Deep near-infrared structure imaging method and system based on composite single-pixel scattering compensation

Abstract

The invention relates to a deep near-infrared structure imaging method and system based on composite single-pixel scattering compensation, and belongs to the technical field of brain function imaging. Aiming at the problems of insufficient penetration depth (1.5-2cm), low spatial resolution and incapability of distinguishing deep neural activities in the existing near-infrared spectral imaging technology, binary spatial-temporal modulation of an incident light field is realized by adopting a combination of an optical fiber bundle and an optical switch, scattered signals are collected through multi-probe aperiodic arrangement, and the near-infrared spectral imaging technology is applied to the near-infrared spectral imaging technology. The method combines composite correlation calculation and a time reversal algorithm to invert tissue three-dimensional scattering characteristics, regulates and controls a light field to focus to a deep target area, synchronously analyzes HbO2 / HbR concentration change, reconstructs a structure image, realizes deep brain function and structure imaging with multiplied penetration depth and millimeter or even below resolution, and has the advantages of high precision and low cost.

Description

Technical Field

[0001] The present invention belongs to the technical field of brain function imaging and relates to a deep near-infrared structure imaging method and system based on composite single-pixel scattering compensation. Background Art

[0002] The brain is the most complex and important organ in the human body, controlling all life activities, including language, movement, hearing, vision, and emotional expression. Functional brain imaging, a key technology for studying brain activity, can monitor patterns of brain activity during specific tasks, revealing the relationship between brain function and behavioral and cognitive processes. With the advancement of neuroscience and medicine, functional brain imaging has played a vital role in studying the functional mechanisms of the healthy brain and those in neuropsychiatric disorders.

[0003] Currently, functional brain imaging techniques primarily include positron emission tomography (PET), electroencephalography (EEG), functional magnetic resonance imaging (fMRI), and functional near-infrared spectroscopy (fNIRS). PET produces brain images by detecting injected or inhaled radioactive substances, with a spatial resolution of approximately 3-4 mm and the ability to penetrate the entire head. However, the equipment is expensive and uses radioactive materials, which can be harmful to the human body. EEG monitors brain activity by recording spontaneous, rhythmic motor potentials from neurons beneath the scalp surface. It has high temporal resolution and the detection device is noninvasive and portable, but its spatial resolution is relatively low (5-9 cm). fMRI stimulates specific senses to induce neural activity in corresponding areas of the cerebral cortex, which is then displayed through magnetic resonance imaging. It uses no radioactive materials and is a noninvasive technique for detecting dynamic brain function. It has high spatial resolution (≥0.5 mm voxel) and can penetrate the entire head. However, it requires exposure to strong magnetic and electric fields, making it unsuitable for people with implanted electronic devices. Furthermore, the detector requires the patient to lie still and quiet, and the equipment is expensive and the imaging speed is slow. fNIRS utilizes the properties of near-infrared light (NIR) to indirectly reflect the functional state of the brain by monitoring changes in the concentrations of oxyhemoglobin (HbO2) and deoxyhemoglobin (HbR). It is a non-invasive, low-cost functional imaging technology that has higher motion robustness, spatial positioning accuracy, and anti-electromagnetic interference capabilities than EEG, as well as higher temporal resolution and lower cost than fMRI. It is suitable for a variety of people, has strong mobility, and can adapt to more scientific research scenarios.

[0004] Under normal circumstances, HbR and HbO2 concentrations remain largely constant due to cerebral vascular autoregulation mechanisms. However, during brain activity, these concentrations change accordingly. Therefore, measuring changes in HbO2 and HbR concentrations can indirectly reflect brain function. The fNIRS measurement process involves a light source placed vertically on the scalp emitting near-infrared light that passes through brain tissue. After absorption and scattering by the tissue, the light is emitted and received by a photodetector placed vertically on the scalp. Light from the source enters the tissue, where it is absorbed and scattered before reaching the detector, following a modified Beer-Lambert law. Given the intensities of the incident and outgoing light, the distribution of optical parameters (absorption coefficient and scattering coefficient) within the tissue can be determined by Monte Carlo simulation and solving the radiation transfer equation, allowing the relative concentrations of HbO2 and HbR to be calculated. fNIRS has been used to study brain function for over three decades, with significant advances in its techniques and applications, and numerous emerging applications for exploring neural mechanisms are emerging. While near-infrared spectroscopy offers a relatively inexpensive and portable method for brain scanning, it is limited in its penetration depth. Furthermore, NIRS cannot provide accurate spatial information; the optical path and light penetration depth can only be estimated.

