A high-sensitivity ATP fluorescence detection method and system

Abstract

The application provides a high-sensitivity ATP fluorescence detection method and system. The application determines the optimal ratio of luciferase and luciferin, configures a reaction mixture containing the optimal ratio of luciferase and luciferin, and ensures that a significantly enhanced fluorescence signal is generated under a low-concentration ATP condition. Based on the reaction mixture, the application applies microfluidic chip technology to control the reaction environment, fully mixes the reaction mixture and the sample to be detected in a dark environment, adjusts the reaction temperature by using a dynamic temperature control technology, and obtains a mixed reaction system. The application uses a high-sensitivity photomultiplier tube to capture the fluorescence signal generated by the mixed reaction system within a set time, separates and generates a pure target fluorescence signal, and measures the intensity of the pure target fluorescence signal. The application analyzes the intensity of the pure target fluorescence signal, calculates and generates the concentration value of ATP in the sample to be detected. The technical scheme provided by the application improves the sensitivity and accuracy of ATP detection.

Description

Technical Field

[0001] This application relates to the field of biosensing and detection technology, and in particular to a highly sensitive ATP fluorescence detection method and system. Background Technology

[0002] In fields such as biomedicine, food safety, and environmental monitoring, there is a growing demand for rapid and accurate detection of adenosine triphosphate (ATP) concentrations in samples. ATP is a key molecule for intracellular energy transfer, and its concentration can reflect important information such as microbial activity or the degree of contamination.

[0003] Currently, various ATP fluorescence detection methods exist on the market, among which the detection method based on the luciferase-luciferin system is the most common. This system indirectly measures the ATP content in a sample by generating a fluorescent signal through the reaction of luciferase with luciferin and ATP. To improve detection efficiency and accuracy, some studies have also combined microfluidic technology and advanced photoelectric detection devices to optimize the entire process.

[0004] However, traditional methods suffer from high detection limits when processing low concentrations of ATP due to weak fluorescence signals that are difficult to capture effectively. Factors such as the difficulty in precisely controlling reaction temperature under typical laboratory conditions also affect the consistency and reproducibility of the fluorescence signal. Existing methods typically require manual correction of background noise and lack effective algorithmic support, making the conversion process from raw fluorescence signal to final result complex and prone to errors.

[0005] In summary, although various ATP fluorescence detection methods exist, they still exhibit limitations when facing specific challenges. This proposal aims to address these issues by optimizing the ratio of luciferase to luciferin, improving reaction conditions using microfluidic chip technology, and combining advanced signal processing techniques with machine learning models, in order to achieve more efficient and accurate measurement of ATP concentration. Summary of the Invention

[0006] This application provides a highly sensitive ATP fluorescence detection method and system to solve the problem of low sensitivity and accuracy of ATP detection in the prior art.

[0007] In a first aspect, embodiments of this application provide a highly sensitive ATP fluorescence detection method, comprising:

[0008] Determine the optimal ratio of luciferase to luciferin, prepare a reaction mixture containing the optimal ratio of luciferase to luciferin, and ensure that a significantly enhanced fluorescence signal is generated under low ATP concentration conditions.

[0009] Based on the reaction mixture, microfluidic chip technology is used to control the reaction environment, so that the reaction mixture and the sample to be tested are fully mixed in the dark environment. At the same time, dynamic temperature control technology is used to adjust the reaction temperature, improve the reaction rate and fluorescence signal stability, and obtain a mixed reaction system.

[0010] A high-sensitivity photomultiplier tube is used to capture the fluorescence signal generated by the mixed reaction system within a set time. The fluorescence signal is then processed using digital lock-in amplification technology and an adaptive noise cancellation algorithm to separate and generate a pure target fluorescence signal. Simultaneously, the intensity of the pure target fluorescence signal is measured.

[0011] Based on the intensity of the pure target fluorescence signal, a machine learning nonlinear regression model trained with a large dataset of ATP standard solutions of different concentrations is used. The model parameters are optimized using a gradient descent algorithm to analyze the intensity of the pure target fluorescence signal. The concentration of ATP in the sample to be tested is calculated and generated by the model prediction, ensuring the accuracy and reliability of the detection results.

[0012] Optionally, the step of using a high-sensitivity photomultiplier tube to capture the fluorescence signal generated by the mixed reaction system within a set time, and processing the fluorescence signal using digital lock-in amplification technology and an adaptive noise cancellation algorithm to separate and generate a pure target fluorescence signal, while simultaneously measuring the intensity of the pure target fluorescence signal, includes:

[0013] Using a high-sensitivity photomultiplier tube, based on a dynamic integral time window of fluorescence signal intensity change, and combined with an adaptive threshold algorithm, the fluorescence signal generated by the mixed reaction system is efficiently captured to obtain a preliminary capture signal;

[0014] Based on the initial captured signal, a frequency-tracking-based digital phase-locked amplification technique is used. The signal spectrum is analyzed using a fast Fourier transform algorithm, and the frequency closest to the target fluorescence signal is selected as the synchronization reference signal. The initial captured signal is then subjected to phase-locking processing to obtain the amplified target fluorescence signal.

[0015] Based on the amplified target fluorescence signal, an adaptive noise reduction algorithm based on deep learning is used. The signal is monitored and analyzed in real time through a convolutional neural network, and the algorithm parameters are automatically adjusted to remove noise from the amplified target fluorescence signal, thereby obtaining a pure target fluorescence signal.

[0016] The pure target fluorescence signal is digitized using a built-in high-precision analog-to-digital converter. A self-calibrating algorithm is used to correct the nonlinear error in the analog-to-digital conversion process, and the intensity of the pure target fluorescence signal is accurately measured to generate the final fluorescence signal intensity data.

[0017] Optionally, the high-sensitivity photomultiplier tube, based on a dynamic integral time window of fluorescence signal intensity change, combined with an adaptive threshold algorithm, efficiently captures the fluorescence signal generated by the mixed reaction system to obtain a preliminary captured signal, including:

[0018] Using a high-sensitivity photomultiplier tube, based on the average fluorescence signal intensity determined in a preliminary experiment, an initial integration time window is set to begin the initial capture of the fluorescence signal generated by the mixed reaction system, and the intensity of the initially captured fluorescence signal is obtained.

[0019] Based on the intensity of the initially captured fluorescence signal, an adaptive threshold algorithm is used to evaluate the signal intensity in real time. When the signal intensity is detected to exceed the preset threshold range, the length of the integration time window is dynamically adjusted to obtain the optimized integration time window.

[0020] Based on the optimized integration time window, a high-sensitivity photomultiplier tube is used to continuously capture the fluorescence signal generated by the mixed reaction system. The integration time window is updated immediately after each capture to ensure that the best signal capture performance is maintained throughout the detection process, thereby obtaining continuously captured fluorescence signals.

[0021] All the continuously captured fluorescence signals are aggregated to form a preliminary capture signal, providing high-quality raw data for subsequent processing using frequency tracking-based digital lock-in amplification technology.

[0022] Optionally, the step of using frequency-tracking-based digital phase-locked amplification technology to analyze the signal spectrum through a fast Fourier transform algorithm based on the initial captured signal, selecting the frequency closest to the target fluorescence signal as a synchronization reference signal, and performing phase-locking processing on the initial captured signal to obtain the amplified target fluorescence signal includes:

[0023] The Fast Fourier Transform algorithm is used to perform spectral analysis on the initially captured signal to identify multiple frequency components in the signal and obtain preliminary spectral analysis results.

