Dual-wavelength time-domain blood oxygen detection system, method, device and medium

Abstract

This application discloses a dual-wavelength time-domain blood oxygen detection system, method, device, and medium, belonging to the fields of optics and biomedical engineering. The system includes a signal generator, a timing controller, two pulsed lasers, an optical fiber coupler, a silicon photomultiplier tube, a signal frequency multiplier, and a time-correlated single-photon counter. The timing controller achieves interleaved excitation of dual-wavelength pulses, and the coupler separates 1% reference light and 99% probe light. At the detection end, the single-photon counter labels photons according to the excitation sequence, filters out shallow noise, and accumulates to construct a dual-wavelength photon histogram. Combined with finite element forward modeling based on the diffusion equation and Robin boundary conditions, the tissue absorption and scattering coefficients are simultaneously inverted, and the total hemoglobin concentration and blood oxygen saturation are calculated by modifying the Lambert-Beer law. This application achieves high-speed acquisition and high-precision reconstruction under the same physiological conditions with dual wavelengths, overcoming the problems of large dynamic errors and strong shallow interference in traditional systems.

Description

Technical Field

[0001] This application relates to the fields of optical and biomedical engineering, and in particular to a dual-wavelength time-domain blood oxygenation detection system, method, device and medium. Background Technology

[0002] Time-domain near-infrared spectroscopy (TD-NIRS), also known as time-domain near-infrared tissue optical functional imaging, is specifically used to address the application of NIRS in neuroimaging. It aims to map and understand the function of the human cerebral cortex and is one of the core technologies in near-infrared optical imaging that combines high information dimensionality and quantitative capabilities. Thanks to the development of ultrashort pulse lasers and single-photon detection technology, TD-fNIRS was first explored and applied by researchers in the fields of optics and biomedicine in the 1990s.

[0003] TD-fNIRS utilizes picosecond-level ultrashort pulses of near-infrared light (typically covering the 650-1000nm biooptical window) to irradiate biological tissues. By detecting the time-of-flight distribution of photons within the tissue and combining this with optical transport models (such as diffusion equations and Monte Carlo methods), it reconstructs the absolute optical coefficients (absorption coefficient, scattering coefficient) of the tissue, thereby quantitatively reflecting functional information such as blood oxygen saturation and cellular metabolic activity within the tissue. Leveraging its characteristics of being non-ionizing, non-invasive, easy to operate, and suitable for bedside application, TD-fNIRS demonstrates unique value in areas such as real-time brain function monitoring, dynamic assessment of muscle oxygenation, and early screening for neonatal brain injury.

[0004] However, optical coefficient reconstruction in a single band cannot reflect the specific information such as oxyhemoglobin and deoxyhemoglobin in biological tissues. Further reconstruction is required using different coefficients obtained from multiple wavelengths. Since the conditions in real biological tissues are constantly moving and changing, it is necessary to explore TD-fNIRS systems and methods with higher accuracy and smaller errors.

[0005] Near-infrared spectroscopy (NIRS) utilizes the characteristic that near-infrared light penetrates deeply into biological tissues (typically within the "optical window" of 650nm-1000nm) and is relatively weakly absorbed by major chromophores (water, lipids, and hemoglobin). Time-domain functional near-infrared spectroscopy (TD-fNIRS), as a higher-order form of NIRS, emits picosecond-level ultrashort laser pulses into tissues and precisely detects the time-of-flight (DTOF) distribution of photons. Combined with optical transport physics models, it can quantitatively invert the absolute optical parameters of tissues (absorption coefficient μa and reduced scattering coefficient μs'), thereby deriving functional information such as tissue oxygen saturation and cellular metabolic activity.

[0006] However, existing TD-fNIRS systems have significant limitations in practical applications: First, single-band optical coefficient reconstruction cannot distinguish the independent concentrations of oxyhemoglobin (HbO2) and deoxyhemoglobin (Hb), making it difficult to directly quantify core physiological indicators such as blood oxygen saturation; second, traditional multi-wavelength systems often rely on mechanically switching filters or alternately triggering different lasers, resulting in long acquisition cycles. Physiological activities of biological tissues (such as heartbeat, respiration, and hemodynamic fluctuations) and subtle environmental changes during the measurement interval can cause dynamic drift in tissue optical coefficients, introducing serious reconstruction errors; third, biological tissues are strong scattering media, and photons arriving at the detector early mostly originate from scattering by the superficial scalp or skull, carrying very little information about deep target tissues, making it difficult for traditional continuous acquisition modes to effectively remove superficial interference; finally, existing systems often treat the differential path factor (DPF) as a fixed empirical value, failing to consider the impact of dynamic changes in the tissue's microscopic scattering structure on optical path extension, leading to limited quantitative calculation accuracy. Therefore, there is an urgent need in this field for a time-domain near-infrared detection system and method that can achieve high-speed excitation of dual wavelengths under the same physiological state, possess deep selective acquisition capabilities, and can accurately and synchronously reconstruct hemoglobin concentration and blood oxygen saturation. Summary of the Invention

[0007] The purpose of this application is to overcome the shortcomings of existing TD-fNIRS technology, such as insufficient information acquisition dimensions, large dynamic measurement errors, and weak anti-interference ability. It provides a dual-wavelength time-domain blood oxygen detection system, method, device, and medium. Through the deep integration of time-series staggered excitation and single-photon time-gated technology, combined with the adjoint gradient reconstruction algorithm based on the average time of flight, high-speed synchronous acquisition of dual-wavelength data and high-precision parameter inversion are achieved while ensuring the consistency of physiological conditions.

