beam splitter cube 32 and the third unpolarized beam splitter cube 33 respectively. A spatial light modulator 5 is disposed on the pose adjustment device 2. The control device is electrically connected to the laser 1, the pose adjustment device 2 and the first image sensor 4.

[0061] It also includes a pixel matching system 6, which includes a pair of first lenses 61 and a pair of mirrors 91. The mirrors 91 include a first mirror 62 and a second mirror 63. The first mirror 62 and the first unpolarized beam splitter cube 31 are located at opposite ends of the optical axis of the pair of lenses, and the reflecting surface of the second mirror 63 is aligned with one side of the second unpolarized beam splitter cube 32.

[0062] It also includes a focusing system 7 disposed between the first non-polarizing beam splitter cube 31 and the optical path adjustment structure 9. The focusing system 7 includes a second lens 71, a medium fixing structure, a fourth non-polarizing beam splitter cube 72, and a second image sensor 73 disposed between the first non-polarizing beam splitter cube 31 and the optical path adjustment structure 9. The second lens 71, the medium fixing structure, and the fourth non-polarizing beam splitter cube 72 are arranged in sequence. The optical path adjustment structure 9 is located on the other side of the fourth non-polarizing beam splitter cube 72 relative to the medium fixing structure. The second image sensor 73 and the medium fixing structure are respectively located symmetrically on the reflective surface of the fourth non-polarizing beam splitter cube 72.

[0063] Phase conjugation systems involve numerous components, each requiring high pose precision, a contradictory requirement for general optical systems. Furthermore, various components exhibit uncontrollable aberrations. The fundamental principle of phase conjugation utilizes the time symmetry of the wave equation's wave function solution to achieve a time-reversal-like effect in electromagnetic wave propagation. In a digital optical phase conjugation (DOPC) system composed of an image sensor and a spatial light modulator (SLM), the image sensor and SLM are positioned symmetrically on the non-polarized beam splitter cube (BS) reflector. A control device, such as a computer, calculates the complex amplitude distribution (or phase map) from the speckle interferogram recorded by the image sensor using a four-step phase-shifting holography technique. It then conjugates this phase map and immediately loads the processed phase map onto the SLM, thereby achieving reverse propagation of light and focusing through the scattering medium. However, since the SLM and image sensor in the DOPC system are on opposite sides of the BS and invisible to each other, the first challenge in calibration is pixel matching between the image sensor and the SLM.

[0064] Therefore, to achieve pixel matching between the image sensor and the SLM, the pose adjustment device 2 is used to adjust the SLM so that the pixels of the first image sensor 4 match the bright and dark grid formed after interference by the SLM. In this process, a pixel matching system 6 is incorporated, wherein a pair of first lenses 61 form an optical 4f system, placing both the SLM and the first image sensor 4 on the front focal plane of the first first lens 61, and the first reflecting mirror 62 on the rear focal plane of the second first lens 61. The advantage of the 4f system is that even if the first lens 61 has a certain offset within its plane, the secondary imaging can correct small displacement errors due to the principle of optical path reversibility. Furthermore, the pitch and tilt angles of the reflecting mirror 91 only affect the angle of the image rather than its position, greatly improving the tolerance of optical path adjustment. Thus, the SLM can perfectly image onto the first image sensor 4, achieving better pixel matching. Simultaneously, a second reflecting mirror 63 is set to reflect the reference light that has not been reflected by the SLM back into the optical path, thereby interfering with the reflected light from the spatial modulator.

[0065] After pixel matching is completed, a scattering medium is placed in the medium-fixed structure. At this point, the second image sensor 73 can only obtain a very blurry focus because pixel matching cannot calibrate the pitch and tilt angles of the SLM to place it in an ideal posture. Therefore, while adjusting each of the six degrees of freedom of the SLM one by one using the pose adjustment device 2, the intensity of the focus within the second image sensor 73 is recorded. The absolute coordinates of the electric displacement stage and rotary stage when the highest peak intensity is reached are recorded to obtain the optimal pose of the SLM. The pose adjustment device 2 can be a six-axis motor or other devices that can be adjusted in six degrees of freedom of SML. Since only simple serial port commands need to be sent to the motor without the need for massive complex amplitude and matrix operations, the entire mechanical calibration process generally only takes 10 minutes. Adaptive stepping accuracy scanning with PID control can also be added, and the mechanical calibration can be shortened to 3-5 minutes, which is much less than the time consumed by other DOPC systems in the world for pure digital six-axis pose compensation.

