Light path calibration method of flow cytometer

Abstract

The invention provides a flow cytometer light path calibration method, which belongs to the technical field of flow cytometers, and comprises the following steps: firstly, collecting light path detection original data of standard fluorescent microspheres, including laser spot position coordinates, detector output voltage value, dark counting rate and signal counting rate; then establishing a detection channel conversion cost matrix, and calculating an optimal light path detection sequence by using a traveling salesman problem optimization algorithm; adjusting the positions of a laser, an optical filter and a detector according to the optimal sequence, and obtaining optical path calibration data; the collected parameters are substituted into an optical path parameter optimization equation set to be solved, and an optical path collimation coefficient, a detector response coefficient, a signal crosstalk coefficient and an optimal gain value are obtained; and finally, respectively adjusting the position of the laser, the working voltage of the detector, the position of the optical filter and the voltage of the photomultiplier according to the coefficients to finish the accurate calibration of the optical path. The problems that an existing method depending on artificial experience is low in efficiency and overall optimization of a system level is difficult to achieve are solved.

Description

Technical Field

[0001] This invention belongs to the field of flow cytometer technology, and more specifically, relates to a flow cytometer optical path calibration method. Background Technology

[0002] Most existing flow cytometers employ a laser-photodetector optical path structure. The laser provides high-brightness, high-power monochromatic excitation light, while various optical components such as lenses, mirrors, and filters focus, separate, and filter scattered light and fluorescence signals. High-sensitivity photodetectors, such as photomultiplier tubes and avalanche photodiodes, convert these optical signals into measurable electrical signals. To meet complex detection requirements, flow cytometers are typically equipped with multiple lasers and detection channels, enabling simultaneous multi-parameter detection of samples. However, due to differences and uncertainties in the optical and mechanical properties of various components within the optical path system, coupled with changes in daily use and environmental conditions, distortion and misalignment of the optical path system can easily occur, leading to a significant decline in key performance indicators such as sensitivity, resolution, and linear range. Therefore, it is necessary to periodically calibrate and optimize the optical path system of the flow cytometer to ensure its long-term stable and reliable operation.

[0003] Currently, the optical path calibration of flow cytometers mainly relies on the operator's experience and intuition. The system performance is usually optimized by adjusting the laser position, detector operating voltage, and filter angle one by one. This trial-and-error adjustment method is inefficient and makes it difficult to achieve overall optimization at the system level. Summary of the Invention

[0004] In view of this, the present invention provides a flow cytometer optical path calibration method, which can solve the problem that the existing flow cytometer optical path calibration work mainly relies on the operator's experience, resulting in low efficiency and difficulty in achieving overall system-level optimization.

[0005] This invention is implemented as follows:

[0006] This invention provides a method for calibrating the optical path of a flow cytometer, comprising the following steps:

[0007] S10. Collect the raw optical path detection data of the standard fluorescent microspheres. The raw optical path detection data includes the laser spot position coordinates, detector output voltage value, dark count rate, and signal count rate.

[0008] S20. Establish an optical path detection channel conversion cost matrix, wherein the cost matrix represents the optical path adjustment time required for switching between different detection channels;

[0009] S30. Calculate the optimal optical path detection sequence based on the Traveling Salesman Problem optimization algorithm;

[0010] S40. Adjust the laser position, filter position, and detector position according to the optimal optical path detection sequence to obtain the optical path calibration data of the standard fluorescent microsphere;

[0011] S50. Substitute the laser spot position coordinates, the detector output voltage value, the dark count rate, the signal count rate, and the filter position into a preset optical path parameter optimization equation set and solve it to obtain the optical path collimation coefficient, detector response coefficient, signal crosstalk coefficient, and optimal gain value.

[0012] S60. Adjust the laser position according to the optical path collimation coefficient, adjust the detector operating voltage according to the detector response coefficient, adjust the filter position according to the signal crosstalk coefficient, and adjust the photomultiplier tube voltage according to the optimal gain value to complete the optical path calibration.

[0013] The steps for calculating the optimal optical path detection sequence based on the Traveling Salesman Problem optimization algorithm specifically include:

[0014] Step 1: Establish a basic directed graph, setting the current state of each optical path detection channel as the starting node in the basic directed graph, and setting the target state of each optical path detection channel as the ending node in the basic directed graph.

[0015] Step 2: Calculate the path weights between nodes based on the cost matrix. The path weights include optical path adjustment time, optical element switching time, and signal stabilization time.

[0016] Step 3: Divide the basic directed graph into a scattered light detection level, a fluorescence detection level, and a data acquisition level;

[0017] Step 4: Execute a dynamic programming algorithm in the basic directed graph to ensure that the transition from the start node to the end node of each optical path detection channel occurs only once, and calculate the state transition sequence with the minimum total path weight;

[0018] Step 5: Generate laser position adjustment command, filter position adjustment command, and detector position adjustment command according to the state transition sequence to form the optimal optical path detection sequence.

[0019] The optical path parameter optimization equation set includes the optical path collimation equation, the detector response equation, the signal crosstalk equation, and the system gain equation.

[0020] The optical path collimation equation is used to calculate the degree of optical path deviation. The inputs include the laser spot position coordinates, the detector response coefficient, and the ideal spot center coordinates. The output is the optical path collimation coefficient.

[0021] The detector response equation is used to evaluate the detector sensitivity. The inputs include the detector output voltage value, the optical path collimation coefficient, and the detector operating voltage. The output is the detector response coefficient.

[0022] The signal crosstalk equation is used to calculate the degree of signal leakage between detection channels. The inputs include the detector output voltage value, the detector response coefficient, and the filter position. The output is the signal crosstalk coefficient.

[0023] The system gain equation is used to determine the photomultiplier tube voltage. The inputs include the dark count rate, the signal count rate, the optical path collimation coefficient, and the detector response coefficient. The output is the optimal gain value.

[0024] The standard fluorescent microspheres refer to polystyrene fluorescent microspheres that have undergone strict quality control and standardization. These microspheres have a highly uniform particle size distribution and stable fluorescence intensity characteristics. Their surface is covalently bonded with wavelength-specific fluorescent dye molecules, which can generate stable fluorescence emission at specific excitation wavelengths. They are widely used in the routine performance verification and quantitative calibration of flow cytometers and can be used to evaluate key parameters of the instrument such as fluorescence detection sensitivity, resolution, and linear range.

[0025] The detectors refer to high-sensitivity photoelectric conversion devices installed in the optical path system of a flow cytometer. They mainly include two types: photomultiplier tubes and avalanche photodiodes. These detectors have excellent single-photon detection capabilities, a wide linear dynamic range, and fast response characteristics. They can convert weak scattered light signals and fluorescence signals from the sample into measurable electrical signal outputs. Their performance directly affects the detection sensitivity and accuracy of the instrument.

[0026] The dark count rate refers to the frequency of the detector's background output signal under conditions without any sample or excitation light irradiation. This parameter reflects the detector's inherent noise level and stability characteristics, and includes the combined effects of various factors such as thermionic emission, stray light interference, and electronic noise. It is an important basis for evaluating detector performance and setting signal judgment thresholds.

[0027] The optical path detection channel conversion cost matrix considers multiple factors such as the physical distance between each detection channel in the optical path system, the mechanical motion characteristics of optical components, the time delay required for signal stabilization, the inherent delay of system control response, and the accuracy limitations of the mechanical positioning system. It optimizes the state switching strategy in the multi-channel detection process by establishing a complete time cost model.

[0028] The optical path calibration data considers the spatial distribution characteristics of the laser spot, energy density distribution, spectral response curve of the detector, quantum efficiency, signal gain characteristics, center wavelength, half width at half maximum, transmittance, attenuation ratio and other core parameters of the optical path system, as well as the comprehensive performance indicators of the system such as signal linear range, dark count level and signal crosstalk degree.

[0029] The equations in the optical path parameter optimization equation set are coupled with each other. The effect is that by establishing the interrelationship between the optical path collimation equation, detector response equation, signal crosstalk equation and system gain equation, the overall optimization of various parameters of the optical path system is achieved. This enables operations such as laser position adjustment, detector operating voltage setting, filter position adjustment and photomultiplier tube gain control to be carried out in a coordinated manner, ensuring that the system reaches the best working state.