[0008] To achieve the above objectives, this application provides the following solution: In a first aspect, this application provides a dual-wavelength time-domain blood oxygenation detection system, comprising: Excitation module and detection module; The excitation module includes a signal generator, a timing controller, a first pulse laser, a second pulse laser, and an optical fiber coupler; The detection module includes a silicon photomultiplier tube, a signal frequency multiplier, and a time-correlated single-photon counter; The output of the signal generator is connected to the input of the timing controller, and the two control outputs of the timing controller are respectively connected to the triggers of the first pulsed laser and the second pulsed laser. The two inputs of the fiber coupler receive the output light from the first pulsed laser and the second pulsed laser, respectively, and the output of the fiber coupler is split into a reference optical path and a probe optical path. The ends of the reference optical path and the probe optical path are both coupled to the photosensitive surface of the silicon photomultiplier tube. The electrical signal output of the silicon photomultiplier tube is connected to the signal input of the time-correlated single-photon counter, the input of the signal frequency multiplier is connected to the signal generator, and the output is connected to the synchronization clock of the time-correlated single-photon counter.

[0009] Optionally, the first pulsed laser is a 633nm pulsed laser, and the second pulsed laser is an 850nm pulsed laser; the fiber coupler is 2 2. Fiber optic coupler, wherein the splitting ratio is configured to allocate 1% of the input optical power to the reference optical path and 99% to the probe optical path.

[0010] Secondly, this application provides a dual-wavelength time-domain blood oxygenation detection method, including: S1. The signal generator outputs a reference pulse signal to the timing controller; the timing controller modulates the reference pulse signal into a first trigger sequence and a second trigger sequence that are strictly non-overlapping on the time axis, respectively driving the first pulse laser and the second pulse laser to alternately emit ultrashort pulse light, forming a dual-wavelength interleaved excitation timing sequence; S2. The dual-wavelength ultrashort pulse light, which is alternately emitted by the first pulse laser and the second pulse laser under the modulation of the timing controller described in step S1, is input into the fiber coupler for spatial beam combining. Then, it is separated into a weak reference light signal and a strong probe light signal according to the preset splitting ratio. The strong probe light signal irradiates the biological tissue to be tested and is scattered and absorbed, forming the output light signal after the tissue action. S3: The silicon photomultiplier tube converts the weak reference light signal and the emitted light signal output in step S2 into single-photon electrical pulses; the time-correlated single-photon counter receives the synchronization clock provided by the signal frequency multiplier, and timestamps and maps wavelength channel tags to the single-photon electrical pulses according to the dual-wavelength interleaved excitation sequence in step S1. At the same time, time-gating is enabled to filter out shallow scattered photons that are earlier than the set threshold, and the deep photons are accumulated and reconstructed into a first wavelength photon histogram and a second wavelength photon histogram respectively. S4. Perform time integration and time-weighted integration on the two photon histograms obtained in step S3 to extract the average flight time at each wavelength as a measured feature quantity. Substitute the measured feature quantity into the finite element forward propagation model based on the diffusion equation and Robin boundary conditions to construct the root mean square error objective function between the measured average flight time and the model predicted average flight time. Use the adjoint method to solve the gradient vector of the root mean square error objective function with respect to the tissue absorption coefficient and the reduced scattering coefficient. Iterate and update the parameters until convergence using the gradient descent method to simultaneously reconstruct the tissue absorption coefficient distribution and the reduced scattering coefficient distribution at both wavelengths. Use the reduced scattering coefficient distribution to perform scattering compensation on the light propagation path and extract the distribution of the change in the absorption coefficient at both wavelengths. S5. Extract the distribution of the change in the dual-wavelength absorption coefficient output in step S4, substitute it into the modified Lambert-Beer law to construct a system of dual-wavelength linear equations, solve the system simultaneously to obtain the change in oxyhemoglobin concentration and the change in deoxyhemoglobin concentration, and then calculate the total tissue hemoglobin concentration and tissue oxygen saturation.