[0024] Based on the preset frequency range and signal strength threshold, the frequency closest to the target fluorescence signal is selected from the preliminary spectrum analysis results as the synchronization reference signal, and the selected synchronization reference signal is generated.

[0025] Based on the selected synchronization reference signal, a digital phase-locked amplification technique based on frequency tracking is adopted. By adjusting the phase-locked loop parameters, the output signal of the phase-locked loop is kept in phase synchronization with the selected synchronization reference signal. The initial captured signal is then subjected to phase-locking processing to generate a phase-locked signal.

[0026] The phase-locked signal is amplified using the adjusted amplifier gain to ensure a significant increase in the intensity of the target fluorescence signal while reducing the influence of background noise, thereby generating an amplified target fluorescence signal.

[0027] Optionally, based on the amplified target fluorescence signal, an adaptive noise reduction algorithm based on deep learning is used. This algorithm monitors and analyzes the signal in real time through a convolutional neural network, automatically adjusts the algorithm parameters, and performs noise removal processing on the amplified target fluorescence signal to obtain a clean target fluorescence signal. This includes:

[0028] Based on the amplified target fluorescence signal, an adaptive noise cancellation model is constructed using a convolutional neural network. The adaptive noise cancellation model is used to monitor and analyze signal characteristics in real time.

[0029] The amplified target fluorescence signal is analyzed in real time using the adaptive noise cancellation model to identify the target component and noise component in the signal and obtain the analysis results.

[0030] Based on the analysis results, the parameters of the adaptive noise cancellation model are automatically adjusted to optimize the noise cancellation effect, ensuring that the noise components in the amplified target fluorescence signal are effectively removed, and generating an optimized noise cancellation model.

[0031] The optimized noise cancellation model is used to remove noise from the amplified target fluorescence signal, generating a clean target fluorescence signal and providing a clean signal source for subsequent signal intensity measurements.

[0032] Optionally, based on the amplified target fluorescence signal, an adaptive noise reduction algorithm based on deep learning is used. This algorithm monitors and analyzes the signal in real time through a convolutional neural network, automatically adjusts the algorithm parameters, and performs noise removal processing on the amplified target fluorescence signal to obtain a clean target fluorescence signal. This includes:

[0033] Based on the amplified target fluorescence signal S amp An adaptive noise cancellation model M is constructed using a convolutional neural network. CNN The adaptive noise cancellation model M CNN Used for real-time monitoring and analysis of signal characteristics;

[0034] Using the adaptive noise cancellation model M CNN The amplified target fluorescence signal S amp Perform real-time analysis to identify target components in the signal.

[0035] Based on the analysis result R, the adaptive noise cancellation model M is automatically adjusted. CNNThe parameters are optimized to improve noise cancellation and ensure the amplified target fluorescence signal S. amp The noise component N is effectively removed to generate an optimized noise cancellation model M'. CNN ;

[0036] The optimized noise cancellation model M' is calculated using the following formula. CNN :

[0037] M′ CNN =M CNN +W·ΔP

[0038] Where ΔP is the parameter increment automatically adjusted based on the analysis result R, and W is the weighting coefficient, which is expressed as:

[0039]

[0040] Where α is the sensitivity coefficient, S threshold It is a preset target component intensity threshold. This formula ensures that when the target component intensity S... target Approaching or exceeding the threshold S threshold When the weighting coefficient W approaches 1, the adjustment of the parameters is increased.

[0041] Using the optimized noise cancellation model M' CNN The amplified target fluorescence signal S amp Noise removal is performed to generate a pure target fluorescence signal S. clean This provides a clean signal source for subsequent signal strength measurements;

[0042] The noise removal formula is calculated using the following formula:

[0043] S clean =S amp -β·N

[0044] Where β is the noise removal coefficient, which is expressed as:

[0045]

[0046] Where γ is the exponential factor, this formula ensures that when the target component intensity S... target When the noise component N is relatively strong, the noise removal coefficient β is close to 1, thus removing noise more effectively.

[0047] Optionally, the step of using a built-in high-precision analog-to-digital converter to digitize the pure target fluorescence signal, employing a self-calibrating algorithm to correct nonlinear errors in the analog-to-digital conversion process, accurately measuring the intensity of the pure target fluorescence signal, and generating final fluorescence signal intensity data includes:

[0048] The pure target fluorescence signal is digitally processed by the built-in high-precision analog-to-digital converter, converting the analog signal into a digital signal to obtain a digital pure target fluorescence signal.

[0049] Based on the digitally purified target fluorescence signal, a self-calibration algorithm is used to analyze the nonlinear errors that may exist in the analog-to-digital conversion process, including quantization error and gain error, and the error analysis results are obtained.

[0050] Based on the error analysis results, the parameters of the self-calibration algorithm are adjusted to correct the nonlinear error in the analog-to-digital conversion process, optimize the quality of the digital signal, and generate the calibrated digital signal.

[0051] Using the corrected digital signal, the intensity of the pure target fluorescence signal is accurately measured to generate the final fluorescence signal intensity data, ensuring the accuracy and reliability of the measurement results.

[0052] Optionally, based on the intensity of the pure target fluorescence signal, a machine learning nonlinear regression model trained using a large dataset of ATP standard solutions of different concentrations is employed. The model parameters are optimized using a gradient descent algorithm to analyze the intensity of the pure target fluorescence signal. The model then predicts and calculates the concentration of ATP in the sample to be tested, ensuring the accuracy and reliability of the detection results. This includes:

[0053] Using a dataset of a large number of ATP standard solutions of different concentrations, which includes ATP standard solutions of known concentrations and their corresponding fluorescence signal intensities, the dataset is preprocessed to obtain a preprocessed dataset.

[0054] Based on the preprocessed dataset, a machine learning nonlinear regression model is constructed, and the gradient descent algorithm is used to optimize the model parameters to ensure that the model can accurately reflect the nonlinear relationship between fluorescence signal intensity and ATP concentration, thereby generating a trained nonlinear regression model.

[0055] Using the trained nonlinear regression model, the intensity of the pure target fluorescence signal is analyzed. Taking the intensity of the pure target fluorescence signal as input, the concentration of ATP in the sample to be tested is predicted and calculated by the model to obtain a preliminary concentration value.

[0056] By iteratively optimizing the model parameters, the accuracy and reliability of the model predictions are verified, and the concentration value of ATP in the sample to be tested is generated to ensure the accuracy and reliability of the detection results.