[0005] The research background of fNIRS can be traced back to 1977, when Frans Jobsis first used near-infrared spectroscopy to detect changes in blood volume in brain tissue, ushering in a new era of NIRS functional brain imaging. Early fNIRS systems were single-channel measurements using a source-detector pair, with extremely low temporal resolution and sensitivity. In 1994, a Japanese team developed the first multi-channel near-infrared functional brain imaging system, significantly improving resolution. However, the spatial resolution of traditional fNIRS remains lower than that of fMRI and is limited by surface tissue signal contamination and a lack of anatomical data. To improve resolution, researchers have continuously optimized the number of channels and source-detector arrangements. fNIRS has evolved from rudimentary topological imaging to diffuse optical tomography (DOT) and high-density diffuse optical tomography (HD-DOT). HD-DOT has improved the application prospects of fNIRS by increasing spatial resolution and reducing surface tissue influences, but its equipment is complex and lacks portability.

[0006] Although fNIRS has limitations in terms of spatial resolution and penetration depth, its non-invasive, portable, and low-cost characteristics give it unique advantages in neuroscience research. In recent years, the multimodal combination of fNIRS with other imaging technologies (such as EEG, ultrasound, MRI) has further improved the imaging quality and provided a new perspective for brain function research. For example, using a high-density diffuse optical sensing array combined with optical tomography reconstruction to improve spatial resolution; combining the use of registered ultrasound (US) to guide light illumination, reception, and image reconstruction, which overcomes the problem of poor light localization of lesions caused by light scattering and improves the reconstruction accuracy of lesions; combining digital breast tomosynthesis (DBT), DOT can provide functional information superimposed on mammogram images, improving the sensitivity and specificity for cancer pathology. The NIRS-EEG combined detection system overcomes the limitations of single imaging technologies by synchronously analyzing hemodynamic changes and electrophysiological signals, providing a new method for clinical brain function detection.

[0007] Since the birth and development of the fNIRS system to date, significant technological progress has been made in related fields. For diverse measurement environments, a wide variety of high-performance fNIRS instruments and advanced technologies have been developed. These innovative achievements have greatly improved the spatial resolution of the system and optimized the imaging quality. However, existing fNIRS also faces many challenges. With the increase in the number of channels, the structure of the device becomes increasingly complex and cumbersome, seriously affecting its portability; relatively simple fNIRS devices have relatively limited functions and can only perform functional imaging of the cortical surface layer, making it difficult to deeply detect deep brain structures under the cortex (such as the striatum and thalamus, etc.), and it is also unable to effectively image deep cortical structures (such as the insula and operculum, etc.). From the imaging principle, the current fNIRS system mainly indirectly reflects the brain's functional state by monitoring the change in hemoglobin concentration, but this method cannot achieve structural imaging, lacks longitudinal resolution, and has a relatively low lateral resolution, only reaching the centimeter level. Summary of the Invention

[0008] In view of this, the purpose of the present invention is to provide a deep near-infrared structural imaging method based on composite single-pixel scattering compensation.

[0009] Due to the scattering and absorption of the scalp and skull, etc., the imaging depth of fNIRS can only reach 1.5 - 2 cm under the scalp (i.e., the surface of the cerebral cortex), and it is impossible to measure the hemodynamic response from deep brain structures through a simple fNIRS device. The change in hemoglobin concentration detected by fNIRS is the result of the combined contribution of all brain tissues through which the light passes between the source and the detector, and it is impossible to distinguish the specific locations where neural activities occur in the brain tissues along the light path. The spatial resolution of fNIRS can only reach the centimeter level.