[0011] Optionally, the specific process of time-gated filtering and wavelength channel label mapping in step S3 is as follows: a dual-channel data buffer is set inside the time-correlated single-photon counter. According to the alternating phase of the first trigger sequence and the second trigger sequence recorded by the timing controller, the arrival timestamp of each single-photon electrical pulse is mapped to the corresponding 633nm or 850nm wavelength channel. Only the photon counts whose arrival times are within the set gating window are retained, and the noise photons outside the window are discarded. Finally, two independent time-resolved photon histograms are output.

[0012] Optionally, the construction process of the finite element forward transmission model includes: Tetrahedral meshing is performed on the three-dimensional geometry of the target organization; The ultrashort pulse light source is approximated as a spatiotemporal Dirac delta function, and a light flux diffusion equation is established by combining Robin boundary conditions; The stiffness matrix and mass matrix are calculated in a finite element discrete framework, and the time-varying luminous flux distribution vector of the mesh nodes is obtained by solving. The time-varying luminous flux distribution vector is mapped to the detector's spatial position using a boundary measurement operator, and the theoretical average flight time is obtained by integral calculation.

[0013] Optionally, the parameter iterative update formula for the gradient descent method in step S4 is: ; in, Indicates the first The optical parameters of the step, Indicates the first The optical parameters of the step, Indicates the step size. Represents the gradient operator, This represents the objective function to be optimized.

[0014] Optionally, the modified Lambert-Beer law is used to construct the two-wavelength linear equation system as follows: ; in, Indicates the first wavelength. Indicates the second wavelength. This represents the change in optical density at the first wavelength. This indicates the change in optical density at the second wavelength. express The molar extinction coefficient at the first wavelength, express The molar extinction coefficient at the second wavelength, express Concentration changes, This represents the geometric distance between the light source and the detector. Represents the difference path factor. express The molar extinction coefficient at the first wavelength, express The molar extinction coefficient at the second wavelength, express The concentration change.

[0015] Optionally, the expression for the total hemoglobin concentration in the tissue is as follows: ; in, ; ; Indicates total hemoglobin concentration; The expression for tissue oxygen saturation is as follows: .

[0016] Thirdly, this application provides a computer device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the dual-wavelength time-domain blood oxygen detection method described in any one of the above.

[0017] Fourthly, this application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the dual-wavelength time-domain blood oxygen detection method described above.

[0018] According to the specific embodiments provided in this application, this application has the following technical effects: This application provides a dual-wavelength time-domain blood oxygenation detection system, method, device, and medium, which has the following significant advantages: Eliminating dynamic drift errors and improving acquisition efficiency: By using a timing controller, 633nm and 850nm pulses are strictly interleaved on the time axis, eliminating the need for mechanical switching of optical paths. This ensures that dual-wavelength data are acquired under completely identical tissue physiological states and environmental conditions, fundamentally avoiding the dynamic drift error of coefficients caused by physiological fluctuations in traditional alternating measurements, while significantly shortening the imaging cycle.

[0019] Depth-selective acquisition suppresses shallow noise: By utilizing the time-gating technology of time-correlated single-photon counters, the arrival time window of photons is precisely set, effectively filtering out scattered photons returning early from the shallow tissue (such as the scalp), selectively accumulating photon events carrying information about deep target tissues, and significantly improving the signal-to-noise ratio and spatial positioning accuracy of deep signals.

[0020] High-precision forward modeling and efficient inverse reconstruction: The mean time of flight (MTOF) is used as the core observation feature. A forward model is constructed by combining the diffusion equation, Robin boundary conditions and finite element discrete framework. The adjoint method is used to efficiently calculate the gradient of the objective function. Combined with the gradient descent method with the Armijo criterion adaptive step size, the absorption coefficient and scattering coefficient are rapidly, stably and synchronously inverted.

[0021] Reliable quantitative calculation of clinical parameters: Based on the reconstructed dual-wavelength absorption coefficient, a linear equation system is constructed by combining the modified Lambert-Beer law to directly analyze the changes in oxygenated / deoxygenated hemoglobin concentration, and then accurately calculate the total hemoglobin concentration and tissue oxygen saturation, providing quantitative and reliable biomedical indicators for brain functional imaging and ischemia-hypoxia monitoring. Attached Figure Description

[0022] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0023] Figure 1 This is a schematic diagram of a dual-wavelength time-domain blood oxygen detection system according to an embodiment of this application; Figure 2 A schematic flowchart of a dual-wavelength time-domain blood oxygen detection method provided in an embodiment of this application; Figure 3 A schematic diagram illustrating dual-wavelength and depth information filtering for a time-domain gating method provided in an embodiment of this application; Figure 4 A source probe diagram of a simulation model provided in an embodiment of this application; Figure 5 This is a schematic diagram of an optical transmission path provided in an embodiment of this application; Figure 6 This is a schematic diagram of the structure of a computer device provided in an embodiment of this application.