[0057] Optionally, based on the intensity of the pure target fluorescence signal, a machine learning nonlinear regression model trained using a large dataset of ATP standard solutions of different concentrations is employed. The model parameters are optimized using a gradient descent algorithm to analyze the intensity of the pure target fluorescence signal. The model then predicts and calculates the concentration of ATP in the sample to be tested, ensuring the accuracy and reliability of the detection results. This includes:

[0058] Using a large dataset of ATP standard solutions of varying concentrations, the dataset includes ATP standard solutions C of known concentrations. i and its corresponding fluorescence signal intensity I i The dataset is preprocessed to obtain the preprocessed dataset (C). i I i );

[0059] Based on the preprocessed dataset (C) i I i A machine learning nonlinear regression model f(I; θ) is constructed, and the model parameters are optimized using the gradient descent algorithm to ensure that the model can accurately reflect the nonlinear relationship between fluorescence signal intensity I and ATP concentration C, thus generating a trained nonlinear regression model f. trained (I;θ);

[0060] The parameter update formula for the gradient descent algorithm is as follows:

[0061]

[0062] Where, θ j It is the j-th parameter in the model parameter vector θ, where θ = (θ0, θ1, ..., θj). n () represents the model's parameter vector, α is the learning rate, m is the number of samples in the dataset, and I ij It is the j-th feature value of the i-th sample;

[0063] Using the trained nonlinear regression model f trained (I; θ), representing the intensity I of the pure target fluorescence signal. clean The intensity I of the pure target fluorescence signal was analyzed. clean As input, the concentration C of ATP in the sample to be tested is calculated through model prediction. pred The preliminary concentration value C was obtained. pred ;

[0064] The prediction formula is as follows: C pred =f trained (I clean ;θ)

[0065] By iteratively optimizing the model parameters, the accuracy and reliability of the model predictions were verified, and the concentration value C of ATP in the sample to be tested was generated. final This ensures the accuracy and reliability of the test results;

[0066] The optimized formula is as follows:

[0067]

[0068] C final =f trained (I clean ;θ opt )

[0069] Where, θ opt It is the optimized model parameter vector; C i It is a standard ATP solution of known concentration, I i It corresponds to the fluorescence signal intensity, (C i I i ) is the preprocessed dataset, θ = (θ0, θ1, ..., θ) n ) is the model's parameter vector, α is the learning rate, m is the number of samples in the dataset, and I ij Let f(I;θ) be the j-th feature value of the i-th sample, and f(I;θ) be a nonlinear regression model. trained (I; θ) is a trained nonlinear regression model, where I... clean It is the intensity of the pure target fluorescence signal, C pred This is the preliminary concentration value, C final It is the concentration of ATP in the final generated test sample, θ opt It is the optimized model parameter vector.

[0070] Secondly, embodiments of this application provide a highly sensitive ATP fluorescence detection system, comprising:

[0071] The configuration module is used to determine the optimal ratio of luciferase to luciferin, and to configure a reaction mixture containing the optimal ratio of luciferase to luciferin to ensure that a significantly enhanced fluorescence signal is generated under low ATP concentration conditions.

[0072] The mixing module is used to control the reaction environment based on the reaction mixture using microfluidic chip technology, so that the reaction mixture and the sample to be tested are fully mixed in the dark environment. At the same time, dynamic temperature control technology is used to adjust the reaction temperature, improve the reaction rate and fluorescence signal stability, and obtain a mixed reaction system.

[0073] The separation measurement module is used to capture the fluorescence signal generated by the mixed reaction system within a set time using a high-sensitivity photomultiplier tube, and to process the fluorescence signal using digital lock-in amplification technology and an adaptive noise cancellation algorithm to separate and generate a pure target fluorescence signal, while measuring the intensity of the pure target fluorescence signal.

[0074] The analysis and calculation module is used to analyze the intensity of the pure target fluorescence signal based on the intensity of the fluorescence signal. It employs a machine learning nonlinear regression model trained on a large dataset of ATP standard solutions with different concentrations, optimizes the model parameters using a gradient descent algorithm, and predicts and calculates the concentration of ATP in the sample to be tested, thereby ensuring the accuracy and reliability of the detection results.

[0075] In this embodiment, the optimal ratio of luciferase to luciferin is determined, and a reaction mixture containing the optimal ratio of luciferase and luciferin is prepared to ensure significantly enhanced fluorescence signal under low ATP concentration conditions. Based on the reaction mixture, microfluidic chip technology is used to control the reaction environment, ensuring thorough mixing of the reaction mixture and the test sample in a dark environment. Simultaneously, dynamic temperature control technology is used to adjust the reaction temperature, improving the reaction rate and fluorescence signal stability, resulting in a mixed reaction system. A high-sensitivity photomultiplier tube is used to capture the fluorescence signal generated by the mixed reaction system within a set time. Digital lock-in amplification technology and an adaptive noise cancellation algorithm are used to process the fluorescence signal, separating and generating a pure target fluorescence signal. The intensity of the pure target fluorescence signal is also measured. Based on the intensity of the pure target fluorescence signal, a machine learning nonlinear regression model trained on a large dataset of ATP standard solutions of different concentrations is used. The model parameters are optimized using a gradient descent algorithm to analyze the intensity of the pure target fluorescence signal. The model predicts and calculates the concentration of ATP in the test sample, ensuring the accuracy and reliability of the detection results.

[0076] The technical solution of this application has the following beneficial effects:

[0077] This invention, by determining the optimal ratio of luciferase to luciferin and preparing a reaction mixture containing this optimal ratio, enables the generation of significantly enhanced fluorescence signals under low ATP concentration conditions, thereby improving detection sensitivity. The application of microfluidic chip technology and dynamic temperature control technology ensures the stability of the reaction environment and reaction speed, enhancing the stability of the fluorescence signal. Utilizing a high-sensitivity photomultiplier tube combined with digital lock-in amplification technology and an adaptive noise cancellation algorithm, the invention efficiently captures and processes fluorescence signals, separating and generating pure target fluorescence signals while accurately measuring their intensity. Based on a machine learning nonlinear regression model trained on a large dataset of ATP standard solutions of varying concentrations, the gradient descent algorithm is used to optimize model parameters, further improving the accuracy of pure target fluorescence signal intensity analysis, thus ensuring the precision and reliability of ATP concentration values ​​in the test sample.

[0078] Furthermore, by using a high-sensitivity photomultiplier tube and combining a dynamic integral time window based on the fluorescence signal intensity change with an adaptive threshold algorithm, this invention can efficiently capture the fluorescence signal generated by the mixed reaction system, obtaining a preliminary captured signal. Employing frequency-tracking-based digital lock-in amplification technology, the signal spectrum is analyzed using a fast Fourier transform algorithm, and the frequency closest to the target fluorescence signal is selected as a synchronization reference signal for phase-locking processing, effectively amplifying the target fluorescence signal. A deep learning-based adaptive noise cancellation algorithm is used, which monitors and analyzes the signal in real time through a convolutional neural network, automatically adjusting algorithm parameters to remove noise components and obtain a pure target fluorescence signal. Finally, the built-in high-precision analog-to-digital converter is used to digitize the signal, and a self-calibrating algorithm is employed to correct nonlinear errors, ensuring the accuracy of the fluorescence signal intensity measurement and providing a high-quality data source for subsequent analysis.

[0079] Furthermore, this application utilizes a large dataset of ATP standard solutions of varying concentrations and preprocesses it to construct a machine learning nonlinear regression model. This model accurately reflects the nonlinear relationship between fluorescence signal intensity and ATP concentration. A gradient descent algorithm is employed to optimize model parameters, ensuring the model's accuracy and generalization ability. The trained nonlinear regression model is used to analyze the fluorescence signal intensity of the pure target sample, and the model predicts and calculates the ATP concentration in the test sample, obtaining a preliminary concentration value. Through multiple iterations to optimize model parameters, the accuracy and reliability of the model's predictions are verified, ultimately generating the ATP concentration value in the test sample, ensuring high precision and reliability of the detection results. This method not only improves detection efficiency but also enhances the credibility of the detection results, making it suitable for various applications requiring high-precision ATP concentration detection.