[0010] In the present invention, time, space, and spatio-temporal coupling modulation of incident light are performed based on a wavefront modulation method. Multiple detectors arranged in a specific distribution function receive the time and space responses of the corresponding output field. An incident spatio-temporal modulation-output spatio-temporal response relationship is established, and the two-dimensional or three-dimensional characteristics of brain tissue are calculated according to the composite correlation imaging calculation method. The incident light field is cyclically partitioned and regulated to amplify the output signal, thereby optimizing the scattering characteristics. A specific input state is calculated by time reversal to control the focusing of light in different regions within the tissue. This process compensates for the influence of multiple scattering within the tissue, can increase the penetration depth of light within the tissue, and obtain deeper neural activity responses. In a specific real-time process, multiple specially arranged output probes are used to receive the output responses, determine the propagation path of light within the tissue, and then focus in a specified area according to the composite correlation calculation or scattering matrix method, making the position where the measured hemoglobin concentration changes more accurate, thereby improving the functional imaging resolution. In addition, by calculating the scattering characteristics of the tissue, brain tissue structure imaging can be achieved simultaneously.

[0011] To achieve the above object, the present invention provides the following technical solutions:

[0012] A deep near-infrared structure imaging method based on composite single-pixel scattering compensation, comprising the following steps:

[0013] S1: Through the combination of an optical fiber bundle and an optical switch, binary spatio-temporal coupling modulation is performed on the incident light field to generate a dynamic two-dimensional spatial light field distribution;

[0014] S2: The modulated incident light field is transmitted to the target tissue through an input probe, and multiple receiving probes arranged according to a specific distribution function are used to collect the output spatio-temporal response signals scattered by the tissue;

[0015] S3: Based on the composite correlation calculation of the output spatio-temporal response signal and the incident spatio-temporal modulation signal, the three-dimensional scattering characteristics of the target tissue are determined;

[0016] S4: According to the three-dimensional scattering characteristics, specific input light field regulation parameters are generated by time reversal calculation, so that the incident light is focused on a preset area inside the target tissue;

[0017] S5: The optical fiber bundle is divided into multiple partitions, and each partition of the optical fiber bundle is successively regulated to achieve coherent superposition of the output signals and iteratively optimize the scattering characteristics;

[0018] S6: Based on the focused light field response, the change amounts of HbO2 and HbR concentrations in the target area are calculated to achieve deep brain functional imaging and simultaneously reconstruct the brain tissue structure image.

[0019] Further, in S1, the fiber optic bundle is an array of independent fiber optic lines arranged in an n×n pattern. Each fiber optic line is independently controlled to be in an "on" or "off" state by an optical switch, and the binary spatio-temporal coupling modulation includes any of the following methods:

[0020] (a) Dynamically switching the state of the optical switch according to a Hadamard matrix, which belongs to spatial modulation;

[0021] (b) Setting multiple fiber optic lines to the same switch state to change the modulation degrees of freedom of the optical field, which belongs to spatial modulation;

[0022] (c) Controlling the frequency of the optical switch being "on" and "off", which belongs to temporal modulation.

[0023] Further, in S3, the composite correlation calculation includes:

[0024] Constructing a multi-detector response matrix based on the distance function between receiving probes, the spatio-temporal correlation of the incident optical field modulation sequence and the output response;

[0025] Generating specific input optical field regulation parameters through a scattering matrix inversion algorithm to obtain the three-dimensional scattering characteristics of the tissue.

[0026] Further, the scattering matrix inversion algorithm obtains the three-dimensional scattering characteristics, regulates the input optical field in a zoned distribution, and successively updates the scattering characteristic parameters of each zone to iteratively optimize the three-dimensional scattering characteristics.

[0027] Further, in S4, the time reversal calculation includes:

[0028] Generating a conjugate optical field distribution according to the three-dimensional scattering characteristics, and dynamically adjusting the switch state of the fiber optic bundle through an optical switch to form a focused light spot of the incident optical field within the target tissue.

[0029] Further, the arrangement of the input probe and the receiving probe satisfies the following conditions:

[0030] According to the spatio-temporal modulation characteristics, the receiving probe and the input probe are arranged according to a specific distribution function.

[0031] Further, the method further includes:

[0032] Replacing the combination of the fiber optic bundle and the optical switch with a spatial light modulator or a digital micromirror array to perform binary modulation on the phase or amplitude of the incident optical field.