[0024] Figure label: Signal generator 1, timing controller 2, first pulse laser 3, second pulse laser 4, fiber optic coupler 5, silicon photomultiplier tube 6, signal frequency multiplier 7, time-correlated single-photon counter 8, and sample under test 9. Detailed Implementation

[0025] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0026] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0027] In one exemplary embodiment, such as Figure 1 As shown, a dual-wavelength time-domain blood oxygen detection system is provided, which includes an excitation module and a detection module; The excitation module includes a signal generator 1, a timing controller 2, a first pulse laser 3, a second pulse laser 4, and an optical fiber coupler 5. The detection module includes a silicon photomultiplier tube 6, a signal frequency multiplier 7, and a time-correlated single-photon counter 8; The output of signal generator 1 is connected to the input of timing controller 2. The two control outputs of timing controller 2 are connected to the triggers of the first pulse laser 3 and the second pulse laser 4, respectively. The two inputs of fiber optic coupler 5 receive the output light from the first pulse laser 3 and the second pulse laser 4, respectively. The output of fiber optic coupler 5 is split into a reference optical path and a probe optical path. The ends of the reference optical path and the probe optical path are both coupled to the photosensitive surface of silicon photomultiplier tube 6. The electrical signal output of silicon photomultiplier tube 6 is connected to the signal input of time-correlated single-photon counter 8. The input of signal frequency multiplier 7 is connected to signal generator 1, and the output is connected to the synchronization clock of time-correlated single-photon counter 8.

[0028] Among them, the first pulsed laser 3 is a 633nm pulsed laser, the second pulsed laser 4 is an 850nm pulsed laser; the fiber coupler 5 is a 2 2. Fiber optic coupler, wherein the splitting ratio is configured to allocate 1% of the input optical power to the reference optical path and 99% to the probe optical path.

[0029] Specifically, in the excitation module of the dual-wavelength TD-fNIRS system, the timing signal is generated by pulse signal generator 1, modulated by timing controller 2 to obtain two signals, which are then passed through a 633nm pulsed laser and an 850nm pulsed laser respectively to generate two timing-designed pulse signals of different wavelengths. After coupling by the fiber optic coupler, signals with signal strengths of 1% and 99% are output for detection, respectively. In the detection module, photons in the pulse signal are sampled as electrons by the silicon photomultiplier tube 6 after being scattered and absorbed by the sample. The obtained electrons are finally statistically analyzed into a histogram by the time-correlated single-photon counter 8 for processing.

[0030] The system utilizes signal generator 1 and timing controller 2 to output strictly staggered excitation signals, ensuring that the two lasers do not overlap in time. A 633nm pulsed laser 3 and an 850nm pulsed laser 4 are used as the light source because near-infrared light experiences relatively weaker scattering and absorption in tissues compared to visible light, which is beneficial for improving the waveform quality acquired by TD-NIRS; 2 2. Fiber optic coupler 5 is used to couple two wavelengths of pulsed laser and output them in separate paths; silicon photomultiplier tube 6 is used to detect optical signals; signal frequency multiplier 7 is used to multiply the signal generator 1 and provide it to time-correlated single-photon counter 8 as a synchronization signal, and on this basis, the photon signals detected by silicon photomultiplier tube 6 are statistically analyzed into a histogram.

[0031] In one exemplary embodiment, such as Figure 2 As shown, a dual-wavelength time-domain blood oxygen detection method is provided. This method is executed by a computer device, specifically by a terminal or server alone, or by both a terminal and a server. In this embodiment, the method includes the following steps: S1. The signal generator outputs a reference pulse signal to the timing controller. The timing controller modulates the reference pulse signal into a first trigger sequence and a second trigger sequence that are strictly non-overlapping on the time axis, respectively driving the first pulse laser and the second pulse laser to alternately emit ultrashort pulse light, forming a dual-wavelength interleaved excitation timing sequence.

[0032] Specifically, the sample to be tested 9 is placed in an optical darkroom, and the signal generator and timing controller are turned on to form an excitation signal.

[0033] S2. The dual-wavelength ultrashort pulse light, which is alternately emitted by the first pulse laser and the second pulse laser under the modulation of the timing controller described in step S1, is input into the fiber coupler for spatial beam combining. Then, it is separated into a weak reference light signal and a strong probe light signal according to the preset splitting ratio. The strong probe light signal irradiates the biological tissue to be tested and is scattered and absorbed, forming the output light signal after the tissue action. The outputs of the two near-infrared lasers are combined and split into two signals with an intensity ratio of 1:99 by an optical fiber coupler. The 1% weak signal is directly converted into an electrical signal by a silicon photomultiplier tube and reconstructed into the first waveform by a time-correlated single-photon counter. The 99% strong signal is input into the silicon photomultiplier tube after passing through the sample under test, and finally the second waveform is obtained, which is broadened and delayed compared to the first waveform.