[0080] These or other aspects of this application will become more apparent in the following description of the embodiments. Attached Figure Description

[0081] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0082] Figure 1 A flowchart illustrating a highly sensitive ATP fluorescence detection method provided in this application embodiment;

[0083] Figure 2 A schematic diagram of a high-sensitivity ATP fluorescence detection system provided in an embodiment of this application;

[0084] Figure 3 This is a schematic diagram of the structure of a computing device provided in an embodiment of this application. Detailed Implementation

[0085] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings.

[0086] In some of the processes described in the specification, claims, and accompanying drawings of this application, multiple operations appearing in a specific order are included. However, it should be clearly understood that these operations may not be executed in the order they appear herein, or may be executed in parallel. The operation numbers, such as 101, 102, etc., are merely used to distinguish different operations and do not themselves represent any execution order. Furthermore, these processes may include more or fewer operations, and these operations may be executed sequentially or in parallel. It should be noted that the descriptions such as "first," "second," etc., in this document are used to distinguish different messages, devices, modules, etc., and do not represent a chronological order, nor do they limit "first" and "second" to different types.

[0087] 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.

[0088] Figure 1 A flowchart of a highly sensitive ATP fluorescence detection method is provided in this application embodiment, as follows: Figure 1 As shown, the method includes:

[0089] 101. Determine the optimal ratio of luciferase to luciferin, prepare a reaction mixture containing the optimal ratio of luciferase to luciferin, and ensure that a significantly enhanced fluorescence signal is generated under low ATP concentration conditions.

[0090] In this step, luciferase is an enzyme that catalyzes the oxidation of luciferin to produce light. Luciferin is a compound that emits light under the action of luciferase. The optimal ratio is determined experimentally to ensure a significantly enhanced fluorescence signal under low ATP concentration conditions.

[0091] First, through a series of preliminary experiments, the optimal ratio of luciferase to luciferin was determined so that the fluorescence signal intensity reached its maximum value in the presence of low concentrations of ATP.

[0092] Secondly, based on the determined optimal ratio, a reaction mixture containing luciferase and luciferin is prepared to provide the necessary chemical reaction conditions for subsequent detection.

[0093] In this application example, it is assumed that in water quality testing, the intensity of fluorescence signal is measured and recorded by adding a known concentration of ATP standard solution to a mixture of luciferase and luciferin in different proportions.

[0094] The ratio with the strongest fluorescence signal was selected as the optimal ratio, and the reaction mixture was prepared using this ratio for ATP detection in actual water samples.

[0095] 102. Based on the reaction mixture, microfluidic chip technology is used to control the reaction environment, so that the reaction mixture and the sample to be tested are fully mixed in the dark environment. At the same time, dynamic temperature control technology is used to adjust the reaction temperature, improve the reaction rate and fluorescence signal stability, and obtain a mixed reaction system.

[0096] In this step, microfluidic chip technology is a technique that utilizes a network of microchannels to process or analyze trace amounts of liquid. Dynamic temperature control technology optimizes reaction conditions by adjusting the reaction temperature in real time.

[0097] First, the prepared reaction mixture and the sample to be tested are introduced into the microfluidic chip and thoroughly mixed in a dark environment.

[0098] Secondly, dynamic temperature control technology is used to adjust the reaction temperature as needed to accelerate the reaction rate and improve the stability of the fluorescence signal.

[0099] In this application example, it is assumed that in water quality testing, water samples and reaction mixtures are injected into different channels of a microfluidic chip, and their flow and mixing are controlled by micropumps and valves. Then, using built-in heating elements and temperature sensors, the reaction temperature is monitored and adjusted in real time to ensure that the reaction proceeds at the optimal temperature, thereby improving detection sensitivity and accuracy.

[0100] 103. Using a high-sensitivity photomultiplier tube, the fluorescence signal generated by the mixed reaction system is captured within a set time. The fluorescence signal is then processed using digital lock-in amplification technology and an adaptive noise cancellation algorithm to separate and generate a pure target fluorescence signal. Simultaneously, the intensity of the pure target fluorescence signal is measured.

[0101] In this step, a high-sensitivity photomultiplier tube is a highly sensitive photoelectric conversion device capable of converting weak light signals into electrical signals. Digital lock-in amplification (LLA) is a signal processing technique used to extract useful signals from noise. Adaptive noise cancellation algorithms are algorithms that automatically adjust parameters to remove noise.

[0102] First, a high-sensitivity photomultiplier tube is used to capture the fluorescence signal generated by the mixed reaction system.

[0103] Secondly, digital lock-in amplification technology and adaptive noise cancellation algorithm are used to process the captured signal and separate the pure target fluorescence signal.

[0104] Finally, the intensity of the purified target fluorescence signal was measured to provide a data basis for subsequent analysis.

[0105] In this application example, it is assumed that in water quality testing, a photomultiplier tube is placed in the fluorescence detection region of a microfluidic chip to capture the fluorescence signal after the water sample reaction. The frequency of the fluorescence signal is locked using digital lock-in amplification technology to remove background noise; an adaptive noise cancellation algorithm is then used to further purify the signal. Finally, pure fluorescence signal intensity data is obtained for subsequent ATP concentration calculation.

[0106] Optionally, step 103, which involves using a high-sensitivity photomultiplier tube to capture the fluorescence signal generated by the mixed reaction system within a set time, and processing the fluorescence signal using digital lock-in amplification technology and an adaptive noise cancellation algorithm to separate and generate a pure target fluorescence signal, while simultaneously measuring the intensity of the pure target fluorescence signal, includes: using a high-sensitivity photomultiplier tube, based on a dynamic integral time window of fluorescence signal intensity changes, combined with an adaptive threshold algorithm, to efficiently capture the fluorescence signal generated by the mixed reaction system and obtain a preliminary captured signal; based on the preliminary captured signal, using frequency-tracking-based digital lock-in amplification technology, and analyzing the signal spectrum using a fast Fourier transform algorithm. The frequency closest to the target fluorescence signal is selected as the synchronization reference signal, and the initial capture signal is phase-locked to obtain the amplified target fluorescence signal. Based on the amplified target fluorescence signal, an adaptive noise cancellation algorithm based on deep learning is used. The signal is monitored and analyzed in real time through a convolutional neural network, and the algorithm parameters are automatically adjusted to remove noise from the amplified target fluorescence signal to obtain a clean target fluorescence signal. The clean target fluorescence signal is digitized using a built-in high-precision analog-to-digital converter. A self-calibrating algorithm is used to correct the nonlinear error in the analog-to-digital conversion process, and the intensity of the clean target fluorescence signal is accurately measured to generate the final fluorescence signal intensity data.