[0033] A deep near-infrared structured imaging system based on composite single-pixel scattering compensation, comprising:

[0034] A light source module for emitting near-infrared light;

[0035] Optical fiber bundle module, including optical fiber lines arranged in an n×n array, each optical fiber line being connected to an independent optical switch;

[0036] Control module, used to regulate the state of the optical switch to generate a binary spatio-temporal coupled modulation optical field;

[0037] Input probe and receiving probe array, used to transmit the modulated optical field to the target tissue and collect scattered response signals;

[0038] Calculation module, used to perform composite correlation calculation, time reversal calculation and iterative optimization of scattering characteristics, and output functional imaging and structural imaging results.

[0039] Furthermore, the optical fiber of the optical fiber bundle module is coated with a biocompatible protective sleeve, and the output end of the optical fiber bundle is integrated in a wearable head-mounted device.

[0040] Furthermore, the basis for the control module to use binary modulation is:

[0041] Adopt binary modulation according to Shannon's theorem to optimize the coding rate, so as to reduce noise interference and improve the focusing efficiency.

[0042] The beneficial effects of the present invention are as follows:

[0043] (1) By means of binary spatio-temporal modulation and scattering compensation algorithm, the influence of tissue multiple scattering is overcome, the penetration depth of near-infrared light is doubled, and the neural activities of deep subcortical structures (such as striatum, thalamus) can be detected.

[0044] (2) Based on composite correlation calculation and time reversal focusing technology, functional imaging with a spatial resolution of millimeter level or below is realized, and fine structure images such as blood vessels and neurons are reconstructed synchronously, breaking through the limitation of the centimeter-level resolution of traditional fNIRS.

[0045] (3) Synchronously obtain brain function (HbO2 / HbR concentration change) and structure (three-dimensional scattering characteristics) information in a single system, make up for the deficiency of the existing technology that can only perform functional imaging, and provide more comprehensive data support for neural mechanism research.

[0046] (4) Use a combination of optical fiber bundle and optical switch to replace the high-cost spatial light modulator, simplify the system structure, improve the portability, and at the same time reduce noise interference through binary coding to ensure the balance of imaging quality and efficiency.

[0047] (5) Compatible with wearable design, suitable for brain function monitoring of people in a moving state or with implanted electronic devices, and provide a high-precision and low-cost imaging solution for the diagnosis of neurological diseases and the development of brain-computer interfaces.

[0048] Other advantages, objects, and features of the present invention will be set forth in part in the following description, and in part will be obvious to those skilled in the art upon examination of the following, or may be learned by practice of the present invention. The objects and other advantages of the present invention may be realized and attained by the means of the instrumentalities and combinations particularly pointed out hereinafter. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] In order to make the objectives, technical solutions, and advantages of the present invention clearer, the present invention will be described in detail below with reference to the accompanying drawings, where:

[0050] Figure 1 Schematic diagram of the device of the present invention;

[0051] Figure 2 Schematic diagram of calculating scattering characteristics by compound correlation calculation;

[0052] Figure 3 Schematic diagram of successively partitioning and regulating an optical fiber bundle and iteratively updating scattering characteristics to enhance the output response;

[0053] Figure 4 Flowchart of the present invention;

[0054] Figure 5 Schematic diagram of the input regulated by a spatial light modulator;

[0055] Figure 6 Schematic diagram of the path of light in tissue;

[0056] Figure 7 Experimental comparison results of binary modulation and continuous modulation for scattering focusing; (a) shows that binary phase modulation realizes spatial modulation of the incident light field to overcome noise and achieve focusing; (b) shows that continuous phase modulation realizes spatial modulation of the incident light field to overcome noise and achieve focusing; (c) shows the comparison of the focusing efficiencies of binary modulation and continuous modulation.

[0057] Reference numerals: outer cortex of the optical fiber 1; optical switch 2; state of the incident light 3; input probe 4; receiving probe 5; focal point 6; brain tissue 7; human scalp 8; optical fiber line 9; light source 11; spatial light modulator 22. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0058] The following describes the embodiments of the present invention through specific examples. Those skilled in the art can easily understand the other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments. Various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that the drawings provided in the following embodiments only illustrate the basic concept of the present invention in a schematic manner. Without conflict, the following embodiments and the features in the embodiments can be combined with each other.