[0034] S3: The silicon photomultiplier tube converts the weak reference light signal and the emitted light signal output in step S2 into single-photon electrical pulses; the time-correlated single-photon counter receives the synchronization clock provided by the signal frequency multiplier, and timestamps and maps wavelength channel tags to the single-photon electrical pulses according to the dual-wavelength interleaved excitation sequence in step S1. At the same time, time-gating is enabled to filter out shallow scattered photons that are earlier than the set threshold, and the deep photons are accumulated and reconstructed into a first wavelength photon histogram and a second wavelength photon histogram respectively.

[0035] S4. Perform time integration and time-weighted integration on the two photon histograms obtained in step S3 to extract the average flight time at each wavelength as a measured feature quantity. Substitute the measured feature quantity into the finite element forward propagation model based on the diffusion equation and Robin boundary conditions to construct the root mean square error objective function between the measured average flight time and the model predicted average flight time. Use the adjoint method to solve the gradient vector of the root mean square error objective function with respect to the tissue absorption coefficient and the reduced scattering coefficient. Iterate and update the parameters until convergence using the gradient descent method to simultaneously reconstruct the tissue absorption coefficient distribution and the reduced scattering coefficient distribution at both wavelengths. Use the reduced scattering coefficient distribution to perform scattering compensation on the light propagation path and extract the distribution of the change in the absorption coefficient at both wavelengths.

[0036] S5. Extract the distribution of the change in the dual-wavelength absorption coefficient output in step S4, substitute it into the modified Lambert-Beer law to construct a system of dual-wavelength linear equations, solve the system simultaneously to obtain the change in oxyhemoglobin concentration and the change in deoxyhemoglobin concentration, and then calculate the total tissue hemoglobin concentration and tissue oxygen saturation.

[0037] This application uses the time-domain diffusion equation to model the light transmission process in an object: (1); in, Represents the absorption coefficient. Represents the diffusion coefficient. Represents the speed of light. Represents light intensity. This represents the derivative of light intensity with respect to time. In this application, given the extremely short pulse width of the pulsed light source used, the light source term can be approximated as a Dirac function in spacetime. For convenience, the start time of the pulse is set to time zero. , Represents a position vector. Represents the position vector of the light source. Indicates the pulse start time. Indicates the time of solution.

[0038] For boundary conditions, this application uses Robin boundary conditions: (2); in, , It is the reflection coefficient within the diffusion transmission. Represents the boundary normal vector. Denotes the solution domain. Indicates the refractive index mismatch factor. Represents the boundary normal derivative.

[0039] The intensity of the emitted light is given by the following formula: (3); Combining the Robin boundary conditions, we can obtain ,in, , This indicates the intensity of the emitted light.

[0040] The emission intensity, which varies with time, can be collected at different times to obtain a time-resolved curve. This application does not directly use the entire time-resolved curve for reconstruction, but uses the average flight time derived from it as the basis for reconstruction. First, the integral intensity is defined. and time-weighted strength as follows: (4); (5); in, The first-order moment represents the intensity of the emitted light.

[0041] For average flight time Then it is given by the following formula: (6); Within the finite element framework, we have: (7); in, Represents the steady-state finite element matrix. Indicates light intensity. Represents the time term of the finite element matrix. Indicates the light source item; , and It can be represented as: (8); (9); in, This represents the terms in the finite element matrix related to the diffusion coefficient. This represents the terms in the finite element matrix related to the absorption coefficient. In the The diffusion coefficient is represented on each unit cell. This represents the number of diffusion coefficients to be reconstructed. The interpolation basis function represents the diffusion coefficient. Indicates the first Units, The interpolation basis function represents the absorption coefficient. This represents the number of absorption coefficients to be reconstructed. Indicates the first Absorption coefficient per unit, This represents the terms in the finite element matrix that are related to the boundary conditions.

[0042] After obtaining the finite element solution vector, the probe data is then processed by the boundary measurement operator. Map the solution vector to the probe vector matrix elements in Representing the The position vectors of the detectors, where... Represents the real number field. Indicates the number of detectors. Indicates the number of nodes. Indicates light intensity. This represents the interpolation basis function.

[0043] In order to reconstruct the optical parameters within the finite element framework, the following optimization objective is established: (10); in, The characteristic quantity representing the time-resolved curve is the integral intensity. or average flight time . This represents the characteristic quantity obtained from experimental measurements. Represents the feature quantity calculated based on the model, where It is a parameter vector composed of absorption coefficient and scattering coefficient. This represents the objective function to be optimized.

[0044] This application uses the gradient descent method to solve the optimization problem. (11); Among them, step size The gradient of the objective function is selected according to the Armijo criterion. , It is the Jacobian matrix of the average flight time with respect to the parameters. Indicates the first The optical parameters of the step, where the elements are obtained by differentiation: (12); in, and The calculation is performed according to the adjoint method as follows: (13); (14); in, The zeroth moment representing the intensity of the emitted light. This represents the first moment of the original problem with respect to time. Finite element interpolation basis functions representing optical parameters This represents the solution to the accompanying problem. The first moment representing the intensity of the emitted light. Indicates the first One optical parameter, This represents the first moment of the original problem with respect to time.