[0030] The steps of adjusting the laser position according to the optical path collimation coefficient are as follows: First, the spatial deviation vector between the actual position and the ideal position of the laser spot is calculated based on the optical path collimation coefficient. Then, the motion parameters of the displacement platform are determined according to the direction and magnitude of the deviation vector. An iterative optimization strategy is used to gradually adjust the laser position until the spot position reaches the preset collimation accuracy requirement.

[0031] The steps of adjusting the detector operating voltage according to the detector response coefficient are as follows: Based on the detector's response characteristic curve and signal detection requirements, the gain characteristics, dark count level and signal linear range of the detector under different operating voltages are analyzed. The optimal operating voltage value is calculated through numerical optimization methods. The detector operating voltage is gradually adjusted using a precise high-voltage control circuit. At the same time, the stability and consistency of the output signal are monitored to ensure that the detector operates in the optimal performance state.

[0032] The steps of adjusting the filter position according to the signal crosstalk coefficient are as follows: First, the source and propagation path of signal crosstalk between each detection channel are identified. Then, the optimal working position of the filter is calculated based on the signal crosstalk coefficient. A high-precision position feedback control system is used to achieve precise positioning of the filter. Finally, the adjustment result is verified by measuring the crosstalk suppression effect.

[0033] The steps for adjusting the photomultiplier tube voltage according to the optimal gain value are as follows: Based on the signal detection range and sensitivity requirements of the system, a suitable operating gain range is determined by analyzing the relationship between the dark count rate and the signal count rate. Then, the optimal operating point is selected based on a comprehensive consideration of the signal-to-noise ratio and the linear dynamic range. A multi-stage voltage divider network is used to precisely control the voltage distribution of each stage of the photomultiplier tube. Finally, the signal detection performance and stability of the system are verified through testing.

[0034] The formulas or equations involved in this invention will be described in detail below:

[0035] 1. Optical path detection channel conversion cost matrix:

[0036] The cost matrix is ​​represented as:

[0037]

[0038] In the formula, c ij This represents the total time cost required to switch from detection channel i to detection channel j; n is the total number of detection channels. Each element c... ij The calculation formula is:

[0039] c ij =t m (d ij )+t s (p ij )+t d +δ ij ;

[0040] Where, t m (d ij ) is a function of mechanical motion time, and is related to the distance d between channels. ij Related; t s (p ij ) is a function of signal settling time, related to optical path parameter p. ij Related; t d δ is the inherent delay time of the system. ij The random error term follows a normal distribution N(0, σ). 2 ).

[0041] 2. Calculation of the optimal optical path detection sequence:

[0042] Optimization of the objective function based on the state transition sequence of dynamic programming:

[0043]

[0044] Constraints:

[0045] s k ∈{1,2,…,n} and are all distinct;

[0046] In the formula, s k This is the number of the k-th detection channel in the sequence; This represents the time cost in the cost matrix.

[0047] 3. Optical path collimation equation:

[0048]

[0049] In the formula, α is the optical path collimation coefficient; x i x represents the actual light spot coordinates; i0 For the ideal light spot coordinates; w i ε is the spatial dimension weighting coefficient; β is the detector response coefficient; γ is the light intensity distribution non-uniformity coefficient; ε1 is the measurement error term.

[0050] 4. Detector response equation:

[0051]

[0052] In the formula, β is the detector response coefficient; V out V is the output voltage. in λ is the input operating voltage; θ is the temperature coefficient; T is the temperature deviation; λ is the attenuation coefficient; t is the operating time; ε2 is the system noise term.

[0053] 5. Signal crosstalk equation:

[0054]

[0055] In the formula, κ is the signal crosstalk coefficient; V ij V represents the interference voltage of channel j on channel i. ii β is the main signal voltage of channel i; i f(Δφ) represents the detector response coefficient for channel i. ij ) is the filter position deviation function; ε3 is the random error term.

[0056] 6. System gain equation:

[0057]

[0058] In the formula, G opt α is the optimal gain value; S is the signal count rate; D is the dark count rate; α is the optical path collimation coefficient; β is the detector response coefficient; μ is the system stability coefficient; t is the response time; ε4 is the gain error term.

[0059] 7. Laser position adjustment vector:

[0060]

[0061] In the formula, The adjusted position vector; α is the initial position vector; α is the optical path collimation coefficient; This is the positional deviation vector; This is the position error vector.

[0062] 8. Detector operating voltage optimization equation:

[0063]

[0064] In the formula, V opt V0 is the optimal operating voltage; β is the current response coefficient; β0 is the standard response coefficient; η is the voltage stability coefficient; t is the operating time; ε6 is the voltage error term.

[0065] 9. Filter position optimization equation:

[0066]

[0067] In the formula, The optimized position vector; κ is the initial position vector; κ is the signal crosstalk coefficient. This is the angle rotation matrix; This is the position error vector.

[0068] 10. Photomultiplier tube voltage distribution equation:

[0069]

[0070] In the formula, V i G is the voltage of the i-th stage multiplier; opt The optimal gain value; k i ρ is the partial pressure coefficient; i ε is the gain nonlinearity coefficient; S is the signal counting rate; D is the dark counting rate; ε8 is the voltage error term.

[0071] The following explains how to obtain each parameter and the principle behind equation construction:

[0072] 1. Method for obtaining parameters in the optical path detection channel conversion cost matrix:

[0073] Mechanical motion time function t m (d ij Obtain it through the following steps:

[0074] Step 1: Use a high-precision displacement sensor to measure the actual physical distance d between each detection channel. ij ;

[0075] Step 2: Record the mechanical motion time at different distances and fit the function to obtain the relationship:

[0076]

[0077] In the formula, a1, a2, and a3 are fitting coefficients, where a1 ranges from 0.1 to 0.5 s / mm, a2 ranges from 0.05 to 0.2 s / mm, and a3 ranges from 0.01 to 0.1 s.

[0078] Signal settling time function t s(p ij How to obtain )

[0079] Step 1: Record the time required for the signal to stabilize under different optical path parameters;

[0080] Step 2: Establish the functional relationship:

[0081] t s (p ih )=b1Δλ ij +b2Δθ ij +b3;

[0082] In the formula, Δλ ij For wavelength difference, Δθ ij The difference in optical path angle is represented by b1, b2, and b3, which are fitting coefficients.

[0083] 2. Obtaining parameters and explaining the principle of optical path collimation equation:

[0084] Spatial dimension weight coefficient w i Obtained through algorithm optimization:

[0085]

[0086] In the formula, I k Let I be the measured light intensity at the k-th test point. k0 Let m be the ideal light intensity, and m be the number of test points.

[0087] Calculation of the light intensity distribution non-uniformity coefficient γ:

[0088]

[0089] In the formula, σ I The standard deviation of light intensity This represents the average light intensity.

[0090] The equation uses weighted Euclidean distance to describe the spot deviation, introduces a light intensity distribution term to consider the energy distribution characteristics, and an exponential term to reflect the system stability, thus constructing a complete optical path collimation evaluation model.

[0091] 3. Explanation of the principle of the detector response equation:

[0092] Methods for measuring the temperature coefficient θ:

[0093] Step 1: Record the detector output at different temperatures;

[0094] Step 2: Obtain the temperature coefficient through linear fitting:

[0095]

[0096] Determination of the attenuation coefficient λ:

[0097]

[0098] In the formula, V out (t) represents the output voltage at time t, V out0 This is the initial output voltage.

[0099] This equation takes into account temperature effects and time decay characteristics, uses an exponential relationship to describe response decay, and a ratio relationship to characterize gain characteristics, thus constructing an accurate detector response model.

[0100] 4. Filter position deviation function in the signal crosstalk equation:

[0101]

[0102] In the formula, k is the steepness coefficient and φ0 is the critical deviation angle.

[0103] Parameter acquisition method:

[0104] Step 1: Measure the crosstalk signal strength at different deviation angles;

[0105] Step 2: Use the Logistic function to fit and obtain the parameters k and φ0.

[0106] The equation uses nonlinear superposition to describe the multi-channel crosstalk effect and uses the Sigmoid function to characterize the effect of position deviation, thus achieving accurate crosstalk assessment.

[0107] 5. Obtaining and Principles of System Gain Equation Parameters:

[0108] Measurement of the system stability coefficient μ:

[0109]

[0110] In the formula, G(t) is the gain value at time t, G ∞ This is the steady-state gain value.