[0059] Among them, the drawings are only for illustrative purposes, showing only schematic diagrams, not physical diagrams, and should not be construed as a limitation to the present invention; in order to better illustrate the embodiments of the present invention, some components in the drawings will be omitted, enlarged or reduced, which does not represent the size of the actual product; for those skilled in the art, it is understandable that some well-known structures and their descriptions in the drawings may be omitted.

[0060] In the 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 there are terms such as "upper", "lower", "left", "right", "front", "rear", etc. indicating the orientation or positional relationship, they are based on the orientation or positional relationship shown in the drawings. It is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation. Therefore, the terms describing the positional relationship in the drawings are only for illustrative purposes and should not be construed as a limitation to the present invention. For those of ordinary skill in the art, the specific meanings of the above terms can be understood according to specific circumstances.

[0061] By modulating the incident light field and measuring the input and output responses to obtain the scattering characteristics of the scattering medium is the key to overcoming scattering and achieving high-resolution deep imaging. Existing wavefront modulation methods usually use high-resolution spatial light modulators and four-step phase shift methods to achieve continuous measurement (0 - 2π) of the scattering matrix. Through previous experiments, it has been proved that using binary modulation to spatially modulate the incident light field has the advantages of fast modulation rate and high efficiency. Based on binary spatial modulation, the present invention proposes a fiber bundle controlled by optical switches, which can achieve binary spatial, temporal, and spatio-temporal compound modulation of the incident light field. A fiber bundle arranged in an n×n array is used as the light input, and each fiber is connected to an optical switch. By the optical switch, the "on" or "off" of any fiber can be realized, and thus binary spatial control of the input light field with the same effect as that of a digital micromirror array can be achieved, as Figure 1 shown. By controlling the optical switch, the degree of freedom and the spatial modulation sequence of the spatial modulation can be changed. In addition, by controlling the frequency of "on" and "off", temporal control of the incident light field can be achieved.

[0062] Through the design of the fiber optic bundle for modulating the incident field, controlling the frequency and quantity of "on" and "off", spatio-temporal coupling modulation of the incident light field is realized. Multiple detectors arranged according to a specific distribution function can receive the corresponding output response information after the modulated incident light is reflected by various levels of the brain. The signal detected by each detector corresponds to both the spatial signal from all incident lights and the temporal signal of the response depth cross-section. Thus, the corresponding relationship between incident spatio-temporal coupling modulation - output detection spatio-temporal response is established. For example, through the response of a detector at one position to all incident signals, only the scattering characteristics of the corresponding cross-section can be obtained by performing single-pixel correlation calculation. By combining the response information of multiple detectors, the scattering characteristics of multiple response cross-sections can be obtained. In actual operation, the arrangement of the detectors can be non-periodically arranged as in Figure 1 non-periodically, or periodically arranged, or arranged according to other distribution functions, and the distance function D(x, y) between each detector is established. Combining the distance function between each detector, the temporal response of the detector to the input, and the spatial response of the detector to the input, multi-detector composite single-pixel correlation calculation can be performed to calculate the three-dimensional scattering characteristics of the tissue. Through time-reversal calculation, a specific input can be determined to make the light focus on different regions inside the tissue, such as Figure 2 shown.

[0063] Due to the strong scattering characteristics of brain tissue, the intensity of the reflected light signal received by the detector decays exponentially with depth, resulting in the reflected signal of the shallow tissue being dominant and the response of the deep tissue being lost. This scheme proposes a method for amplifying the output signal by controlling the fiber optic bundle switch based on partition control of the input light field. This method divides the fiber optic bundle into multiple partitions, first separately controls the "on" and "off" states of the fiber optic lines in the first partition, calculates the scattering characteristics corresponding to the first partition, and determines the optimal light field control parameters based on the conjugate inversion algorithm and keeps them unchanged. Then, successively control the fiber optic lines in the remaining partitions to make their output light fields coherently superpose with the first partition, thereby performing iterative calculation on the scattering characteristics of the brain tissue to improve the accuracy of the output spatio-temporal response information, such as Figure 3 shown.

[0064] From the perspective of structural composition, the fiber optic bundle system consists of the following parts:

[0065] (1) Fiber optic bundle design: It consists of n×n = N independent fiber optic lines. The outer layer of the fiber is coated with a protective sheath and integrated into a single fiber optic bundle;

[0066] (2) Design for connection with the optical switch: The optical switch is connected to the fiber optic lines one by one, and can independently control the "on" (on) and "off" (off) states of each fiber optic line;

[0067] (3) Input - Output Connection: One end of the fiber optic bundle is coupled to the optical switch to regulate the state of the input light, and the other end transmits the optical signal to the target tissue (such as brain tissue) through the input probe.