[0066] After the coarse adjustment of pixel matching and the fine adjustment of the six degrees of freedom of the SLM, a very sharp focus has appeared on the second image sensor 73.

[0067] It also includes an ultrasonic focusing system 8, which includes an ultrasonic transducer 81 and a light-blocking structure 82. The ultrasonic transducer 81 is located between the medium fixing structure and the fourth non-polarized beam splitter cube 72. The light-blocking structure 82 can make the fourth non-polarized beam splitter cube 72 and the optical path adjustment structure 9 either allow light to pass through or block light from passing through.

[0068] The frequency of use is f US The self-focusing ultrasonic transducer 81 emits an ultrasonic beam into the scattering medium. The focal point of the emitted ultrasonic beam is located between the fourth unpolarized beam splitter cube 72 and the scattering medium. When the initially incident light passes through the ultrasonic focal point, due to the strong acousto-optic effect, some of the reflected light undergoes a frequency shift. This portion of light has a frequency close to that of the reference light and can therefore be detected by heterodyne interferometry. Subsequently, the phase conjugate system only inverts the reflected light that has passed through the ultrasonic focal point, thus focusing the reflected light into the scattering medium. Furthermore, the DOPC system is easily integrated with various guiding targets to further achieve focusing within the scattering medium. In biological tissues, compared to the strong scattering characteristics of light, ultrasound has excellent penetrability due to its longer wavelength, making it a high-performance guiding target.

[0069] The ultrasonic focusing system 8 also includes a transparent water tank 83, and the fourth unpolarized beam splitter cube 72 is designed inside the transparent water tank 83. The focal point of the ultrasonic transducer 81 is located inside the transparent water tank 83, and a semi-transparent reflector 84 is provided between the focal point of the ultrasonic transducer 81 and the fourth unpolarized beam splitter cube 72.

[0070] Because air has high impedance to ultrasound, the ultrasonic transducer 81 should be immersed in water to better conduct ultrasound waves. To facilitate observation of the focusing process, the TRUE system typically uses a sandwich structure (scattering medium-water-scattering medium) instead of a large scattering medium, allowing light to propagate through a distance of free space. Conventional iTRUE systems add a fluorescent film at the focal beam location to photograph the scattering medium from the side with a camera. This method is extremely difficult to implement in the invisible light band, has very low energy efficiency, and only provides a cross-sectional view of the beam, failing to visually reflect the two-dimensional distribution of the focal point. This invention replaces one of the scattering media with a semi-transparent, semi-reflective mirror, allowing the frequency-shifted light energy to be returned and received by the second image sensor 73, while also enabling excellent observation of the focal point.

[0071] The optical path adjustment structure 9 includes a mirror 91 that allows the laser emitted by the laser 1 to enter the second unpolarized beam splitter cube 32, a half-wave plate 92 located on the laser optical path, and a first beam expander structure 93 located between the half-wave plate 92 and the second unpolarized beam splitter cube 32.

[0072] The optical path adjustment structure 9 also includes a fifth unpolarized beam splitting cube 94 for splitting the laser emitted by the laser 1, an acousto-optic modulator 95 (AOM) and a second beam expander 96 arranged sequentially in the direction of reflected light of the fifth unpolarized beam splitting cube, and an acousto-optic modulator 95 disposed between the third unpolarized beam splitting cube 33 and the half-wave plate 92. The fifth unpolarized beam splitting cube 94 is provided with half-wave plates 92 in both the incident and reflection directions.