[0111] The equation uses a logarithmic relationship to balance the signal-to-noise ratio and an exponential term to describe the stabilization process, thus achieving adaptive calculation of the optimal gain value.

[0112] 6. Parameter acquisition and principle in laser position adjustment vector:

[0113] Position deviation vector The calculation process:

[0114]

[0115] In the formula, (x t ,y t ,z t(x0, y0, z0) represents the target position coordinates, and (x0, y0, z0) represents the current position coordinates.

[0116] Position error vector Expression:

[0117]

[0118] Each component follows a normal distribution:

[0119] This vector equation uses a rectangular coordinate system to describe spatial position, takes into account system positioning errors, and achieves high-precision spatial position adjustment.

[0120] 7. Obtaining parameters for the detector operating voltage optimization equation:

[0121] Determination of nominal operating voltage V0:

[0122] Step 1: Measure the detector response curve under standard conditions;

[0123] Step 2: Calculate the voltage corresponding to the maximum signal-to-noise ratio:

[0124]

[0125] In the formula, S(V) is the signal strength and N(V) is the noise strength.

[0126] Measurement of voltage stability coefficient η:

[0127]

[0128] In the formula, V(t) is the voltage value at time t, V ∞ This is the steady-state voltage value.

[0129] This equation uses an exponential relationship to describe the voltage stabilization process and a ratio relationship to characterize the response characteristics, thus achieving optimized regulation of the operating voltage.

[0130] 8. Detailed explanation of the parameters in the filter position optimization equation:

[0131] Angle rotation matrix Expression:

[0132]

[0133] In the formula, θ is the rotation angle.

[0134] Position error vector Acquisition:

[0135] Statistical results obtained from repeated location experiments:

[0136]

[0137] In the formula, ∑ is the covariance matrix:

[0138]

[0139] Where ρ is the correlation coefficient.

[0140] This equation, combining rotational transformation and error analysis, enables precise optimization of the filter position.

[0141] 9. Obtaining parameters for the voltage distribution equation of a photomultiplier tube:

[0142] Pressure dividing coefficient k i Determination:

[0143] Step 1: Determine the initial voltage division ratio based on the characteristics of the dynode structure;

[0144] Step 2: Optimize the voltage divider coefficient:

[0145]

[0146] In the formula, k i0 Let ξ be the initial partial pressure coefficient. i is the correction factor, and n is the doubling series.

[0147] Gain nonlinear coefficient ρ i Measurement:

[0148]

[0149] In the formula, G i This represents the gain for the i-th level.

[0150] This equation takes into account the nonlinear characteristics of the gain and uses a logarithmic relationship to optimize the voltage division ratio, thus achieving efficient voltage distribution.

[0151] 10. Principle of Overall Optimization Strategy:

[0152] The coupling relationships between the equations are achieved through the following iterative optimization:

[0153]

[0154] In the formula, k is the number of iterations, and f1, f2, f3, f4 are the functional forms of the corresponding equations.

[0155] This optimization strategy enables the coordinated adjustment of parameters, ensuring the overall optimal performance of the system.

[0156] Compared with existing technologies, the optical path calibration method for flow cytometers provided by this invention first collects raw optical path detection data, including laser spot position coordinates, detector output voltage, dark count rate, and signal count rate, to establish a cost matrix describing the switching time overhead between different detection channels. Then, a traveling salesman problem optimization algorithm is applied to calculate the optimal optical path detection sequence that minimizes the total time overhead. Next, based on this optimal detection sequence, key parameters of the optical path system, such as laser position, filter angle, and detector operating voltage, are adjusted sequentially, and optical path calibration data of standard fluorescent microspheres are collected. Finally, by establishing and solving a set of optical path parameter optimization equations, including optical path collimation, detector response, signal crosstalk, and system gain, the optimal operating state of each component is obtained, and the optical path system is precisely adjusted accordingly to ensure that the entire system reaches its optimal performance level.

[0157] Compared with existing empirical adjustment and local optimization methods, the optical path calibration method proposed in this invention has the following advantages:

[0158] 1) A mathematical model describing the characteristics of each component of the optical path system was established, which can comprehensively analyze the system state and provide a basis for overall optimization.

[0159] 2) The optimal optical path detection sequence is calculated by using algorithm optimization, which greatly improves calibration efficiency and repeatability.

[0160] 3) Data acquisition and parameter optimization are performed based on standard fluorescent microspheres to ensure the reliability and objectivity of the calibration process.

[0161] 4) Coordinated adjustments were made to core components such as lasers, detectors, and filters to optimize the overall performance of the optical path system.

[0162] In summary, the optical path calibration method proposed in this invention fully integrates optical modeling, algorithm optimization, and systems engineering techniques, which can significantly improve the detection performance and operational stability of flow cytometers. It solves the problem that the optical path calibration of existing flow cytometers mainly relies on the operator's experience, resulting in low efficiency and difficulty in achieving overall system-level optimization. Attached Figure Description

[0163] Figure 1 A flowchart of the method provided by the present invention;

[0164] Figure 2 This is a diagram showing the laser spot position distribution in Example 2;

[0165] Figure 3 This is a diagram of the optical path detection channel conversion network in Example 2;

[0166] Figure 4This is a three-dimensional diagram of the detector response characteristics in Example 2;

[0167] Figure 5 This is a graph showing the system gain optimization in Example 2. Detailed Implementation

[0168] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings.

[0169] like Figure 1 The diagram shown is a flowchart of a flow cytometer optical path calibration method provided by the present invention. The specific implementation of each step is described in detail below:

[0170] The specific implementation of step S10 involves acquiring raw optical path detection data from standard fluorescent microspheres. First, a high-precision photodetector is used to acquire the horizontal and vertical coordinates of the laser spot. This step accurately measures the spatial distribution of the laser spot within the optical path system. Second, a digital voltmeter is used to measure the detector's output voltage at different operating voltages, forming a detector voltage output data sequence. This step allows for the acquisition of the detector's response characteristics. Third, the detector is set to dark-field detection mode, and the dark count rate data per unit time is recorded. This step allows for the acquisition of the detector's inherent noise level. Finally, the standard fluorescent microspheres are placed in the detection area, the detector is set to normal detection mode, and the signal count rate data per unit time is recorded. This step allows for the acquisition of the detector's actual response to the fluorescence signal. After completing the above data acquisition, statistical averaging and variance analysis are performed on the detector output voltage values, and outliers are removed to obtain stable voltage output data. This step ensures the reliability and repeatability of the data. Through these steps, key detection parameters of the flow cytometer's optical path system can be obtained, laying the foundation for subsequent optical path calibration.

[0171] The specific implementation of step S20 involves establishing an optical path detection channel switching cost matrix. First, the physical distance data between each detection channel is acquired, and a channel distance matrix is ​​established. This step describes the spatial relationship between different detection channels. Second, the mechanical motion characteristics of the optical elements in each detection channel are measured, including acceleration time, uniform motion time, and deceleration time, generating a mechanical motion time matrix. This step reflects the switching capability of each component in the optical path system. Third, for different combinations of detection channels, the time required for the signal to stabilize from instability is measured, forming a signal stabilization time matrix. This step describes the signal re-stabilization process during switching between detection channels. Next, the inherent response delay time of the system control unit is acquired, and a system delay matrix is ​​established. This step reflects the inherent performance limitations of the system itself. Finally, the above four time cost matrices are weighted and superimposed to generate the optical path detection channel switching cost matrix. This step comprehensively considers multiple factors in the optical path system, providing a basis for subsequent optimal detection sequence calculation. By establishing the cost matrix, the time consumption characteristics of switching between detection channels can be comprehensively described, laying the foundation for optimizing optical path calibration.

[0172] The specific implementation of step S30 is to calculate the optimal optical path detection sequence based on the Traveling Salesman Problem optimization algorithm. First, a directed graph data structure is constructed for the aforementioned cost matrix, with detection channels as nodes and channel transition times as edge weights. This step transforms the detection channel transition problem in the optical path system into a graph theory problem. Second, based on the sequential dependencies of scattered light detection, fluorescence detection, and data acquisition, the directed graph is divided into multiple levels. This step fully utilizes the functional characteristics of the optical path system and simplifies the calculation process. Next, a dynamic programming algorithm is used to calculate the shortest path between nodes in the graph, generating an initial state transition sequence. This step ensures the detection sequence with the minimum total time cost is obtained. Then, the initial state transition sequence is constrained to ensure that each detection channel is accessed only once. This step can meet the specific requirements of practical applications. Finally, an optical path component adjustment instruction set is generated based on the state transition sequence, forming the optimal optical path detection sequence. This step directly converts the calculation results into executable control commands. Through the above steps, an optimal optical path detection sequence that balances time cost and functional requirements can be obtained, providing a basis for subsequent optical path calibration work.