[0068] The method of modulating the incident light using the fiber optic bundle is as follows: By controlling the optical switch, different two - dimensional spatial pattern distributions can be formed at the output end of the fiber optic bundle. Among them, the 'on' state corresponds to '1' of the binary spatial pattern, and the 'off' state corresponds to '0', thus realizing spatial incident modulation. Specifically, the switch state can be dynamically changed according to a specific mathematical sequence (such as the Hadamard matrix), so that the two - dimensional spatial light field distribution changes regularly with time; also, by setting multiple fiber lines to the same switch state, the degree of freedom of light field modulation can be changed, thereby realizing spatial modulation with different precisions.

[0069] The fiber optic bundle and the optical switch have the advantages of low cost, and the design of the fiber optic bundle is simple and portable, which is more practical for brain tissue imaging. In addition, the incident light can also be spatially modulated in other ways, such as DMD modulating the incident light.

[0070] Focus the light on the target position. The change in hemoglobin concentration calculated according to the modified Beer - Lambert law (the input - output intensity correspondence) can more accurately reflect the physiological changes in the specific neural activity area of the light in the tissue, improve the resolution of functional imaging, obtain a finer brain function image, and at the same time compensate for the multiple scattering effects in the tissue, and obtain the blood oxygen change conditions of deeper tissues.

[0071] The two - dimensional or three - dimensional characteristics of the tissue are obtained through composite correlation calculation based on the input - output relationship. By inversely calculating to determine a specific input, not only can the aggregation of light in the tissue be controlled, but also direct imaging of the brain tissue structure can be achieved, realizing the visualization of the fine structure of the brain tissue, such as blood vessels, neurons, etc.

[0072] The spatial light modulator (DMD, SLM) contains many independent units, and each unit can independently receive the control of optical signals or electrical signals to perform high - speed spatial modulation on the light wave illuminating it. As Figure 5 Shown is a schematic diagram of modulating the incident light state using a spatial light modulator, and the optical path ignores processes such as collimation and beam expansion. The light emitted by the light source is collimated and expanded and then modulated by the spatial light modulator (modulating the phase or amplitude of the incident light field, similar to the principle of modulating the fiber optic bundle with an optical switch, each micromirror unit of the spatial light modulator has two states of 'on' and 'off' to control the opening and closing of the light beam), and then reflected. The diameter of the reflected light beam is reduced and coupled with a single - core fiber, and the light is transmitted through the fiber. The fiber is coupled with the light source probe and transmitted into the tissue.

[0073] In addition, laser input can be coupled with single-mode optical fiber to transmit light, LED can be used as input and coupled with multi-mode optical fiber to transmit light, and an optical module can be used to control the optical fiber bundle to transmit light for spatial modulation of the incident light field. The basic principle is to modulate the spatial switching state of the input light beam to perform spatial modulation on the incident light. Due to the influence of scattering and absorption, the propagation path of light changes randomly, but there is an average migration path for a large number of photons, and this path is curved like a banana shape, such as Figure 6 Path A in it. Using scattering imaging technology, the switching state of the input light is controlled to compensate for part of the scattering effect when light propagates in the tissue, focus the light inside the tissue, and increase the penetration depth of light in the tissue, such as Figure 6 Path B in it.

[0074] By modulating the incident light field and measuring the response of the input and output, the scattering characteristics of the scattering medium are obtained, which is the key to overcoming scattering and achieving high-resolution deep imaging. Existing wavefront modulation methods usually use high-resolution spatial light modulators and four-step phase shift and other methods to achieve continuous measurement (0-2π) of the scattering matrix.