[0107] Optionally, step 103, which utilizes a high-sensitivity photomultiplier tube to efficiently capture the fluorescence signal generated by the mixed reaction system based on a dynamic integration time window of fluorescence signal intensity changes and an adaptive threshold algorithm to obtain a preliminary captured signal, includes: using a high-sensitivity photomultiplier tube, setting an initial integration time window based on the average fluorescence signal intensity determined in a pre-experiment, and starting to initially capture the fluorescence signal generated by the mixed reaction system to obtain the initially captured fluorescence signal intensity; based on the initially captured fluorescence signal intensity, using an adaptive threshold algorithm to evaluate the signal intensity in real time, and dynamically adjusting the length of the integration time window when the detected signal intensity exceeds a preset threshold range to obtain an optimized integration time window; based on the optimized integration time window, continuously capturing the fluorescence signal generated by the mixed reaction system using a high-sensitivity photomultiplier tube, updating the integration time window immediately after each capture to ensure optimal signal capture performance throughout the detection process, and obtaining continuously captured fluorescence signals; summarizing all the continuously captured fluorescence signals to form a preliminary captured signal, providing high-quality raw data for subsequent processing using frequency tracking-based digital lock-in amplification technology.

[0108] Optionally, step 103, which involves using frequency-tracking digital phase-locked amplification (PLA) technology based on the initial captured signal, analyzing the signal spectrum using a fast Fourier transform (FFT) algorithm, selecting the frequency closest to the target fluorescence signal as a synchronization reference signal, and performing phase-locking processing on the initial captured signal to obtain the amplified target fluorescence signal, includes: performing spectral analysis on the initial captured signal using a FFT algorithm to identify multiple frequency components in the signal and obtain preliminary spectral analysis results; selecting the frequency closest to the target fluorescence signal from the preliminary spectral analysis results as a synchronization reference signal based on a preset frequency range and signal strength threshold, generating a selected synchronization reference signal; using frequency-tracking digital PLA technology based on the selected synchronization reference signal, adjusting the PLA parameters to ensure that the PLA output signal is phase-synchronized with the selected synchronization reference signal, performing phase-locking processing on the initial captured signal to generate a phase-locked signal; and amplifying the phase-locked signal using the adjusted amplifier gain to ensure a significant increase in the target fluorescence signal intensity while reducing the influence of background noise, thereby generating the amplified target fluorescence signal.

[0109] Optionally, step 103, which involves using a deep learning-based adaptive noise cancellation algorithm based on the amplified target fluorescence signal to perform noise removal processing on the amplified target fluorescence signal and obtain a clean target fluorescence signal, includes: constructing an adaptive noise cancellation model based on the amplified target fluorescence signal using a convolutional neural network, wherein the adaptive noise cancellation model is used to monitor and analyze signal characteristics in real time; using the adaptive noise cancellation model to perform real-time analysis on the amplified target fluorescence signal, identifying target components and noise components in the signal, and obtaining analysis results; automatically adjusting the parameters of the adaptive noise cancellation model according to the analysis results to optimize the noise cancellation effect, ensuring effective removal of noise components in the amplified target fluorescence signal, and generating an optimized noise cancellation model; and using the optimized noise cancellation model to perform noise removal processing on the amplified target fluorescence signal to generate a clean target fluorescence signal, providing a clean signal source for subsequent signal intensity measurements.

[0110] Optionally, step 103, which involves using a built-in high-precision analog-to-digital converter to digitize the purified target fluorescence signal, employing a self-calibrating algorithm to correct nonlinear errors during the analog-to-digital conversion process, accurately measuring the intensity of the purified target fluorescence signal, and generating final fluorescence signal intensity data, includes: using the built-in high-precision analog-to-digital converter to digitize the purified target fluorescence signal, converting the analog signal into a digital signal to obtain a digitized purified target fluorescence signal; based on the digitized purified target fluorescence signal, using a self-calibrating algorithm to analyze possible nonlinear errors during the analog-to-digital conversion process, including quantization errors and gain errors, to obtain error analysis results; adjusting the parameters of the self-calibrating algorithm according to the error analysis results to correct nonlinear errors during the analog-to-digital conversion process, optimizing the quality of the digital signal, and generating a calibrated digital signal; and using the calibrated digital signal to accurately measure the intensity of the purified target fluorescence signal to generate final fluorescence signal intensity data, ensuring the accuracy and reliability of the measurement results.

[0111] In this step, the dynamic integration time window is an integration time setting that automatically adjusts according to changes in fluorescence signal intensity to optimize signal capture efficiency. The adaptive threshold algorithm is an algorithm that adjusts the threshold in real time based on signal strength to determine whether the integration time window needs adjustment. The Fast Fourier Transform (FFT) algorithm is an efficient algorithm for calculating the Discrete Fourier Transform, used for analyzing signal spectra. Digital lock-in amplification (PLA) is a technique that extracts useful signals by locking the signal frequency, often used to separate target signals from noise. Convolutional neural networks (CNNs) are deep learning models particularly suitable for image and signal processing tasks, capable of automatically identifying and classifying features in signals. A high-precision analog-to-digital converter (ADC) is a device that converts analog signals into digital signals, possessing high precision characteristics. The self-calibration algorithm is an algorithm used to correct nonlinear errors during analog-to-digital conversion, ensuring the accuracy of the digital signal.

[0112] First, a high-sensitivity photomultiplier tube is used to capture fluorescence signals. An initial integration time window is set based on the average fluorescence signal intensity determined in preliminary experiments. An adaptive threshold algorithm is used to evaluate the signal intensity in real time; when the signal intensity exceeds a preset threshold range, the integration time window length is dynamically adjusted. Based on the optimized integration time window, fluorescence signals are continuously captured, and the integration time window is updated in real time to ensure optimal signal capture performance. All continuously captured fluorescence signals are then aggregated to form the initial capture signal.

[0113] Secondly, a fast Fourier transform algorithm is used to perform spectral analysis on the initially captured signal to identify multiple frequency components. Based on a preset frequency range and signal strength threshold, the frequency closest to the target fluorescence signal is selected as the synchronization reference signal. Digital phase-locked amplification technology is employed, and by adjusting the phase-locked loop parameters, the output signal is kept in phase synchronization with the synchronization reference signal. The initially captured signal undergoes phase-locking processing, and the amplified target fluorescence signal is generated using the adjusted amplifier gain.

[0114] Furthermore, an adaptive noise cancellation model based on a convolutional neural network is constructed for real-time monitoring and analysis of signal characteristics. This model is used to analyze the amplified target fluorescence signal in real time, identifying target and noise components. Based on the analysis results, the model parameters are automatically adjusted to optimize the noise cancellation effect. The optimized noise cancellation model is then used to remove noise from the amplified target fluorescence signal, generating a clean target fluorescence signal.

[0115] Finally, a high-precision analog-to-digital converter (ADC) is used to convert the purified target fluorescence signal into a digital signal. A self-calibrating algorithm is employed to analyze the nonlinear errors in the ADC process, including quantization and gain errors. Based on the error analysis results, the parameters of the self-calibrating algorithm are adjusted to optimize the digital signal quality. Using the calibrated digital signal, the intensity of the purified target fluorescence signal is accurately measured to generate the final fluorescence signal intensity data.

[0116] In the embodiments of this application, it is assumed that the above-mentioned scheme can be adopted in order to improve the sensitivity and accuracy of ATP fluorescence detection in water quality testing.

[0117] First, in water sample testing, a high-sensitivity photomultiplier tube is used to capture fluorescence signals. An initial integration time window is set based on the average fluorescence signal intensity determined in pre-experiments. During detection, an adaptive threshold algorithm is used to evaluate the signal intensity in real time and dynamically adjust the integration time window to ensure optimal signal capture performance. All captured fluorescence signals are then aggregated to form a preliminary capture signal.