[0173] The specific implementation of step S40 involves adjusting the optical path system according to the optimal optical path detection sequence. First, based on the optimal optical path detection sequence, a high-precision displacement platform is controlled to adjust the laser's position in three-dimensional space. This step ensures that the laser spot is precisely focused on the sample area. Second, based on the optimal optical path detection sequence, the filter wheel mechanism is controlled to adjust the filter's angle and operating state. This step ensures that the filter efficiently blocks interfering light. Third, based on the optimal optical path detection sequence, the detector support is controlled to adjust the detector's spatial position and incident angle. This step ensures that the detector accurately receives the signal light from the sample. Then, after each adjustment, standard fluorescent microspheres are used for optical path detection, recording the laser spot position coordinates, detector output voltage, dark count rate, and signal count rate. This step obtains the actual performance data of the adjusted optical path system. Finally, the optical path detection data is analyzed in real time to determine whether a stable state has been reached; if not, the process returns to the previous step for further adjustment. This step ensures that the optical path system reaches its optimal operating state. By following the steps above, the components of the optical path system can be adjusted according to the pre-calculated optimal sequence, and the system performance indicators can be monitored in real time to ensure the effectiveness of the optical path calibration work.

[0174] The specific implementation of step S50 involves calculating the optical path calibration coefficient using the optical path parameter optimization equations. First, based on the optical path collimation equation, the laser spot position coordinates are substituted into the calculation to obtain the optical path collimation coefficient. This step quantifies the degree of deviation between the laser spot and the ideal position. Second, based on the detector response equation, the detector output voltage value and the optical path collimation coefficient are substituted into the calculation to obtain the detector response coefficient. This step evaluates the detector's sensitivity characteristics. Third, based on the signal crosstalk equation, the detector output voltage value, the detector response coefficient, and the filter position are substituted into the calculation to obtain the signal crosstalk coefficient. This step quantifies the degree of signal leakage between detection channels. Next, based on the system gain equation, the dark count rate, the signal count rate, the optical path collimation coefficient, and the detector response coefficient are substituted into the calculation to obtain the optimal gain value. This step determines the optimal operating gain level of the system. Finally, the calculated coefficients are numerically verified to ensure they meet the system performance requirements. This step ensures the coordination between the parameters. Through the above steps, key calibration parameters such as optical path collimation coefficient, detector response coefficient, signal crosstalk coefficient, and optimal gain value can be obtained using the optical path parameter optimization equations, providing a basis for subsequent optical path adjustment.

[0175] The specific implementation of step S60 involves adjusting the optical path system according to the optical path calibration parameters. First, based on the optical path collimation coefficient, the spatial position deviation vector of the laser is calculated, and the three-dimensional displacement platform is controlled for position compensation. This step ensures that the laser spot can be accurately focused on the sample area. Second, based on the detector response coefficient, the optimal operating voltage of the detector is calculated, and the high-voltage power supply module is controlled for voltage adjustment. This step ensures the detector operates at its optimal performance state. Third, based on the signal crosstalk coefficient, the optimal angular position of the filter is calculated, and the precision angle adjustment mechanism is controlled for position compensation. This step minimizes signal interference between detection channels. Next, based on the optimal gain value, the voltage distribution of each stage of the photomultiplier tube's dynodes is calculated, and the high-voltage divider network is controlled for voltage setting. This step ensures the system operates at optimal sensitivity and dynamic range. Finally, after all adjustments are completed, a system performance verification test is performed using standard fluorescent microspheres, and the final optical path calibration parameters are recorded. This step ensures that the calibration results meet the usage requirements. Through the above steps, key optical path system components such as the laser position, detector operating voltage, filter angle, and photomultiplier tube voltage can be adjusted according to the aforementioned optical path calibration parameters, ultimately achieving high-precision calibration of the flow cytometer's optical path.

[0176] The solution process for the optical path parameter optimization equations is as follows: First, a set of nonlinear equations is constructed, including the optical path collimation equation, detector response equation, signal crosstalk equation, and system gain equation. These equations are coupled, enabling the overall optimization of various parameters of the optical path system. Second, an iterative optimization algorithm is used to numerically solve the equations, setting initial values ​​and convergence conditions. In each iteration, the parameters of the equations are updated, the objective function value and gradient are calculated, and the optimization direction is adjusted based on the gradient information until the convergence condition or the maximum number of iterations is reached. Finally, the final optimization results are output, including the optical path collimation coefficient, detector response coefficient, signal crosstalk coefficient, and optimal gain value. By establishing the optical path parameter optimization equations and solving them using an iterative optimization algorithm, it is possible to ensure that the core parameters of the optical path system reach their optimal state, providing a basis for subsequent optical path adjustments.

[0177] The process of obtaining variables in the optical path parameter optimization equations is as follows: First, a high-precision light intensity distribution analyzer is used to measure the energy density distribution of the laser spot to obtain the light intensity distribution non-uniformity coefficient. This step describes the characteristics of the laser source. Second, a temperature control system is used to measure the detector response characteristics under different temperature conditions to obtain the detector temperature coefficient. This step reflects the temperature dependence of the detector performance. Third, a long-term stability testing system is used to record the changes in the detector output signal over time to obtain the detector attenuation coefficient. This step reflects the time stability of the detector performance. Next, a multi-channel crosstalk testing system is used to measure the signal crosstalk degree of different detection channel combinations to obtain the signal crosstalk correction coefficient. This step reflects the signal coupling characteristics in the optical path system. Finally, a system stability evaluation module is used to measure the changes in system parameters over time to obtain the system stability coefficient. This step describes the long-term operating characteristics of the entire optical path system. Through the above steps, the variables required in the optical path parameter optimization equations can be obtained, providing basic data support for the establishment and solution of the equations.

[0178] The optimization process of the optical path detection channel switching cost matrix is ​​as follows: First, a multi-dimensional data model is established, including the physical distance, mechanical characteristics, and signal characteristics of the channels. This step comprehensively describes the various time overhead factors of detection channel switching in the optical path system. Second, time series analysis is used to process the delay data during the detection channel switching process. This step can statistically characterize the changing characteristics of various time overheads. Third, an adaptive weighting algorithm is used to dynamically weight different types of time overheads. This step can flexibly adjust the relative importance of each factor according to actual application requirements. Next, matrix factorization algorithms are used to perform eigenvalue analysis and singular value decomposition on the cost matrix. This step can uncover the potential patterns contained in the cost matrix. Finally, the optimal detection channel switching strategy is generated based on the matrix optimization results. This step can support the subsequent calculation of the optimal optical path detection sequence. Through the above steps, a dynamically adaptive optical path detection channel switching cost matrix model can be established, providing an efficient and reliable computational foundation for optical path calibration.

[0179] Specifically, the principle of this invention is:

[0180] First, a cost matrix C describing the time overhead is established for the switching process between different detection channels in the optical path system. Each element c of this matrix... ij This represents the total time cost required to switch from detection channel i to detection channel j, including the mechanical movement time t. m (d ij Signal settling time t s (p ij ) and the system's inherent delay time t dFactors such as these can be considered. By establishing such a comprehensive time cost model, the coupling characteristics between various components in the optical path system can be described more accurately, providing a basis for subsequent calculation of the optimal detection sequence.

[0181] Secondly, a dynamic programming algorithm based on the Traveling Salesman Problem is used to optimize the cost matrix, resulting in the optical path detection sequence s with the minimum total time cost. Specifically, each detection channel is treated as a node in a graph theory, with the weights between nodes representing the switching time cost. Then, based on the sequential dependencies of scattered light detection, fluorescence detection, and data acquisition, the directed graph is divided into multiple levels. Next, the dynamic programming algorithm is used to calculate the shortest path between nodes, ensuring that each channel is visited only once. Finally, adjustment commands for laser position, filter angle, and detector operating voltage are generated based on the obtained state transition sequence. This global optimization-based method effectively reduces the time cost of the optical path system during multi-channel detection, laying the foundation for subsequent accurate calibration.