[0075] According to Shannon's theorem, in a noisy system, the channel capacity C defines the upper limit of the information transmission rate that can be achieved under any low bit error rate condition. When the actual transmission rate R of the system is lower than the channel capacity C, there is a coding method that can transmit information with an arbitrarily small error rate. Applying this theory to the optical scattering system, it is inferred that: the lower the transmission rate, the more accurate the information transmission, and the binary coding method can achieve the lowest transmission rate. As Figure 7 shown, when the binary modulation method and the continuous modulation method are used to achieve focusing under the same noise conditions, the binary spatial modulation method shows significant advantages in both the modulation rate and overcoming the scattering effect, and the above inference is experimentally proven. Figure 7 In it, (a) is the binary phase modulation to achieve spatial modulation of the incident light field and overcome noise to achieve focusing. (b) is the continuous phase modulation to achieve spatial modulation of the incident light field and overcome noise to achieve focusing. (c) is the comparison of the focusing efficiency between binary modulation and continuous modulation.

[0076] Embodiment 1: Deep brain function imaging based on Hadamard matrix modulation

[0077] System configuration: A 16×16 fiber bundle (256 independent fiber lines) is adopted, each fiber line is connected to a high-speed optical switch, and the receiving probes are arranged in a Gaussian distribution and integrated into a wearable head-mounted device.

[0078] Optical field modulation: Control the optical switch to dynamically switch the "on / off" state of the optical fiber according to the Hadamard matrix sequence to generate 256 binary spatial light field distributions, and the modulation frequency is 1 kHz.

[0079] Signal acquisition: Near-infrared light (wavelength 690nm / 830nm) is incident through the scalp, and 16 receiving probes simultaneously collect the scattered light intensity to record the spatio-temporal response signal with a time resolution of 10 ms.

[0080] Scattering characteristic calculation: A 256×16-dimensional response matrix is constructed, and the three-dimensional scattering coefficient of the brain tissue is calculated through phase conjugate inversion, and the optical transmission path model is iteratively corrected in partitions.

[0081] Focusing and imaging: Based on the scattering characteristics, the conjugate light field parameters are generated to regulate the output of the fiber optic bundle, so that the light is focused inside the human brain tissue, the concentration changes of HbO2 / HbR are analyzed (accuracy ±0.1 μM), and the vascular distribution image is reconstructed synchronously.

[0082] Example 2: Subcortical structure imaging with partitioned and stepwise optimization

[0083] Partition design: A 32×32 fiber optic bundle is divided into 8 rectangular partitions, and each rectangular partition contains 128 fiber optic lines.

[0084] Sequential regulation: The fiber optic lines in each partition are activated in sequence. In the first round, the first partition is regulated, and after collecting the signals, the scattering characteristics of this area are calculated, and the optimal optical switch combination is fixed; subsequent partitions are regulated by superimposition in sequence, and the signal-to-noise ratio of deep signals is improved by 30 dB through the coherent enhancement algorithm.

[0085] Iterative inversion: After calculating the scattering parameters of each partition, iterative inversion is performed until the response of the deep brain tissue area is clear and stable.

[0086] Dynamic monitoring: The concentration change curve of HbO2 / HbR is displayed in real time (time resolution 50 ms), and the three-dimensional tissue image is output synchronously.

[0087] Example 3: Multimodal imaging assisted by a spatial light modulator

[0088] System alternative: Replace the fiber optic bundle with a digital micromirror device (DMD, 1024×768 pixels), and combine binary phase modulation (0-2π) to generate a high-precision light field.

[0089] Modulation and acquisition: The DMD switches the light field mode at a frequency of 22 kHz, and 64 receiving probes record the spatio-temporal response of the scattered light, and EEG electrodes are synchronously connected to obtain EEG signals.

[0090] Multimodal fusion: The inversion results of the light scattering characteristics are fused with the EEG time-frequency characteristics to construct a functional-electrophysiological joint atlas to locate the activated area of the motor cortex.

[0091] Clinical application: It can be used for preoperative target verification of deep brain stimulation in Parkinson's disease patients.

[0092] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the present technical solution, and they should all be covered within the scope of the claims of the present invention.