[0045] in, For the equation The solution, For the equation The solution, For the equation The solution, For the equation The solution. For the time-resolved curves collected in the experiment, first calculate the zero-order moment and the first-order moment equations, and then calculate the gradient according to formulas (12), (13), and (14), that is, the gradient in formula (11) is... renew.

[0046] Since biological tissues are strong scattering media, light does not travel in a straight line within them. Therefore, a modified Lambert-Beer law is needed to describe the relationship between light attenuation and concentration. (15); in, Wavelength The change in optical density (logarithm of the ratio of incident light intensity to outgoing light intensity); , express and At wavelength The molar extinction coefficient at the given time (a known constant); , express and The concentration change; This represents the geometric distance between the light source and the detector. This represents the differential path factor (the ratio of the actual optical path length to the geometric distance, reflecting the optical path lengthening caused by scattering; it is typically 4–6 and is related to tissue type). This indicates background scattering and other interference (which can be ignored in a short time or eliminated by the difference between two wavelengths).

[0047] Dual-wavelength measurement and equation solving: selecting two wavelengths (e.g., 630nm) and (e.g., 850nm), establish a system of equations: (16); By solving algebraically, we can obtain and Concentration changes: (17); (18); The change in total hemoglobin concentration is as follows: (19); Tissue oxygen saturation (StO2) is defined as the proportion of oxyhemoglobin to total hemoglobin. (20); The aforementioned dual-wavelength time-domain blood oxygenation detection method aims to simultaneously excite and randomly probe two wavelengths under the same conditions in the traditional TD-fNIRS system. This ensures that the pulse broadening obtained under both wavelengths has the same tissue and environmental conditions, avoiding noise and errors caused by biological activities or environmental changes. The system uses a timing controller to non-overlap the high-frequency pulse signals of the two wavelengths, and then uses an optical fiber coupler to couple and output superimposed signals with 1% and 99% power, respectively. The 1% power signal is directly fed to a silicon photomultiplier tube to solve the system's instrument response function, while the 99% power signal is used to input the test sample to obtain the broadening signal. By comparing the two, the true broadening capability of the test sample to the signal is restored.

[0048] The aforementioned dual-wavelength time-domain blood oxygenation detection method is based on the time-gating capability of a time-correlated single-photon counter. In this method, a silicon photomultiplier tube alternately receives signals at 850nm and 633nm at a fixed frequency. The time-correlated single-photon counter can capture photons from each single pulse waveform with a low probability. By utilizing the time-gating capability, the wavelength information corresponding to each captured photon can be recorded. Finally, the corresponding photons are accumulated into a histogram to reconstruct the pulse waveform and obtain the broadening information. This method can maintain the consistency of conditions during the broadening of pulses at different wavelengths and can also filter out photon information returned from the shallow layer of the sample through gating technology, thereby selectively obtaining information from the deeper layers of the sample.

[0049] To address the limitations of existing TD-fNIRS systems, such as limited information acquisition and significant interference, this application improves the excitation mode and innovatively proposes a dual-wavelength time-domain blood oxygenation detection system. This system can improve detection speed and anti-interference capability under multi-wavelength conditions while maintaining reconstruction accuracy. This application has the following advantages: The dual-wavelength time-domain blood oxygenation detection system proposed in this application achieves simultaneous excitation of the sample with two wavelengths within the same time period without switching optical paths, and the number of wavelengths can still be increased in the same way. By using two wavelengths to alternately excite the sample during the same time period and acquiring photons with tagged information to reconstruct pulse waveforms in the form of a histogram, it solves the shortcomings of insufficient reconstruction information and accuracy in traditional TD-fNIRS systems. The line-scan FDOT forward process modeling method proposed in this application, by discretizing the line source into multiple point sources and treating the Green's function of the line source as a weighted sum of the Green's functions of the point sources, can accurately model the diffusion motion of the line source within the sample, ensuring the accuracy of forward modeling and inverse reconstruction. A timing controller enables strictly staggered excitation of 633nm and 850nm pulses on the time axis, eliminating the need for mechanical switching of optical paths. This ensures that dual-wavelength data are acquired under completely identical tissue physiological states and environmental conditions, fundamentally avoiding the dynamic drift error of coefficients caused by physiological fluctuations in traditional alternating measurements, while significantly shortening the imaging cycle.

[0050] This application models the propagation process of light within an object based on the diffusion equation. Traditionally, the propagation of light in tissues is modeled using the radiative transfer equation (RTE). However, this equation contains six independent variables and is difficult to solve directly. Therefore, the modeling method used in this application applies an approximation model—the diffusion equation (DE) derived by performing a first-order spherical harmonic expansion of the luminous flux.