[0182] Furthermore, by collecting optical path detection data of standard fluorescent microspheres under different adjustment states, an optimization equation set was established, including the optical path collimation equation, detector response equation, signal crosstalk equation, and system gain equation. These equations are coupled and can describe the optical characteristics of core components such as the laser, detector, and filter, and their mutual influences. Taking the optical path collimation equation as an example, it uses parameters such as the laser spot coordinates {x1, x2}, the detector response coefficient β, and the light intensity distribution non-uniformity γ to calculate the degree of optical path deviation α; based on the detector response equation, the output voltage V can be used to calculate the optical path deviation. out Operating voltage V in Factors such as temperature characteristics are considered to evaluate the detector's sensitivity β; the signal crosstalk equation describes the degree of mutual interference between different detection channels κ, which is related to the detector response β and the filter position Δφ; finally, the system gain equation, combined with the dark count rate D, signal count rate S, collimation α, and response β, determines the optimal operating gain G. opt By solving this set of equations, the optimal operating state of each core component of the optical path system can be obtained, providing a basis for subsequent precise adjustments.

[0183] Finally, based on the key calibration parameters obtained from the above optimization calculations, the laser position, detector operating voltage, filter angle, and photomultiplier tube voltage were precisely adjusted. Specifically, the laser position deviation vector was calculated using the optical path collimation coefficient α. Compensation is performed using a high-precision three-dimensional displacement platform; the optimal operating voltage V is determined based on the detector's response coefficient β. opt A high-voltage power supply module is used for precise adjustment; the optimal angle position of the filter is calculated based on the signal crosstalk coefficient χ. Compensation is performed using a precision angle adjustment mechanism; finally, based on the optimal gain value G... opt Setting the voltage distribution V of each stage of the photomultiplier tube's dynamo electrode i This ensures the system operates at optimal sensitivity and dynamic range. Through the aforementioned series of coordinated adjustments, the entire optical path system can be ensured to achieve optimal performance, meeting the stringent requirements of flow cytometers for detection sensitivity, resolution, and linear range.

[0184] In summary, the flow cytometer optical path calibration method proposed in this invention fully integrates optical modeling, algorithm optimization, and systems engineering techniques, enabling overall performance optimization of each core component of the optical path system. This method first establishes a cost matrix model describing the switching time overhead between detection channels and calculates the optimal detection sequence using a dynamic programming algorithm. Then, based on the optical path detection data of standard fluorescent microspheres, it establishes a set of optimization equations including optical path collimation, detector response, signal crosstalk, and system gain, and obtains the optimal operating state of each component through numerical solution. Finally, based on the optimization results, parameters such as laser position, detector operating voltage, filter angle, and photomultiplier tube voltage are precisely adjusted to ensure the entire optical path system reaches its optimal performance level. Compared to existing empirical adjustment and local optimization methods, the method of this invention has a high degree of automation, optimization effect, and repeatability, and can effectively solve the performance degradation problem of flow cytometer optical path systems during long-term use.

[0185] The following is a specific embodiment 1 of the present invention. Each step in this embodiment 1 is described in detail below: Step S10 involves acquiring the raw optical path detection data of the standard fluorescent microspheres. First, a high-precision photodetector is used to acquire the horizontal position coordinates x1 and vertical position coordinates x2 of the laser spot. This step can accurately measure the spatial distribution of the laser spot in the optical path system. Second, a digital voltmeter is used to measure the detector at different operating voltages V. in Output voltage V out First, a sequence of detector voltage output data is generated. This step allows us to obtain the detector's response characteristics. Next, the detector is set to dark-field detection mode, and the dark count rate D per unit time is recorded. This step allows us to obtain the detector's inherent noise level. Finally, standard fluorescent microspheres are placed in the detection area, the detector is set to normal detection mode, and the signal count rate S per unit time is recorded. This step allows us to obtain the detector's actual response to the fluorescence signal. After completing the above data acquisition, statistical averaging and variance analysis are performed on the detector output voltage values ​​to remove outliers and obtain stable voltage output data. This step ensures the reliability and repeatability of the data. Through these steps, key detection parameters of the flow cytometer optical path system can be obtained, laying the foundation for subsequent optical path calibration.

[0186] The specific implementation of step S20 is to establish the optical path detection channel conversion cost matrix. This cost matrix C can be expressed as:

[0187]

[0188] Where, c ij This represents the total time cost required to switch from detection channel i to detection channel j, where n is the total number of detection channels. Each element c... ij The calculation formula is:

[0189] c ij =t m (d ij )+t s (p ij )+t d +δ ij ;

[0190] Where, t m (d ij ) is a function of mechanical motion time, and is related to the distance d between channels. ij Related; t s (p ij ) is a function of signal settling time, related to optical path parameter p. ij Related; t d δ is the inherent delay time of the system. ij The random error term follows a normal distribution. First, the physical spacing data of each detection channel is acquired, and a channel distance matrix is ​​established. This step describes the spatial relationship between different detection channels. Second, the mechanical motion characteristics of the optical elements in each detection channel are measured, including acceleration time, uniform motion time, and deceleration time, generating a mechanical motion time matrix. This step reflects the switching capability of each component in the optical path system. Third, for different combinations of detection channels, the time required for the signal to stabilize from instability is measured, forming a signal stabilization time matrix. This step describes the signal re-stabilization process when switching between detection channels. Next, the inherent response delay time of the system control unit is acquired, and a system delay matrix is ​​established. This step reflects the inherent performance limitations of the system itself. Finally, the above four time cost matrices are weighted and superimposed to generate an optical path detection channel switching cost matrix. This step comprehensively considers multiple factors in the optical path system, providing a basis for subsequent optimal detection sequence calculation. By establishing the cost matrix, the time consumption characteristics of switching between detection channels can be comprehensively described, laying the foundation for optimizing optical path calibration.

[0191] The specific implementation of step S30 is to calculate the optimal optical path detection sequence based on the Traveling Salesman Problem optimization algorithm. First, a directed graph data structure is constructed for the aforementioned cost matrix C, with the detection channels as nodes and the transition times between channels as edge weights. This step transforms the detection channel transition problem in the optical path system into a graph theory problem. Second, based on the sequential dependencies of scattered light detection, fluorescence detection, and data acquisition, the directed graph is divided into multiple levels. This step fully utilizes the functional characteristics of the optical path system and simplifies the calculation process. Next, a dynamic programming algorithm is used to calculate the shortest path between nodes in the graph, generating an initial state transition sequence s = {s1, s2, ..., s...}. n This step ensures the detection sequence with the minimum total time overhead. Then, the initial state transition sequence is constrained to ensure each detection channel is accessed only once. This step meets specific requirements in practical applications. Finally, an optical path element adjustment instruction set is generated based on the state transition sequence, forming the optimal optical path detection sequence. This step directly converts the calculation results into executable control commands. Through the above steps, an optimal optical path detection sequence that balances time overhead and functional requirements can be obtained, providing a basis for subsequent optical path calibration. The optimization objective function of the optimal optical path detection sequence is:

[0192]

[0193] The constraint is: s k ∈{1,2,…,n} and are all distinct.

[0194] The specific implementation of step S40 involves adjusting the optical path system according to the optimal optical path detection sequence. First, based on the optimal optical path detection sequence, a high-precision displacement platform is controlled to adjust the laser's position in three-dimensional space. This step ensures that the laser spot is precisely focused on the sample area. Second, based on the optimal optical path detection sequence, the filter wheel mechanism is controlled to adjust the filter's angle and operating state. This step ensures that the filter efficiently blocks interfering light. Third, based on the optimal optical path detection sequence, the detector support is controlled to adjust the detector's spatial position and incident angle. This step ensures that the detector accurately receives the signal light from the sample. Then, after each adjustment, standard fluorescent microspheres are used for optical path detection, recording the laser spot position coordinates {x1, x2} and the detector output voltage value V. outThe steps involve calculating the dark count rate (D) and the signal count rate (S). This step obtains the actual performance data of the adjusted optical path system. Finally, the optical path detection data is analyzed in real time to determine whether a stable state has been reached. If not, the process returns to the previous step for further adjustment. This step ensures that the optical path system reaches its optimal operating state. Through these steps, the various components of the optical path system can be adjusted according to a pre-calculated optimal sequence, and system performance indicators can be monitored in real time, ensuring the effectiveness of the optical path calibration.