[0073] The optical path adjustment structure 9 also includes a sixth unpolarized beam splitting cube 97 for splitting the laser emitted by the laser 1, an acousto-optic modulator 95 disposed between the fourth unpolarized beam splitting cube 72 and the sixth unpolarized beam splitting cube 97, and an acousto-optic modulator 95 disposed between the sixth unpolarized beam splitting cube 97 and the half-wave plate 92. The sixth unpolarized beam splitting cube 97 is provided with half-wave plates 92 in both the incident and reflection directions.

[0074] This scheme arranges the optical path simply and directly using the above structure. The position and number of the reflector 91 and the non-polarizing beam splitter cubes can be changed as needed, allowing the light emitted by the laser 1 to enter the first non-polarizing beam splitter cube 31, the second non-polarizing beam splitter cube 32, and the third non-polarizing beam splitter cube 33 respectively. The first beam expander structure 93 and the second beam expander structure 96 can be beam expanders composed of two lenses, or other structures that can change the beam size; these will not affect the implementation of this scheme.

[0075] Laser 1 is a near-infrared light generator.

[0076] Deep biological tissue focusing and imaging has long been hampered by the massive absorption and excessive scattering of photons by various particles, hindering its application. Wavefront shaping techniques can only overcome the scattering of light by the scattering medium, but cannot eliminate the effects of absorption. Most biological tissues absorb visible light strongly, causing unnecessary energy loss. Furthermore, although phase-conjugated systems possess powerful performance, they are limited by the size of the spatial light modulator (SLM), making them only partially spatially controlled systems. This means they cannot control scattered light outside the optical path. Near-infrared light, with its longer wavelength than visible light and a size comparable to various scatterers within the body, exhibits weaker Mie scattering and can penetrate deeper tissues. Simultaneously, the superior properties of near-infrared light allow it to penetrate thicker muscle tissue and thinner bone, making this technology promising for applications such as subcutaneous laser therapy, non-invasive high-resolution brain microscopy, and optogenetics research. However, in the invisible light band, the optical path construction and error control of phase-conjugated systems are particularly difficult. This invention provides a focusing system with rapid calibration and high focusing accuracy, significantly improving the usability of focusing light through scattering media using infrared light sources.

[0077] Example 2

[0078] like Figures 1 to 3 As shown, a near-infrared deep tissue focusing method based on rapid mechanical calibration includes the following steps:

[0079] S1: The reference light formed by the light emitted by the laser 1 after passing through the optical path adjustment structure 9 causes the reference light to interfere with the reflected light of the spatial light modulator 5, thereby making the image generated by the spatial light modulator 5 have a specific interference pattern.

[0080] The reference light, formed by the acousto-optic modulator 95 (AOM), enters the third unpolarized beam splitter cube 33. The laser 1, after entering the second unpolarized beam splitter cube 32 through the optical path adjustment structure 9, enters the spatial light modulator 5 (SLM) to form the reflected light of the spatial light modulator 5.

[0081] S2: Adjust the pose adjustment device 2 to adjust the Δx, Δy, Δz and Δθ of the spatial light modulator 5. z The four degrees of freedom are coarsely adjusted so that the first image sensor 4 senses the image generated by the spatial light modulator 5, and so that the image generated by the spatial light modulator 5 corresponds to each pixel of the first image sensor 4.

[0082] S3: A scattering medium is placed in a medium-fixed structure so that the reference light and the sample light passing through the scattering medium form heterodyne interference. The control device continuously records the multi-frame interference diagrams identified by the first image sensor 4 and calculates the phase diagram. Then, the calculated phase diagram is loaded onto the spatial light modulator 5 so that the light beam penetrates the scattering medium in reverse, re-forms a parallel light beam that enters the sample, and focuses it onto the second image sensor 73.

[0083] Laser 1 enters the fourth unpolarized beam splitter cube through the optical path adjustment structure and then enters the scattering medium, forming sample light that passes through the scattering medium. The sample light enters from one side of the scattering medium, and the newly formed parallel beam incident on the sample re-enters the scattering medium from the other side. The first sensor records at least two frames of interferograms.