[0051] like Figure 3 As shown, Figure 3 Parts (a) and (b) of this application illustrate the principle of dual-wavelength and depth information filtering based on the time-domain gating method. Figure 3 As can be seen, this method can effectively separate signals of different wavelengths and simultaneously obtain the depth information of the target, thereby achieving joint screening of wavelength and spatial dimension.

[0052] like Figure 4 As shown, Figure 4 The image shows the probed object and the spatial distribution of the source probe, from... Figure 4 It can be seen that the light source and detector are distributed at equal intervals in On the surface of the object, signals of different wavelengths are emitted by a light source, and these signals are then received by a detector for subsequent inversion of optical parameters.

[0053] like Figure 5 As shown, Figure 5 This shows the spatial sensitivity distribution of the absorption coefficient under the condition of maximum source distance, from Figure 5The distribution shows a clear depth dependence, with higher sensitivity in the shallow region (closer to the detection surface), while the sensitivity gradually decreases with increasing depth, and the sensitivity in the deep region is relatively low.

[0054] Given the location of the light source and the geometric structure of the object, the forward problem involves solving for surface measurements based on known optical parameters. The solution can be summarized as follows: First, the object is meshed using tetrahedrons. Then, the stiffness matrix is ​​calculated, and a finite element equation is established. This equation is then solved to obtain the time-varying luminous flux distribution at each node. Finally, combined with time information, the optical measurements are calculated. In this application, the first moment of the luminous flux, i.e., the Mean Time of Flight (MTOF), is used as the measurement information. For the inverse problem, which involves reversing the diffusion equation based on known surface measurements, the solution steps are as follows: An optimization objective function is constructed based on the difference between the surface measurements and the forward predictions of the physical model. Given an initial estimate of the optical parameters, the gradient information of the objective function with respect to the optical parameters is calculated using the adjoint method. The optical parameters are iteratively updated based on the gradient information until convergence or the algorithm's stopping condition is met.

[0055] In one exemplary embodiment, a computer device is provided, which may be a server or a terminal, and its internal structure diagram may be as follows. Figure 6 As shown, the computer device includes a processor, memory, input / output (I / O) interfaces, and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is also connected to the system bus via the I / O interfaces. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and a database. The internal memory provides the environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The database stores dual-wavelength time-domain blood oxygenation detection data. The I / O interfaces are used for information exchange between the processor and external devices. The communication interface is used for communication with external terminals via a network connection. When the computer program is executed by the processor, it implements a dual-wavelength time-domain blood oxygenation detection method.

[0056] Those skilled in the art will understand that Figure 6The structures shown are merely block diagrams of some structures related to the present application and do not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than shown in the figures, or combine certain components, or have different component arrangements. In an exemplary embodiment, a computer device is provided, including a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the steps in the above-described method embodiments.

[0057] In one exemplary embodiment, a computer-readable storage medium is provided storing a computer program that, when executed by a processor, implements the steps in the above-described method embodiments.

[0058] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0059] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments described above. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM).

[0060] The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0061] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0062] This document uses specific examples to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this application. Furthermore, those skilled in the art will recognize that, based on the ideas of this application, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A dual-wavelength time-domain blood oxygen detection system, characterized in that, The dual-wavelength time-domain blood oxygen detection system includes: an excitation module and a detection module; The excitation module includes a signal generator, a timing controller, a first pulse laser, a second pulse laser, and an optical fiber coupler; The detection module includes a silicon photomultiplier tube, a signal frequency multiplier, and a time-correlated single-photon counter; The output of the signal generator is connected to the input of the timing controller, and the two control outputs of the timing controller are respectively connected to the triggers of the first pulsed laser and the second pulsed laser. The two inputs of the fiber coupler receive the output light from the first pulsed laser and the second pulsed laser, respectively, and the output of the fiber coupler is split into a reference optical path and a probe optical path. The ends of the reference optical path and the probe optical path are both coupled to the photosensitive surface of the silicon photomultiplier tube. The electrical signal output of the silicon photomultiplier tube is connected to the signal input of the time-correlated single-photon counter, the input of the signal frequency multiplier is connected to the signal generator, and the output is connected to the synchronization clock of the time-correlated single-photon counter.

2. The dual-wavelength time-domain blood oxygenation detection system according to claim 1, characterized in that, The first pulsed laser is a 633 nm pulsed laser and the second pulsed laser is an 850 nm pulsed laser; the fiber coupler is a 2 2 a fiber coupler with a splitting ratio configured to allocate 1% of the input optical power to the reference light path and 99% to the probe light path.