[0118] Secondly, a fast Fourier transform algorithm is used to perform spectral analysis on the initially captured signal to identify multiple frequency components. Based on a preset frequency range and signal strength threshold, the frequency closest to the target fluorescence signal is selected as the synchronization reference signal. Digital phase-locked loop (PLL) amplification technology is employed, adjusting the PLL parameters to maintain phase synchronization between the output signal and the synchronization reference signal, thereby amplifying the target fluorescence signal.

[0119] Furthermore, an adaptive noise cancellation model based on a convolutional neural network is constructed to monitor and analyze the amplified target fluorescence signal in real time. The model can identify the target component and noise component in the signal and automatically adjust parameters to optimize the noise cancellation effect. Finally, a pure target fluorescence signal is generated.

[0120] Finally, a high-precision analog-to-digital converter (ADC) was used to convert the purified target fluorescence signal into a digital signal. A self-calibrating algorithm was employed to analyze and correct nonlinear errors during the ADC process, ensuring the accuracy of the digital signal. Using the corrected digital signal, the fluorescence signal intensity was accurately measured to generate the final fluorescence signal intensity data for subsequent ATP concentration calculations.

[0121] These steps enable highly sensitive and accurate ATP fluorescence detection in water quality testing, ensuring the reliability and repeatability of test results.

[0122] This application considers that in highly sensitive ATP fluorescence detection methods, noise removal processing is required for the amplified target fluorescence signal to improve signal purity and detection accuracy. Here, a deep learning-based adaptive noise reduction algorithm is employed. This algorithm uses a convolutional neural network to monitor and analyze signal characteristics in real time, automatically adjusting algorithm parameters to effectively remove noise components.

[0123] Optionally, step 103, which involves using a deep learning-based adaptive noise reduction algorithm on the amplified target fluorescence signal, to automatically adjust algorithm parameters by monitoring and analyzing the signal in real time through a convolutional neural network, and performing noise removal processing on the amplified target fluorescence signal to obtain a clean target fluorescence signal, includes:

[0124] Based on the amplified target fluorescence signal S amp An adaptive noise cancellation model M is constructed using a convolutional neural network. CNN The adaptive noise cancellation model N CNN Used for real-time monitoring and analysis of signal characteristics;

[0125] Using the adaptive noise cancellation model N CNN The amplified target fluorescence signal S amp Perform real-time analysis to identify target components in the signal.

[0126] Based on the analysis result R, the adaptive noise cancellation model M is automatically adjusted. CNN The parameters are optimized to improve noise cancellation and ensure the amplified target fluorescence signal S. amp The noise component N is effectively removed to generate an optimized noise cancellation model M'. CNN ;

[0127] The optimized noise cancellation model M' is calculated using the following formula. CNN :

[0128] M′ CNN =M CNN+W·ΔP

[0129] Where ΔP is the parameter increment automatically adjusted based on the analysis result R, and W is the weighting coefficient, which is expressed as:

[0130]

[0131] Where α is the sensitivity coefficient, S threshold It is a preset target component intensity threshold. This formula ensures that when the target component intensity S... target Approaching or exceeding the threshold S threshold When the weighting coefficient W approaches 1, the adjustment of the parameters is increased.

[0132] Using the optimized noise cancellation model M' CNN The amplified target fluorescence signal S amp Noise removal is performed to generate a pure target fluorescence signal S. clean This provides a clean signal source for subsequent signal strength measurements;

[0133] The noise removal formula is calculated using the following formula:

[0134] S clean =S amp -β·N

[0135] Where β is the noise removal coefficient, which is expressed as:

[0136]

[0137] Where γ is the exponential factor, this formula ensures that when the target component intensity S... target When the noise component N is relatively strong, the noise removal coefficient β is close to 1, thus removing noise more effectively.

[0138] The adaptive noise cancellation model M is dynamically adjusted using weighting coefficients W and parameter increments ΔP. CNN The parameters allow the model to self-optimize based on the current signal characteristics. The design of the weight coefficients W ensures that when the target component intensity S... target Approaching or exceeding the threshold S threshold At the same time, the parameter adjustment is increased to better adapt to signal changes. The design of the noise removal coefficient β ensures that when the target component intensity S... target It can remove noise more effectively when the noise component N is strong, while preserving the target signal.

[0139] The following is a brief introduction to how the parameters of this formula are obtained:

[0140] The sensitivity coefficient α can be determined experimentally. Choosing an appropriate α value ensures that W in S target Approaching S threshold There is a noticeable response at times. Target component intensity threshold S threshold Through statistical analysis of experimental data, a reasonable threshold was determined, which should be able to distinguish between normal and noise signals. The exponential factor γ was determined experimentally, and an appropriate value of γ was selected so that β... target When the parameter is stronger than N, it approaches 1, while decreasing as N becomes larger. The parameter increment ΔP is calculated through gradient descent or other optimization algorithms during model training, representing the amount of parameter change in each iteration.

[0141] Assume the following parameters have been obtained in the water quality testing:

[0142] Sensitivity coefficient α = 10; preset target component intensity threshold S threshold =500; Exponential factor γ=2;

[0143] Furthermore, the following data was obtained from the actual testing:

[0144] Amplified target fluorescence signal S amp =800; Preliminary analysis results show that the target component intensity S target =700; Noise component N=300;

[0145] Calculate the weighting coefficient W:

[0146]

[0147] W≈1

[0148] Because of e -2000 It is very close to 0, therefore W is close to 1.

[0149] Calculate the parameter increment ΔP:

[0150] Assume the parameter increment ΔP obtained through the model training process is 0.01;

[0151] Generate the optimized noise cancellation model M' CNN :

[0152] M′ CNN =M CNN +W·ΔP

[0153] M′ CNN =M CNN +1·0.01

[0154] M′ CNN =M CNN +0.01

[0155] Calculate the noise removal factor β:

[0156]

[0157] β = 0.49

[0158] Generate pure target fluorescence signal S clean :

[0159] S clean =S amp -β·N

[0160] S clean =800 - 0.49·300

[0161] S clean =653

[0162] The pure target fluorescence signal S was obtained through the above calculations. clean =653. This indicates that after processing by the adaptive noise cancellation algorithm, the noise components in the original signal were effectively removed, and the target signal was enhanced. The weighting coefficient W≈1 indicates that the current target component intensity S target =700 is much higher than the threshold S threshold =500, therefore the model parameters are adjusted significantly, which helps to better adapt to signal changes. The noise removal coefficient β = 0.49 indicates that the noise removal coefficient is relatively high at the target component intensity S. target When the noise component N=300 and the noise component N=700, the noise removal effect is strong, but it will not completely remove the noise to avoid distortion of the target signal. Pure target fluorescence signal S clean =653, compared to the original signal S amp =800, which removes some noise components, resulting in a purer signal and improving the accuracy and reliability of detection.

[0163] This treatment method can significantly improve the quality of ATP fluorescence signals in water quality testing, ensuring the accuracy and reliability of the test results.

[0164] 104. Based on the intensity of the pure target fluorescence signal, a machine learning nonlinear regression model trained with a large dataset of ATP standard solutions of different concentrations is used. The model parameters are optimized using a gradient descent algorithm. The intensity of the pure target fluorescence signal is analyzed, and the concentration of ATP in the sample to be tested is calculated and generated through model prediction, ensuring the accuracy and reliability of the detection results.