[0084] S4: The control device controls the pose adjustment device 2 to adjust the Δx, Δy, Δz, and Δθ of the spatial light modulator 5. x , Δθ y , Δθ z The six degrees of freedom are adjusted one by one, and at the same time the control device records the peak intensity of the focal point in the image identified by the second image sensor 73 and the corresponding coordinate position and angle of the spatial light modulator 5 to obtain the optimal pose of the spatial light modulator 5.

[0085] Step S3 also includes the following steps:

[0086] Adjust the frequency of the acousto-optic modulator 95 (AOM) to f US And make the beat frequency f of both M The frame rate is set to 10Hz, allowing the sample light to form heterodyne interference with the reference light generated by the AOM. The first image sensor 4 is a scientific-grade CMOS camera (sCMOS), and the second image sensor 73 is an industrial-grade CMOS camera. The sCMOS frame rate is set to 4fps. M After recording four consecutive interferograms, the conjugate of the calculated phase diagram is loaded onto the spatial light modulator 5 to make the beam penetrate the scattering medium in reverse, re-form a parallel beam that enters the sample, and focus it on the CMOS.

[0087] It also includes the following steps:

[0088] S5: Take the multi-order Zernike polynomial, scan the coefficients of each order in order from low to high order, and obtain the optimal coefficients of each order through the focal peak intensity of the second image sensor 73. Then, superimpose the final Zernike compensated phase map on the conjugate phase map in step S3.

[0089] After step S4, a very sharp focus appears on the second image sensor 73, but it still falls short of the DOPC system's performance limits. One reason is that the surface curvature of the SLM is not ideal. Current wafer manufacturing processes still have certain defects, and the backplane of a single LCoS (liquid crystal on silicon) wafer may have some curvature. Although the height difference from the center to the edge of this deformation is generally less than 10 μm, having almost no impact on geometric optics imaging systems, it degrades the modulation wavefront in phase-modulated DOPC systems. Secondly, limited by the manufacturing processes of various optical components in the optical path, such as insufficient flatness of the mirror 91 and the inability to fully correct higher-order aberrations of the lens, slight wavefront aberrations inevitably appear in all beams. These defects are completely unrelated to the SLM's pose and can only be digitally compensated by loading the corresponding phase map onto the SLM.

[0090] For typical DOPC devices, the SLM at the system's end functions as both optical field control and error compensation. Specifically, during system debugging, the accumulated phase distortion of the system needs to be calculated using a specific feedback algorithm. This phase map is then preloaded onto the SLM, and the wavefront of the scattered light from the sample is phase-conjugated to achieve high-quality focusing through the scattering medium. However, this digital compensation method consumes a significant amount of time in the initial calculations, has limited accuracy and a small adjustment range, and cannot compensate for larger errors in invisible light systems.

[0091] Therefore, this invention performs digital compensation based on the mechanical calibration of the focusing system 7. Characterized by Zernike polynomials, these polynomials, through linear superposition, can generate almost arbitrary phase diagrams. Solving for the coefficients of each order of the Zernike polynomials is similar to the 6-DOF correction of the SLM pose. Therefore, this invention uses Zernike polynomials for digital compensation, thereby obtaining a sharper focus. Step S4 of this invention performs fine-tuning by adjusting the SLM displacement. Due to the mechanical calibration, the speed is very fast. Then, in step S5, digital compensation further improves the calibration accuracy. This invention combines mechanical calibration and digital compensation, and its calibration speed and final focusing accuracy are higher than those of typical visible light DOPC systems.

[0092] In step S5, an 8th-order Zernike polynomial is taken; the scanning range of each order coefficient is [-2,2], and the step precision is 0.01.

[0093] The phase map is multiplied by coefficients of different Zernickel phases to detect changes in focus brightness in real time, and the coefficient corresponding to the strongest focus is recorded. Focusing accuracy is significantly improved when performing scans of orders 1 to 3, but higher-order Zernickel aberrations result in greater shape changes and increased scan time. In this scheme, an 8th-order Zernickel polynomial is used.