3. A dual-wavelength time-domain blood oxygen detection method, characterized in that, The dual-wavelength time-domain blood oxygenation detection method includes: S1. The signal generator outputs a reference pulse signal to the timing controller; the timing controller modulates the reference pulse signal into a first trigger sequence and a second trigger sequence that are strictly non-overlapping on the time axis, respectively driving the first pulse laser and the second pulse laser to alternately emit ultrashort pulse light, forming a dual-wavelength interleaved excitation timing sequence; S2. The ultrashort pulse light emitted alternately by the first pulse laser and the second pulse laser under the modulation of the timing controller described in step S1 is input into the fiber coupler for spatial beam combining. Then, it is separated into a weak reference light signal and a strong probe light signal according to the preset splitting ratio. The strong probe light signal irradiates the biological tissue to be tested and undergoes scattering and absorption to form the output light signal after the tissue action. S3. The silicon photomultiplier tube converts the weak reference light signal and the emitted light signal output in step S2 into single-photon electrical pulses; the time-correlated single-photon counter receives the synchronization clock provided by the signal frequency multiplier, and records the timestamps and maps the wavelength channel labels of the single-photon electrical pulses according to the dual-wavelength interleaved excitation sequence in step S1. At the same time, time-gating is enabled to filter out shallow scattered photons that are earlier than the set threshold, and the deep photons are accumulated and reconstructed into the first wavelength photon histogram and the second wavelength photon histogram respectively. S4. Perform time integration and time-weighted integration on the two photon histograms obtained in step S3 to extract the average flight time at each wavelength as a measured feature quantity. Substitute the measured feature quantity into the finite element forward propagation model based on the diffusion equation and Robin boundary conditions to construct the root mean square error objective function between the measured average flight time and the model predicted average flight time. Use the adjoint method to solve the gradient vector of the root mean square error objective function with respect to the tissue absorption coefficient and the reduced scattering coefficient. Iterate and update the parameters until convergence using the gradient descent method to simultaneously reconstruct the tissue absorption coefficient distribution and the reduced scattering coefficient distribution at both wavelengths. Use the reduced scattering coefficient distribution to perform scattering compensation on the light propagation path and extract the distribution of the change in the absorption coefficient at both wavelengths. S5. Extract the distribution of the change in the dual-wavelength absorption coefficient output in step S4, substitute it into the modified Lambert-Beer law to construct a system of dual-wavelength linear equations, solve the system simultaneously to obtain the change in oxyhemoglobin concentration and the change in deoxyhemoglobin concentration, and then calculate the total tissue hemoglobin concentration and tissue oxygen saturation.

4. The dual-wavelength time-domain blood oxygen detection method according to claim 3, wherein, The specific process of time-gated filtering and wavelength channel label mapping in step S3 is as follows: A dual-channel data buffer is set inside the time-correlated single-photon counter. According to the alternating phase of the first trigger sequence and the second trigger sequence recorded by the timing controller, the arrival timestamp of each single-photon electrical pulse is mapped to a 633nm or 850nm wavelength channel. Only the photon counts whose arrival times are within the set gating window are retained, and the noise photons outside the window are discarded. Finally, two independent time-resolved photon histograms are output.

5. The dual-wavelength time-domain blood oxygen detection method according to claim 3, wherein, The construction process of the finite element forward transmission model includes: Tetrahedral meshing is performed on the three-dimensional geometry of the target organization; The ultrashort pulse light source is approximated as a spatiotemporal Dirac delta function, and a light flux diffusion equation is established by combining Robin boundary conditions; The stiffness matrix and mass matrix are calculated in a finite element discrete framework, and the time-varying luminous flux distribution vector of the mesh nodes is obtained by solving. The time-varying luminous flux distribution vector is mapped to the detector's spatial position using a boundary measurement operator, and the theoretical average flight time is obtained by integral calculation.

6. The dual-wavelength time-domain blood oxygen detection method according to claim 3, wherein, The parameter iterative update formula for the gradient descent method in step S4 is as follows: ； wherein, represents the optical parameter of the first step, represents the optical parameter of the first step, represents the step size, represents the gradient operator, represents the optimization objective function.

7. The dual-wavelength time-domain blood oxygenation detection method according to claim 3, characterized in that, The modified Lambert-Beer law is expressed as follows for the two-wavelength linear equation system: ； in, Indicates the first wavelength. Indicates the second wavelength. This represents the change in optical density at the first wavelength. This indicates the change in optical density at the second wavelength. express The molar extinction coefficient at the first wavelength, express The molar extinction coefficient at the second wavelength, express Concentration changes, This represents the geometric distance between the light source and the detector. Represents the difference path factor. express The molar extinction coefficient at the first wavelength, express The molar extinction coefficient at the second wavelength, express The concentration change.

8. The dual-wavelength time-domain blood oxygenation detection method according to claim 7, characterized in that, The expression for the total hemoglobin concentration in the tissue is as follows: ； in, ; ； Indicates total hemoglobin concentration; The expression for tissue oxygen saturation is as follows: 。 9. A computer device, comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that the processor executes the computer program to implement the dual-wavelength time-domain blood oxygen detection method according to any one of claims 3-8.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the computer program implements the dual-wavelength time-domain blood oxygen detection method as described in any one of claims 3-8.

Images