[0195] The specific implementation of step S50 involves calculating the optical path calibration coefficients using the optical path parameter optimization equations. First, based on the optical path collimation equations:

[0196]

[0197] Substituting the laser spot position coordinates {x1, x2} into the calculation, the optical path collimation coefficient α is obtained. This step can quantify the degree of deviation between the laser spot and the ideal position. Secondly, based on the detector response equation:

[0198]

[0199] The detector output voltage value V out Substituting the collimation coefficient α into the calculation, the detector response coefficient β is obtained. This step can evaluate the detector's sensitivity characteristics. Again, based on the signal crosstalk equation:

[0200]

[0201] The detector output voltage value V out The detector response coefficient β and the filter position Δφ ij Substituting the values ​​into the calculation, we obtain the signal crosstalk coefficient κ. This step quantifies the degree of signal leakage between detection channels. Next, based on the system gain equation:

[0202]

[0203] Substituting the dark count rate D, the signal count rate S, the optical path collimation coefficient α, and the detector response coefficient β into the calculation, the optimal gain value G is obtained. opt This step determines the optimal operating gain level of the system. Finally, the calculated coefficients are numerically verified to ensure they meet the system performance requirements. This step ensures the coordination between the parameters. Through the above steps, the optical path collimation coefficient α, detector response coefficient β, signal crosstalk coefficient κ, and optimal gain value G can be obtained using the optical path parameter optimization equations. opt These key calibration parameters provide a basis for subsequent optical path adjustments.

[0204] The specific implementation of step S60 involves adjusting the optical path system according to the optical path calibration parameters. First, the spatial position deviation vector of the laser is calculated based on the optical path collimation coefficient α. Position compensation is performed by controlling a three-dimensional displacement platform.

[0205]

[0206] This step ensures that the laser spot is precisely focused on the sample area. Next, based on the detector response coefficient β, the optimal operating voltage V of the detector is calculated. opt Voltage regulation is achieved by controlling the high-voltage power supply module.

[0207]

[0208] This step ensures the detector operates at its optimal performance. Next, based on the signal crosstalk coefficient κ, the optimal angular position of the filter is calculated. Position compensation is achieved by controlling a precision angle adjustment mechanism.

[0209]

[0210] This step can suppress signal interference between detection channels to the greatest extent. Next, according to the optimal gain value G... opt Calculate the voltage distribution V of each stage of the photomultiplier tube. i The voltage is set by controlling the high-voltage divider network.

[0211]

[0212] This step ensures the system operates at optimal sensitivity and dynamic range. Finally, after all adjustments are completed, system performance is verified using standard fluorescent microspheres, and the final optical path calibration parameters are recorded. This step ensures the calibration results meet usage requirements. Through these steps, key optical path system components such as laser position, detector operating voltage, filter angle, and photomultiplier tube voltage can be adjusted according to the aforementioned optical path calibration parameters, ultimately achieving high-precision calibration of the flow cytometer's optical path.

[0213] To further understand and implement this invention, a specific application scenario is provided below as Example 2: An engineering team is performing optical path calibration on a certain model of flow cytometer. Before starting the calibration, the engineering team thoroughly reviewed the instrument's usage history and current status. It is understood that this flow cytometer has been used continuously for three years, primarily for lymphocyte subset analysis and tumor cell phenotype detection. The daily detection workload is substantial, with the instrument averaging about 8 hours of operation per day. Recently, researchers at the institute reported a decrease in the instrument's detection sensitivity and resolution, as well as some repeatability issues, thus necessitating a comprehensive calibration.

[0214] After a detailed inspection of the instrument, the engineering team determined the specific content and requirements for this optical path calibration:

[0215] 1. Calibration range: including the spot position and intensity distribution of the two lasers, the response characteristics and dark noise level of the four photodetectors, and the spectral characteristics and attenuation ratio of the three sets of filters, etc.

[0216] 2. Calibration objectives: Ensure that the laser spot diameter is less than 20μm, the detector signal-to-noise ratio is greater than 1000, the signal crosstalk is less than 5%, and the system sensitivity meets the requirement of CV≤3%.

[0217] 3. Calibration method: Standardized fluorescent microspheres are used as the optical path detection source, and data acquisition and signal analysis software are used to comprehensively test and evaluate the instrument's performance indicators.

[0218] 4. Calibration process: First, each subsystem is individually debugged and optimized, and then the entire optical path system is coordinated and adjusted to ensure that the working state of each component reaches the best match.

[0219] 5. Calibration time: This optical path calibration is expected to take half a day to complete, and the calibration results will be submitted in the form of a written report.

[0220] Based on the above requirements, the engineering team developed a detailed optical path calibration plan and began its implementation. The entire process consisted of the following steps:

[0221] 1. Acquire raw data from optical path detection

[0222] First, the engineering team placed a box of standardized polystyrene fluorescent microspheres in the sample channel area. These microspheres have a highly uniform particle size distribution and stable fluorescence intensity characteristics, and are widely used in the routine performance verification and quantitative calibration of flow cytometers. Using a high-precision CCD camera, the engineering team accurately measured the center coordinates of the two laser spots in the X and Y planes, as shown in Table 1.

[0223] Table 1. Coordinates of laser spot position

[0224] laser X coordinate Y coordinate 488nm 125.3 234.6 637nm 210.5 301.8

[0225] Figure 2 This is a laser spot position distribution diagram, showing the initial positions and calibrated displacement directions of the 488nm and 637nm laser spots in the XY plane. The blue circles represent the 488nm laser spot, and the red circles represent the 637nm laser spot. The arrows indicate the direction and distance that need to be adjusted.

[0226] Next, the engineering team used digital voltmeters to measure the output voltage values ​​of the four photodetectors under different operating voltages, and recorded the dark count rate and signal count rate per unit time. Specific data are shown in Table 2.

[0227] Table 2 Detector Performance Data

[0228] detector Operating voltage / V Output voltage / V Dark count rate / cps Signal count rate / cps PMT-1 800 1.26 45 12458 PMT-2 850 1.38 52 11927 APD-1 280 1.04 38 9742 APD-2 300 1.15 42 10306

[0229] The above data represents the initial test results. Any abnormal data will be verified and optimized in the future.

[0230] 2. Establish the optical path detection channel conversion cost matrix

[0231] Based on the optical path detection data collected above, the engineering team began to establish a cost matrix describing the switching time overhead between different detection channels. First, the engineering team measured the physical distances between the four detection channels, as shown in Table 3:

[0232] Table 3. Distance between detection channels

[0233] aisle PMT-1 PMT-2 APD-1 APD-2 PMT-1 0 58.2 73.5 81.6 PMT-2 58.2 0 52.8 67.3 APD-1 73.5 52.8 0 43.1 APD-2 81.6 67.3 43.1 0

[0234] Secondly, the engineering team tested the mechanical characteristics of each optical component, including acceleration time, constant speed time, and deceleration time, and established the corresponding time matrix. Taking the filter wheel as an example, switching from PMT-1 to PMT-2 requires 0.8 seconds of acceleration time, 1.2 seconds of constant speed time, and 0.5 seconds of deceleration time. Similarly, switching from APD-1 to APD-2 requires 0.6 seconds of acceleration time, 0.9 seconds of constant speed time, and 0.4 seconds of deceleration time. In addition, the inherent response delay time of the system control circuit is approximately 50ms.

[0235] Furthermore, for different combinations of detection channels, the engineering team measured the time required for the signal to stabilize from unstable to stable. For example, switching from PMT-1 to PMT-2 required approximately 0.3 seconds for the signal to stabilize; switching from APD-1 to APD-2 required 0.2 seconds.

[0236] Finally, the three types of time overhead data and the inherent system delay time are weighted and superimposed to generate the following optical path detection channel conversion cost matrix C:

[0237]

[0238] Where, matrix element c ij This represents the total time required to switch from detection channel i to detection channel j, in seconds.