[0094] It also includes the following steps:

[0095] S6: The ultrasonic transducer 81 emits an ultrasonic beam into the scattering medium, and the incident light in the scattering medium passes through the ultrasonic intersection point, causing a frequency shift in some of the reflected light. Then, the reference light generated by the laser 1 performs heterodyne interference with this part of the reflected light to detect the reflected light. Finally, the control device inverts the reflected light that has passed through the ultrasonic focal point, and can then focus it into the scattering medium.

[0096] It also includes the following steps:

[0097] S7: Iterate the inverted light as the incident light in the scattering medium multiple times;

[0098] In step S6, a regularly focused beam is first incident into the medium. A weak signal light (marker light) is reflected from the point where the ultrasound focal point is located. This reflected light is then inverted using phase conjugation. It is observed that the full width at half maximum (FWHM) of the initially formed focal point is close to the FWHM of the ultrasound focal point. Because the signal-to-noise ratio (SNR) of the first reflected light is very low, the quality of the inverted focal point within the medium is poor. However, this time, more light hits near the ultrasound focal point, so the SNR is significantly improved when it is reflected again. The reflected marker light is then conjugated again, focusing it within the scattering medium. These steps are repeated continuously, and the focal point within the medium becomes increasingly stronger and smaller.

[0099] In this embodiment, after 5 iterations, the focused light energy rapidly converges to form a sharp focal point, whose full width at half maximum (FWHM) is about 1 / 6 of the ultrasonic focal point, which is very close to the theoretical limit.

[0100] Example 3

[0101] One of the important metrics for evaluating the focusing performance of a wavefront shaping system is the PBR (peak-to-background ratio), which is the ratio of the peak intensity of the focal spot to the average intensity of the speckle background outside the full width at half maximum (FWHM).

[0102] like Figure 4 As shown, Figure 4 This is a schematic diagram illustrating the improvement of the focusing brightness through the scattering medium during the SLM mechanical calibration process of this invention.

[0103] like Figure 5 The diagram illustrates the improved focusing effect through the scattering medium during system calibration and the final focusing performance achieved by this invention. Figure 5 (a) This is the speckle pattern formed when uncontrolled (random phase loaded) light emitted from the SLM illuminates a scattering medium (such as a frosted card) with a suitable scattering coefficient. This speckle background serves as a control. Figure 5 (b) This is the focus obtained through the medium using the DOPC system without any calibration; the PBR is only 60. For example... Figure 5(c) After initial pixel matching, the focal point PBR formed by the DOPC through its focusing is approximately 200. However, after completing the six-DOF SLM pose mechanical calibration, such as Figure 5 (d) The focal PBR was improved by several hundred times, reaching approximately 6 × 10⁻⁶. 4 Upon testing, Δθ x and Δθ y The final accuracy is 0.0005°, Δθ z The final accuracy is 0.02°, the final accuracy of Δx and Δy is 4 μm, and the final accuracy of Δz is 50 μm. After compensating for the system wavelet aberration using Zernike polynomials, as follows: Figure 5 (e) The focal PBR has been further improved to 1.1 × 10⁻⁶. 5 It reached 70% of the theoretical limit.

[0104] like Figure 6 (a) Speckle pattern obtained by SLM without phase modulation. Figure 6 (b)-(d) represent the focal points obtained inside the medium during the 1st, 3rd, and 5th iterations, respectively.

[0105] Obviously, the above embodiments of the present invention are merely examples for clearly illustrating the present invention, and are not intended to limit the implementation of the present invention. Those skilled in the art can make other variations or modifications based on the above description. It is neither necessary nor possible to exhaustively describe all embodiments here. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the scope of protection of the claims of the present invention.