[0165] In this step, the machine learning nonlinear regression model is a data-trained model capable of fitting nonlinear relationships. The gradient descent algorithm is an optimization algorithm used to minimize the loss function, thereby optimizing the model parameters.

[0166] First, a machine learning nonlinear regression model was trained using a large dataset of ATP standard solutions of different concentrations.

[0167] Secondly, the gradient descent algorithm was used to optimize the model parameters to ensure that the model could accurately reflect the nonlinear relationship between fluorescence signal intensity and ATP concentration.

[0168] Finally, the trained model is used to analyze the intensity of the pure target fluorescence signal to predict and generate the concentration of ATP in the sample to be tested.

[0169] In this application example, it is assumed that in water quality testing, a series of ATP standard solutions of known concentrations are used to collect corresponding fluorescence signal intensity data to construct a training dataset. Next, a nonlinear regression model is trained, and the model parameters are optimized using a gradient descent algorithm. Finally, the pure target fluorescence signal intensity is input into the trained model to predict and output the ATP concentration value in the water sample being tested, ensuring the accuracy and reliability of the detection results.

[0170] Optionally, step 104, which involves analyzing the intensity of the pure target fluorescence signal using a machine learning nonlinear regression model trained on a large dataset of ATP standard solutions of varying concentrations, and optimizing the model parameters using a gradient descent algorithm, to predict and generate the concentration of ATP in the sample to be tested, thereby ensuring the accuracy and reliability of the detection results, includes: preprocessing the dataset of a large dataset of ATP standard solutions of varying concentrations, which contains known concentrations of ATP standard solutions and their corresponding fluorescence signal intensities, to obtain a preprocessed dataset. Based on the preprocessed dataset, a machine learning nonlinear regression model is constructed. The gradient descent algorithm is used to optimize the model parameters to ensure that the model can accurately reflect the nonlinear relationship between fluorescence signal intensity and ATP concentration, generating a trained nonlinear regression model. Using the trained nonlinear regression model, the intensity of the pure target fluorescence signal is analyzed. The intensity of the pure target fluorescence signal is used as input, and the model predicts and calculates the ATP concentration value in the sample to be tested, obtaining a preliminary concentration value. Through multiple iterations to optimize the model parameters, the accuracy and reliability of the model prediction are verified, and the ATP concentration value in the sample to be tested is generated, ensuring the accuracy and reliability of the detection results.

[0171] In this step, the machine learning nonlinear regression model is a data-trained model capable of fitting the nonlinear relationship between fluorescence signal intensity and ATP concentration. The gradient descent algorithm is an optimization algorithm that iteratively updates model parameters to minimize the error between predicted and actual values, thereby optimizing model performance. Preprocessing involves cleaning and standardizing the raw data to improve the effectiveness of model training. The dataset is a collection of data containing ATP standard solutions of known concentrations and their corresponding fluorescence signal intensities, used for training and validating the model.

[0172] First, a large number of ATP standard solutions of different concentrations were collected, and their corresponding fluorescence signal intensities were recorded. The collected data underwent preprocessing, including outlier removal, normalization, and other steps to ensure data quality. A preprocessed dataset was then generated to prepare for subsequent model training.

[0173] Secondly, a suitable machine learning algorithm is selected to construct a nonlinear regression model. Gradient descent is used to optimize the model parameters, and multiple iterations are employed to reduce the error between predicted and actual values. Model performance is evaluated using methods such as cross-validation to ensure that the model accurately reflects the nonlinear relationship between fluorescence signal intensity and ATP concentration. The trained nonlinear regression model is then generated for subsequent concentration prediction.

[0174] Furthermore, the intensity of the pure target fluorescence signal is used as input to a trained nonlinear regression model. Based on the input fluorescence signal intensity, the model predicts and outputs the concentration of ATP in the sample, thus obtaining a preliminary concentration value.

[0175] Furthermore, the model parameters are optimized through multiple iterations to ensure that the model maintains good predictive performance on new data. The predictive accuracy of the model is verified using a validation set or test set to ensure the reliability of the model in practical applications.

[0176] Finally, the concentration value of ATP in the sample to be tested is generated to ensure the accuracy and reliability of the test results.

[0177] In the embodiments of this application, it is assumed that the above-mentioned scheme can be used in order to accurately determine the ATP concentration in a water sample during water quality testing.

[0178] First, a series of ATP standard solutions of known concentrations were collected, and their corresponding fluorescence signal intensities were measured. These data may come from laboratory experiments or historical data records. Data preprocessing was performed, including outlier removal, normalization, or standardization, to ensure data consistency and high quality.

[0179] Secondly, a neural network-based nonlinear regression model was constructed using the preprocessed dataset. The model parameters were optimized using gradient descent to ensure that the model accurately reflects the nonlinear relationship between fluorescence signal intensity and ATP concentration. Cross-validation was used to evaluate the model's performance and ensure its generalization ability on new data.

[0180] Furthermore, in actual water quality testing, the intensity of the pure target fluorescence signal is input into a trained nonlinear regression model. Based on the input fluorescence signal intensity, the model predicts and outputs the concentration of ATP in the water sample, thus obtaining a preliminary concentration value.

[0181] Finally, the model parameters were further optimized through multiple iterations to ensure that the model maintained good predictive performance on new water quality samples. The model's predictive accuracy was verified using a validation set or test set to ensure its reliability in practical applications. Finally, the concentration value of ATP in the water sample was generated to ensure the accuracy and reliability of the detection results.

[0182] These steps enable high-precision and high-reliability determination of ATP concentration in water quality testing, ensuring the effectiveness and accuracy of water quality monitoring.

[0183] This application considers that, in highly sensitive ATP fluorescence detection methods, to accurately predict the ATP concentration in the sample, a machine learning nonlinear regression model f(I;θ) is used to establish the relationship between fluorophore signal intensity I and ATP concentration C. The model is trained using a large dataset of ATP standard solutions with different concentrations, and the model parameters are optimized using a gradient descent algorithm to ensure that the model accurately reflects this nonlinear relationship.

[0184] Optionally, in step 104, based on the intensity of the pure target fluorescence signal, a machine learning nonlinear regression model trained using a large dataset of ATP standard solutions of different concentrations is employed. The model parameters are optimized using a gradient descent algorithm to analyze the intensity of the pure target fluorescence signal. The model then predicts and calculates the concentration of ATP in the sample to be tested, ensuring the accuracy and reliability of the detection results. This includes:

[0185] Using a large dataset of ATP standard solutions of varying concentrations, the dataset includes ATP standard solutions C of known concentrations. i and its corresponding fluorescence signal intensity I i The dataset is preprocessed to obtain the preprocessed dataset (C). i I i );

[0186] Based on the preprocessed dataset (C) i I iA machine learning nonlinear regression model f(I; θ) is constructed, and the model parameters are optimized using the gradient descent algorithm to ensure that the model can accurately reflect the nonlinear relationship between fluorescence signal intensity I and ATP concentration C, thus generating a trained nonlinear regression model f. trained (I;θ);

[0187] The parameter update formula for the gradient descent algorithm is as follows:

[0188]

[0189] Where, θ j It is the j-th parameter in the model parameter vector θ, where θ = (θ0, θ1, ..., θj). n () represents the model's parameter vector, α is the learning rate, m is the number of samples in the dataset, and I ij It is the j-th feature value of the i-th sample;