[0239] 3. Calculate the optimal optical path detection sequence

[0240] Based on the cost matrix c established above, the engineering team used a dynamic programming algorithm to calculate the optimal optical path detection sequence that minimizes the total time cost. The specific steps are as follows:

[0241] First, each detection channel is treated as a node in a directed graph, and the edge weights between nodes represent the corresponding time costs. Considering the functional characteristics of the optical path system, the engineering team divided this directed graph into three levels:

[0242] First level - Scattered light detection channels: PMT-1, PMT-2;

[0243] Second-level - Fluorescence detection channels: APD-1, APD-2;

[0244] Third level - Data acquisition channel;

[0245] Then, the shortest path from the starting node to the ending node is calculated using a dynamic programming algorithm. The starting node corresponds to the initial state of each detection channel, and the ending node corresponds to the target state of each detection channel. The algorithm process is as follows:

[0246] 1) Starting from PMT-1 in the first level, calculate the shortest path to PMT-2 and obtain the total time cost c. 12 = 2.8 seconds;

[0247] 2) Return from PMT-2 to PMT-1, calculate the shortest path to the second-level APD-1, and obtain the total time cost c. 21 +c 1,3 =2.8 + 3.5 = 6.3 seconds;

[0248] 3) Calculate the shortest path from APD-1 to APD-2 to obtain the total time cost c. 3,4 = 2.0 seconds;

[0249] 4) Finally, the data is returned from APD-2 to the third-level data acquisition channel, with a total time cost of c. 4,3 +c 3,1 =2.0 + 3.5 = 5.5 seconds.

[0250] Based on the above calculations, the state transition sequence with the minimum total time cost is: PMT-1→PMT-2→APD-1→APD-2→PMT-1, with a total time cost of 2.8+6.3+2.0+5.5=16.6 seconds.

[0251] Figure 3 This is a network diagram of optical path detection channel switching, showing the optimal detection sequence from PMT-1 to the data acquisition system (DAQ). Nodes in the diagram represent different detectors, and the numbers on the sides indicate the time required for channel switching.

[0252] Finally, based on the obtained optimal detection sequence, the engineering team generated the following optical path component adjustment instructions:

[0253] 1) Move the 488nm laser 85.2μm along the positive X-axis and 92.4μm along the negative Y-axis;

[0254] 2) Move the 637nm laser 72.5μm along the negative X-axis and 65.3μm along the positive Y-axis;

[0255] 3) Adjust the operating voltage of PMT-1 to 850V and PMT-2 to 800V;

[0256] 4) Adjust the operating voltage of APD-1 to 320V and APD-2 to 280V;

[0257] 5) Rotate the diffuse light channel filter to the 0° position and the fluorescence channel filter to the 45° position.

[0258] Through the above adjustments, it can be ensured that the optical path system can efficiently complete the switching operation according to the pre-calculated optimal sequence each time the detection channel is switched.

[0259] 4. Acquire optical path calibration data

[0260] After completing the initial adjustments to the optical path system parameters, the engineering team began collecting optical path calibration data from the standard fluorescent microspheres. The team first placed a suspension of fluorescent microspheres with a concentration of 1×10^6 / mL in the sample area and then tested each detection channel sequentially according to the optimal detection sequence calculated earlier. Specific test data are shown in Table 4.

[0261] Table 4 Optical Path Calibration Data

[0262] Detection channel Laser spot center / μm Output voltage / V Dark count rate / cps Signal count rate / cps PMT-1 (125.1,234.8) 1.28 47 12573 PMT-2 (210.2,301.5) 1.35 50 12032 APD-1 (73.4,95.6) 1.07 40 9868 APD-2 (95.2,118.3) 1.18 43 10432

[0263] The data in the table above shows that the parameters such as the laser spot center coordinates, detector output voltage, dark count rate, and signal count rate in this test are basically consistent with the original data acquired initially, indicating that the optical system is operating relatively stably. However, a few parameters still have some deviations and require further optimization and adjustment.

[0264] Figure 4 This is a 3D plot of the detector's response characteristics, showing the PMT detector's response under different operating voltage and time conditions. In the plot, the X-axis represents the operating voltage, the Y-axis represents time, and the Z-axis represents the normalized detector response intensity. The color gradient reflects the changes in response intensity.

[0265] 5. Calculate optical path calibration parameters

[0266] Based on the optical path calibration data obtained above, the engineering team conducted an in-depth analysis of the system using the optical path parameter optimization equations. The specific process is as follows:

[0267] First, according to the optical path collimation equation:

[0268]

[0269] Where, x i x is the actual light spot coordinate. i0 For the ideal light spot coordinates, w i ε is the spatial dimension weighting coefficient, β is the detector response coefficient, γ is the light intensity distribution non-uniformity coefficient, and ε1 is the measurement error. Substituting the spot coordinate data in Table 4 into the calculation, the optical path collimation coefficients α are obtained as 0.85, 0.91, 0.78, and 0.83, respectively.

[0270] Secondly, according to the detector response equation:

[0271]

[0272] Among them, V out V is the output voltage. in Let θ be the operating voltage, θ be the temperature coefficient, T be the temperature deviation, λ be the attenuation coefficient, t be the operating time, and ε2 be the system noise. Substituting the voltage data from Table 4 and the aforementioned optical path collimation into the calculation, the detector response coefficients β are obtained as 0.92, 0.96, 0.88, and 0.91, respectively.

[0273] Secondly, according to the signal crosstalk equation:

[0274]

[0275] Among them, V ij V represents the interference voltage of channel j on channel i. iif(Δφ) is the main signal voltage of channel i. ij Let ε0 be the filter position deviation function, and ε3 be the random error. Substituting the voltage data, detector response coefficient, and filter position angle from Table 4 into the calculation, the signal crosstalk coefficients κ are obtained as 2.8%, 3.1%, 2.4%, and 2.6%, respectively.

[0276] Finally, according to the system gain equation:

[0277]

[0278] Among them, G opt Let S be the optimal gain value, D be the signal count rate, μ be the system stability coefficient, and ε₄ be the gain error. Substituting the data from Table 4, along with the aforementioned optical path collimation coefficient and detector response coefficient, into the calculation, the optimal gain value G can be obtained. opt The values ​​are 3200, 3400, 2800, and 3100 respectively.

[0279] Through the above steps, the engineering team obtained the key calibration parameters of the flow cytometer optical system, including the optical collimation coefficient α, the detector response coefficient β, the signal crosstalk coefficient κ, and the optimal gain value G. opt These parameters provide a basis for subsequent precise adjustments.

[0280] Figure 5 This is the system gain optimization curve, showing the signal-to-noise ratio (SNR) changes under different system gain values ​​(Gopt). The X-axis represents the system gain value, the Y-axis represents the SNR, and the curve reflects the nonlinear relationship between them.

[0281] 6. Adjust the optical path system parameters

[0282] Based on the optical path calibration parameters calculated above, the engineering team began to precisely adjust the optical path system of the XYZ-8000 flow cytometer. The specific steps are as follows:

[0283] First, based on the optical path collimation coefficient α, the position deviation vectors of the 488nm and 637nm lasers in the X and Y planes are calculated as follows:

[0284]

[0285] Then, the position compensation of the two lasers is performed by a high-precision three-dimensional displacement platform to ensure that the laser spot can be accurately focused on the sample area.

[0286] Secondly, based on the detector response coefficient β, the optimal operating voltages for PMT-1, PMT-2, APD-1, and APD-2 were calculated to be 860V, 790V, 310V, and 270V, respectively. Subsequently, a precision high-voltage power supply module was used to adjust the operating voltages of these four detectors to ensure they operate at their optimal sensitivity.

[0287] Next, based on the signal crosstalk coefficient k, the optimal angle position of the scattered light channel filter was calculated to be 0°, and the optimal angle position of the fluorescence channel filter was calculated to be 42°. Then, the filters were precisely positioned using a precise angle adjustment mechanism to suppress signal interference between different detection channels to the greatest extent possible.

[0288] Finally, based on the optimal gain value G opt The high-voltage divider networks of the PMT and APD are adjusted to ensure that they can amplify the signal within their optimal operating range. Specifically, the cascaded voltage distribution scheme for PMT-1 and PMT-2 is 800V-810V-820V-830V-840V-850V, and the cascaded voltage distribution scheme for APD-1 and APD-2 is 270V-280V-290V-300V-310V-320V.

[0289] Through a series of meticulous adjustments, the engineering team ultimately brought the optical path system of the XYZ-8000 flow cytometer to its optimal operating state. The team then used standard fluorescent microspheres to verify the calibration results, which showed that:

[0290] 1) The laser spot diameter is less than 18μm, meeting the requirement of ≤20μm;

[0291] 2) The signal-to-noise ratio of all four detectors exceeds 1200, reaching the specification of >1000;

[0292] 3) The signal crosstalk between different detection channels is less than 4%, which is better than the target of ≤5%;

[0293] 4) The CV values ​​of the system sensitivity test are all within 2.5%, which meets the requirement of CV≤3%.