Claims

1. A near-infrared deep tissue focusing system based on fast mechanical calibration, characterized in that, The system comprises a control device, a laser (1), a pose adjustment device (2), a first unpolarized beam splitting cube (31), a second unpolarized beam splitting cube (32), a third unpolarized beam splitting cube (33), a first image sensor (4), and a spatial light modulator (5). The first image sensor (4) and the spatial light modulator (5) are respectively arranged at the symmetrical positions of the reflecting surface of the first unpolarized beam splitting cube (31). The second unpolarized beam splitting cube (32) is arranged between the first unpolarized beam splitting cube (31) and the spatial light modulator (5). The third unpolarized beam splitting cube (33) is arranged between the first unpolarized beam splitting cube (31) and the first image sensor (4). The laser (1) is provided with an optical path adjusting structure (9) between the first unpolarized beam splitting cube (31), the second unpolarized beam splitting cube (32), and the third unpolarized beam splitting cube (33), so that the laser beam respectively passes through the first unpolarized beam splitting cube (31), the second unpolarized beam splitting cube (32), and the third unpolarized beam splitting cube (33). The spatial light modulator (5) is arranged on the pose adjustment device (2). The control device is electrically connected with the laser (1), the pose adjustment device (2), and the first image sensor (4). The system further comprises a pixel matching system (6) which comprises a pair of first lenses (61) and a pair of mirrors (91). The pair of mirrors (91) comprises a first mirror (62) and a second mirror (63). The first mirror (62) and the first unpolarized beam splitting cube (31) are respectively located at the two ends of the optical axis of the pair of first lenses. The reflecting surface of the second mirror (63) is aligned with one side of the second unpolarized beam splitting cube (32). The system further comprises a focusing system (7) arranged between the first unpolarized beam splitting cube (31) and the optical path adjusting structure (9). The focusing system (7) comprises a second lens (71) arranged between the first unpolarized beam splitting cube (31) and the optical path adjusting structure (9), a medium fixing structure, a fourth unpolarized beam splitting cube (72), and a second image sensor (73). The second lens (71), the medium fixing structure, and the fourth unpolarized beam splitting cube (72) are arranged in sequence. The optical path adjusting structure (9) is located on the other side of the fourth unpolarized beam splitting cube (72) relative to the medium fixing structure. The second image sensor (73) and the medium fixing structure are respectively located at the symmetrical positions of the reflecting surface of the fourth unpolarized beam splitting cube (72).

2. The fast mechanical calibration based near-infrared deep tissue focusing system of claim 1, wherein, The system further comprises an ultrasonic focusing system (8) which comprises an ultrasonic transducer (81) and a light blocking structure (82). The ultrasonic transducer (81) is arranged between the medium fixing structure and the fourth unpolarized beam splitting cube (72). The light blocking structure (82) can make the fourth unpolarized beam splitting cube (72) and the optical path adjusting structure (9) be in a state of allowing light to pass through or blocking light to pass through.

3. The fast mechanical calibration based near-infrared deep tissue focusing system of claim 2, wherein, The ultrasonic focusing system (8) further comprises a transparent water tank (83), the fourth unpolarized beam splitting cube (72) is designed in the transparent water tank (83), the focal point of the ultrasonic transducer (81) is located in the transparent water tank (83), and a half mirror (84) is arranged between the focal point of the ultrasonic transducer (81) and the fourth unpolarized beam splitting cube (72).

4. The fast mechanical calibration based near-infrared deep tissue focusing system of claim 2, wherein, The light path adjusting structure (9) comprises a mirror (91) for enabling the laser emitted by the laser (1) to enter the second unpolarized beam splitting cube (32), a half wave plate (92) located on the laser light path, and a first beam expanding structure (93) located between the half wave plate (92) and the second unpolarized beam splitting cube (32). The light path adjusting structure (9) further comprises a fifth unpolarized beam splitting cube (94) for splitting the laser emitted by the laser (1), an acousto-optic modulator (95) and a second beam expanding structure (96) arranged in sequence in the reflection direction of the fifth unpolarized beam splitting cube (94), and an acousto-optic modulator (95) arranged between the third unpolarized beam splitting cube (33) and the half wave plate (92), and half wave plates (92) are arranged in the incident and reflection directions of the fifth unpolarized beam splitting cube (94). The light path adjusting structure (9) further comprises a sixth unpolarized beam splitting cube (97) for splitting the laser emitted by the laser (1), an acousto-optic modulator (95) arranged between the fourth unpolarized beam splitting cube (72) and the sixth unpolarized beam splitting cube (97), and an acousto-optic modulator (95) arranged between the sixth unpolarized beam splitting cube (97) and the half wave plate (92), and half wave plates (92) are arranged in the incident and reflection directions of the sixth unpolarized beam splitting cube (97).