Claims

1. A deep near-infrared structural imaging method based on composite single-pixel scattering compensation, characterized in that: It includes the following steps: S1: Through the combination of an optical fiber bundle and an optical switch, perform binary spatio-temporal coupling modulation on the incident light field to generate a dynamic two-dimensional spatial light field distribution; S2: Transmit the modulated incident light field to the target tissue through an input probe, and use receiving probes arranged with multiple specific distribution functions to collect the output spatio-temporal response signal scattered by the tissue; S3: Based on the composite correlation calculation of the output spatio-temporal response signal and the incident spatio-temporal modulation signal, determine the three-dimensional scattering characteristics of the target tissue; S4: According to the three-dimensional scattering characteristics, generate specific input light field control parameters through time reversal calculation, so that the incident light focuses on a preset area inside the target tissue; S5: Divide the optical fiber bundle into multiple partitions, and successively control each partition of the optical fiber bundle to achieve coherent superposition of the output signals and iteratively optimize the scattering characteristics; S6: Based on the focused light field response, calculate the change in the concentrations of HbO2 and HbR in the target area, achieve deep brain functional imaging, and synchronously reconstruct the brain tissue structure image.

2. The deep near-infrared structured imaging method based on compound single-pixel scattering compensation according to claim 1, wherein: In the above S1, the optical fiber bundle is an independent optical fiber line arranged in an n×n pattern, and each optical fiber line is independently controlled to be in an "on" or "off" state by an optical switch. And the binary spatio-temporal coupling modulation includes any of the following methods: (a) Dynamically switch the state of the optical switch according to the Hadamard matrix, which belongs to spatial modulation; (b) Set multiple optical fiber lines to the same switch state to change the modulation degree of freedom of the light field, which belongs to spatial modulation; (c) Control the "on" and "off" frequencies of the optical switch, which belongs to time modulation.

3. The deep near-infrared structured imaging method based on compound single-pixel scattering compensation according to claim 1, wherein: In the above S3, the composite correlation calculation includes: Construct a multi-detector response matrix according to the distance function between the receiving probes, the time and spatial correlation of the incident light field modulation sequence and the output response; Generate specific input light field control parameters through the scattering matrix inversion algorithm to obtain the three-dimensional scattering characteristics of the tissue.

4. The deep near-infrared structured imaging method based on composite single-pixel scattering compensation according to claim 3, wherein: The three-dimensional scattering characteristics are obtained by the scattering matrix inversion algorithm, and the input light field is controlled in a partitioned distribution manner, and the scattering characteristic parameters of each partition are updated successively to iteratively optimize the three-dimensional scattering characteristics.

5. The deep near-infrared structured imaging method based on composite single-pixel scattering compensation according to claim 1, characterized in that: In the above S4, the time reversal calculation includes: Generate a conjugate light field distribution according to the three-dimensional scattering characteristics, and dynamically adjust the switch state of the optical fiber bundle through the optical switch to form a focused light spot of the incident light field in the target tissue.

6. The deep near-infrared structured imaging method based on composite single-pixel scattering compensation according to claim 1, characterized in that: The arrangement of the input probe and the receiving probe satisfies the following conditions: According to the spatio-temporal modulation characteristics, the receiving probe and the input probe are arranged according to a specific distribution function.

7. The method for deep near-infrared structured imaging based on compound single-pixel scattering compensation according to claim 1, wherein: The method further includes: Replace the combination of the optical fiber bundle and the optical switch with a spatial light modulator or a digital micromirror array to perform binary modulation on the phase or amplitude of the incident light field.

8. A deep near-infrared structured imaging system based on composite single-pixel scattering compensation, characterized in that: It includes: A light source module for emitting near-infrared light; An optical fiber bundle module containing optical fiber lines arranged in an n×n pattern, and each optical fiber line is connected to an independent optical switch; A control module for controlling the state of the optical switch to generate a binary spatio-temporal coupling modulation light field; An input probe and a receiving probe array for transmitting the modulated light field to the target tissue and collecting the scattered response signal; A calculation module for performing composite correlation calculation, time reversal calculation and iterative optimization of scattering characteristics, and outputting functional imaging and structural imaging results.

9. The deep near-infrared structured imaging system based on compound single-pixel scattering compensation according to claim 8, wherein: The outer layer of the optical fiber of the optical fiber bundle module is coated with a biocompatible protective sleeve, and the output end of the optical fiber bundle is integrated into a wearable head-mounted device.

10. The deep near-infrared structured imaging system based on composite single-pixel scattering compensation according to claim 8, wherein: The basis for the control module to use binary modulation is as follows: According to Shannon's theorem, binary modulation is adopted to optimize the coding rate, reduce noise interference and improve the focusing efficiency.

Images