[0294] In summary, after detailed optical path calibration and parameter optimization, the XYZ-8000 flow cytometer has significantly improved its performance indicators such as detection sensitivity, resolution, and linear range, meeting the actual application needs of the research institute.

[0295] It should be noted that the variables involved in this invention are explained in detail in Table 5 below.

[0296] Table 5. Variable Explanation Table

[0297]

[0298]

[0299] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for calibrating the optical path of a flow cytometer, characterized in that, Includes the following steps: S10. Collect the raw optical path detection data of the standard fluorescent microspheres. The raw optical path detection data includes the laser spot position coordinates, detector output voltage value, dark count rate, and signal count rate. S20. Establish an optical path detection channel conversion cost matrix, wherein the cost matrix represents the optical path adjustment time required for switching between different detection channels; S30. Calculate the optimal optical path detection sequence based on the Traveling Salesman Problem optimization algorithm; S40. Adjust the laser position, filter position, and detector position according to the optimal optical path detection sequence to obtain the optical path calibration data of the standard fluorescent microsphere; S50. Substitute the laser spot position coordinates, the detector output voltage value, the dark count rate, the signal count rate, and the filter position into a preset optical path parameter optimization equation set and solve it to obtain the optical path collimation coefficient, detector response coefficient, signal crosstalk coefficient, and optimal gain value. S60. Adjust the laser position according to the optical path collimation coefficient, adjust the detector operating voltage according to the detector response coefficient, adjust the filter position according to the signal crosstalk coefficient, and adjust the photomultiplier tube voltage according to the optimal gain value to complete the optical path calibration.

2. The flow cytometer optical path calibration method according to claim 1, characterized in that, Step S10 specifically includes: Step 101: Using standard fluorescent microspheres as the light source, the horizontal and vertical position coordinates of the laser spot are obtained through a high-precision photodetector; Step 102: Use a digital voltmeter to measure the output voltage value of the detector under different operating voltages to form a detector voltage output data sequence; Step 103: Set the detector to work in dark field detection mode and record the dark count rate data per unit time. Step 104: Place the standard fluorescent microspheres in the detection area, set the detector to normal detection mode, and record the signal count rate data per unit time. Step 105: Perform statistical averaging and variance analysis on the output voltage value of the detector, and obtain stable voltage output data after removing outliers.

3. The flow cytometer optical path calibration method according to claim 1, characterized in that, Step S20 specifically includes: Step 201: Obtain the physical spacing data of each detection channel and establish a channel distance matrix; Step 202: Measure the mechanical motion characteristics of the optical elements in each detection channel, including acceleration time, uniform motion time, and deceleration time, and generate a mechanical motion time matrix; Step 203: For different combinations of detection channels, measure the time required for the signal to go from unstable to stable, and form a signal stabilization time matrix; Step 204: Obtain the inherent response delay time of the system control unit and establish the system delay matrix; Step 205: Weight and superimpose the inter-channel distance matrix, the mechanical motion time matrix, the signal stabilization time matrix, and the system delay matrix to generate the optical path detection channel conversion cost matrix.

4. The flow cytometer optical path calibration method according to claim 1, characterized in that, Step S30 specifically includes: Step 301: Construct a directed graph data structure for the cost matrix, using the detection channels as nodes of the graph and the transition time between channels as the weight of the edges; Step 302: Based on the sequential dependency of scattered light detection, fluorescence detection, and data acquisition, the directed graph is divided into multiple levels; Step 303: Use dynamic programming algorithm to calculate the shortest path between nodes in the graph and generate an initial state transition sequence; Step 304: Perform constraint verification on the initial state transition sequence to ensure that each detection channel is accessed only once; Step 305: Generate an optical path element adjustment instruction set based on the state transition sequence to form an optimal optical path detection sequence.

5. The flow cytometer optical path calibration method according to claim 1, characterized in that, Step S40 specifically includes: Step 401: Based on the optimal optical path detection sequence, control the high-precision displacement platform to adjust the position of the laser in three-dimensional space; Step 402: Based on the optimal optical path detection sequence, control the filter wheel mechanism to adjust the angle position and working state of the filter; Step 403: Based on the optimal optical path detection sequence, control the detector support to adjust the spatial position and incident angle of the detector; Step 404: After each adjustment, use standard fluorescent microspheres to perform optical path detection and record the laser spot position coordinates, detector output voltage value, dark count rate and signal count rate. Step 405: Analyze the optical path detection data in real time to determine whether a stable state has been reached. If a stable state has not been reached, return to step 401 to continue adjustment.

6. The flow cytometer optical path calibration method according to claim 1, characterized in that, Step S50 specifically includes: Step 501: Based on the optical path collimation equation, substitute the laser spot position coordinates into the calculation to obtain the optical path collimation coefficient; Step 502: Based on the detector response equation, substitute the detector output voltage value and the optical path collimation coefficient into the calculation to obtain the detector response coefficient; Step 503: Based on the signal crosstalk equation, substitute the detector output voltage value, the detector response coefficient and the filter position into the calculation to obtain the signal crosstalk coefficient; Step 504: Based on the system gain equation, substitute the dark count rate, the signal count rate, the optical path collimation coefficient, and the detector response coefficient into the calculation to obtain the optimal gain value; Step 505: Verify the calculated coefficients numerically to ensure they meet the system performance requirements.

7. The flow cytometer optical path calibration method according to claim 1, characterized in that, Step S60 specifically includes: Step 601: Calculate the spatial position deviation vector of the laser based on the optical path collimation coefficient, and control the three-dimensional displacement platform to perform position compensation; Step 602: Calculate the optimal operating voltage of the detector based on the detector response coefficient, and control the high-voltage power supply module to regulate the voltage. Step 603: Calculate the optimal angular position of the filter based on the signal crosstalk coefficient, and control the precision angle adjustment mechanism to perform position compensation; Step 604: Calculate the voltage distribution of each stage of the photomultiplier tube based on the optimal gain value, and control the high voltage divider network to set the voltage. Step 605: After completing all adjustments, use standard fluorescent microspheres to perform system performance verification tests and record the final optical path calibration parameters.

8. The flow cytometer optical path calibration method according to claim 1, characterized in that, The solution process for the optical path parameter optimization equations specifically includes: Step 801: Construct a set of nonlinear equations including the optical path collimation equation, the detector response equation, the signal crosstalk equation, and the system gain equation; Step 802: Numerically solve the system of equations using an iterative optimization algorithm, setting initial values ​​and convergence conditions; Step 803: Update the parameters of the system of equations in each iteration, and calculate the objective function value and gradient; Step 804: Adjust the optimization direction based on the gradient information until the convergence condition or the maximum number of iterations is reached; Step 805: Output the final optimization results, including the optical path collimation coefficient, detector response coefficient, signal crosstalk coefficient, and optimal gain value.

9. The flow cytometer optical path calibration method according to claim 1, characterized in that, The process of obtaining the variables in the optical path parameter optimization equation set specifically includes: Step 901: Measure the energy density distribution of the laser spot using a high-precision light intensity distribution analyzer to obtain the light intensity distribution non-uniformity coefficient; Step 902: Measure the detector response characteristics under different temperature conditions using a temperature control system to obtain the detector temperature coefficient; Step 903: Use a long-term stability test system to record the changes in the detector output signal over time and obtain the detector attenuation coefficient; Step 904: Use a multi-channel crosstalk testing system to measure the signal crosstalk level of different detection channel combinations and obtain the signal crosstalk correction coefficient; Step 905: Use the system stability assessment module to measure the changes in system parameters over time and obtain the system stability coefficient.

10. The flow cytometer optical path calibration method according to claim 1, characterized in that, The optimization process of the optical path detection channel conversion cost matrix specifically includes: Step 1001: Establish a multi-dimensional data model that includes the physical distance of the channel, mechanical characteristics, and signal characteristics; Step 1002: Process the delay data during the detection channel switching process using time series analysis methods; Step 1003: Use an adaptive weighting algorithm to dynamically weight different types of time costs; Step 1004: Perform eigenvalue analysis and singular value decomposition on the cost matrix using a matrix factorization algorithm; Step 1005: Generate the optimal detection channel conversion strategy based on the matrix optimization results.

Images