5. A near-infrared deep tissue focusing method based on fast mechanical calibration, characterized in that, The near-infrared deep tissue focusing system based on fast mechanical calibration according to any one of claims 2-4, comprising the following steps: S1: the light emitted by the laser (1) forms a reference light after passing through the light path adjusting structure (9), and the reference light and the reflected light of the spatial light modulator (5) form interference, so that the image generated by the spatial light modulator (5) has a specific interference pattern; S2: Adjust the pose adjustment device (2) to adjust the spatial light modulator (5). , , and The four degrees of freedom are coarsely adjusted so that the first image sensor (4) senses the image generated by the spatial light modulator (5) and the image generated by the spatial light modulator (5) corresponds to each pixel of the first image sensor (4); S3: the scattering medium is placed in the medium fixing structure, so that the reference light and the sample light passing through the scattering medium form heterodyne interference, the control device continuously records multiple frames of interference patterns recognized by the first image sensor (4) and calculates a phase pattern, and then loads the calculated phase pattern on the spatial light modulator (5) to make the light beam penetrate the scattering medium in reverse, re-form a parallel light beam incident on the sample, and focus on the second image sensor (73); S4: The control device controls the pose adjustment device (2) to adjust the spatial light modulator (5). , , , , , The six degrees of freedom are adjusted one by one, and the control device records the peak intensity of the focal point in the image identified by the second image sensor (73) and the corresponding coordinate position and angle of the spatial light modulator (5) to obtain the optimal pose of the spatial light modulator (5).

6. The method of claim 5, wherein the near-infrared deep tissue focusing method is based on fast mechanical calibration. The step S3 further comprises the following steps: The frequency of the acousto-optic modulator (95) is adjusted to be f US and the beat frequency of the sample light and the reference light is f M 10 Hz, so that the sample light and the reference light generated by the acousto-optic modulator form heterodyne interference, the first image sensor (4) is a scientific CMOS camera, the second image sensor (73) is an industrial CMOS camera, and the frame rate of the scientific CMOS camera is set to 4 f M After 4 frames of interference patterns are recorded in succession, the conjugate of the calculated phase pattern is loaded on the spatial light modulator (5) to make the light beam penetrate the scattering medium in the reverse direction, re-form a parallel light beam into the sample, and focus on the industrial CMOS camera.

7. The method of claim 5, wherein the near-infrared deep tissue focusing method is based on fast mechanical calibration. Further comprising the following steps: S5: taking a plurality of Zernike polynomials, scanning each order coefficient in order from low order to high order, and feeding back the focal peak intensity of the second image sensor (73) to obtain optimal coefficients of each order, and superimposing the final Zernike compensation phase pattern on the phase pattern in step S3.

8. The method of claim 7, wherein the near-infrared deep tissue focusing method is based on fast mechanical calibration. The step S5 further comprises the following steps: In the step S5, the 8th Zernike polynomial is taken; the scanning range of each order coefficient is [-2, 2], and the step precision is 0.

01.

9. The fast mechanical calibration based near-infrared deep tissue focusing method according to any one of claims 6 to 8, characterized in that, Further comprising the following steps: S6: The ultrasonic transducer (81) emits an ultrasonic beam inside the scattering medium, and makes the incident light of the scattering medium pass through the ultrasonic intersection point to make part of the reflected light frequency shift, then the reference light generated by the laser (1) and the part of the reflected light perform heterodyne interference to detect the part of the reflected light, finally the control device inverses the reflected light passing through the ultrasonic focus point, i.e. focuses on the inside of the scattering medium.

10. The method of claim 9, wherein the near-infrared deep tissue focusing method is based on fast mechanical calibration. Further comprising the following steps: S7: The inverted light is iterated for multiple times as the incident light of the scattering medium.

Images