[0190] Using the trained nonlinear regression model f trained (I; θ), representing the intensity I of the pure target fluorescence signal. clean The intensity I of the pure target fluorescence signal was analyzed. clean As input, the concentration C of ATP in the sample to be tested is calculated through model prediction. pred The preliminary concentration value C was obtained. pred ;

[0191] The prediction formula is as follows: C pred =f trained (I clean ;θ)

[0192] By iteratively optimizing the model parameters, the accuracy and reliability of the model predictions were verified, and the concentration value C of ATP in the sample to be tested was generated. final This ensures the accuracy and reliability of the test results;

[0193] The optimized formula is as follows:

[0194]

[0195] C final =f trained (I clean ;θ opt )

[0196] Where, θ opt It is the optimized model parameter vector; C i It is a standard ATP solution of known concentration, I i It corresponds to the fluorescence signal intensity, (C i I i) is the preprocessed dataset, θ = (θ0, θ1, ..., θ) n ) is the model's parameter vector, α is the learning rate, m is the number of samples in the dataset, and I ij Let f(I;θ) be the j-th feature value of the i-th sample, and f(I;θ) be a nonlinear regression model. trained (I; θ) is a trained nonlinear regression model, where I... clean It is the intensity of the pure target fluorescence signal, C pred This is the preliminary concentration value, C final It is the concentration of ATP in the final generated test sample, θ opt It is the optimized model parameter vector.

[0197] The raw data undergoes preprocessing such as cleaning and standardization to improve the model training effect. The preprocessed dataset (C...) i I i ) Contains a standard ATP solution of known concentration C i and its corresponding fluorescence signal intensity I i A nonlinear regression model f(I; θ) is constructed, where θ is the model parameter vector. The gradient descent algorithm is used to optimize the model parameters θ, enabling the model to accurately fit the nonlinear relationship between light signal intensity I and ATP concentration C. The trained model f is then used... trained (I; θ) represents the intensity of the pure target fluorescence signal I clean The analysis was performed to predict the ATP concentration C in the sample to be tested. pred By iteratively optimizing the model parameters, the final ATP concentration value C was generated. final。

[0198] The following is a brief introduction to the design rationale behind each term of the formula:

[0199] This sub-item determines the direction and magnitude of model parameter adjustment by calculating the average of the product of the prediction error and the corresponding feature value for each sample, thereby gradually reducing the overall prediction error.

[0200] The following is a brief introduction to how the parameters of this formula are obtained:

[0201] Dataset (C) i I i The dataset was obtained through experimental measurements, containing a series of ATP standard solutions of known concentrations and their corresponding fluorescence signal intensities. The learning rate α was determined experimentally, with an appropriate α chosen to balance learning speed and stability. The sample size m was determined based on the actual dataset size. The eigenvalues ​​I... ij The model parameters are obtained directly from experimental data. The model parameter vector θ is automatically learned during the training process.

[0202] Assume the following data is already obtained:

[0203] Dataset (C) i I i It contains 100 samples, where C i and I i These represent ATP standard solutions of known concentrations and their corresponding fluorescence signal intensities; the learning rate α = 0.01; the model parameter vector θ = (θ0, θ1, ..., θ2). n The initial value is a random value; the intensity I of the pure target fluorescence signal. clean =653;

[0204] Data preprocessing: Assuming the dataset (C) i I i The data has already undergone preprocessing, such as normalization or standardization.

[0205] Constructing a nonlinear regression model:

[0206] Assume the model is a multinomial regression model f(I; θ) = θ0 + θ1I + θ2I 2 +…+θ n I n ;

[0207] Gradient descent algorithm optimizes model parameters:

[0208] The initial parameter vector is θ = (0.1, 0.2, 0.3, ..., 0.1).

[0209] The parameter θ is iteratively updated using the gradient descent algorithm until the model converges.

[0210] The trained nonlinear regression model:

[0211] After multiple iterations, the trained model parameter vector θ is obtained. trained = (0.1, 0.2, 0.3, ..., 0.1).

[0212] Predict ATP concentration;

[0213] Preliminary forecast:

[0214] C pred =f trained (I clean ;θ trained )

[0215] C pred =0.1 + 0.2·653 + 0.3·653 2 +…+0.1·653 n

[0216] Assuming that C is obtained through calculationpred =10.5μM.

[0217] Optimize model parameters:

[0218] By iteratively optimizing the model parameters, the optimized parameter vector θ is obtained. opt = (0.1, 0.2, 0.3, ..., 0.1).

[0219] Final prediction:

[0220] C final =f trained (I clean ;θ opt )

[0221] C final =0.1 + 0.2·653 + 0.3·653 2 +…+0.1·653 n

[0222] Assuming that C is obtained through calculation final =10.7μM.

[0223] The above calculations yielded the ATP concentration C in the sample to be tested. final =10.7 μM. This indicates that the trained and optimized nonlinear regression model can accurately predict the ATP concentration in water samples. Preliminary prediction C pred =10.5μM is the intensity I of the pure target light signal obtained by the trained model. clean Preliminary predictions were made using a value of 653, yielding preliminary concentration values. The optimized prediction value C... final =10.7μM was obtained by iteratively optimizing the model parameters, which further improved the accuracy of the prediction and yielded the final ATP concentration value.

[0224] This treatment method can significantly improve the accuracy and reliability of ATP concentration detection in water quality testing, ensuring the accuracy and consistency of test results.

[0225] Figure 2 This application provides a schematic diagram of the structure of a high-sensitivity ATP fluorescence detection system, as shown in the embodiment. Figure 2 As shown, the system includes:

[0226] Configuration module 21 is used to determine the optimal ratio of luciferase to luciferin, and to configure a reaction mixture containing the optimal ratio of luciferase to luciferin to ensure that a significantly enhanced fluorescence signal is generated under low ATP concentration conditions.

[0227] The mixing module 22 is used to control the reaction environment based on the reaction mixture using microfluidic chip technology, so that the reaction mixture and the sample to be tested are fully mixed in the dark environment. At the same time, dynamic temperature control technology is used to adjust the reaction temperature, improve the reaction rate and fluorescence signal stability, and obtain a mixed reaction system.

[0228] The separation measurement module 23 is used to capture the fluorescence signal generated by the mixed reaction system within a set time using a high-sensitivity photomultiplier tube, and to process the fluorescence signal using digital lock-in amplification technology and adaptive noise cancellation algorithm, to separate and generate a pure target fluorescence signal, and to measure the intensity of the pure target fluorescence signal at the same time.

[0229] The analysis and calculation module 24 is used to analyze the intensity of the pure target fluorescence signal based on the intensity of the pure target fluorescence signal, using a machine learning nonlinear regression model trained with a large dataset of ATP standard solutions of different concentrations, and optimizing the model parameters using a gradient descent algorithm. The module predicts and calculates the concentration value of ATP in the sample to be tested, ensuring the accuracy and reliability of the detection results.

[0230] Figure 2 The aforementioned high-sensitivity ATP fluorescence detection system can perform... Figure 1 The implementation principle and technical effects of the high-sensitivity ATP fluorescence detection method described in the illustrated embodiments will not be repeated here. The specific operation methods of each module and unit in the high-sensitivity ATP fluorescence detection system described above have been detailed in the embodiments related to this method, and will not be elaborated upon here.

[0231] In one possible design, Figure 2 The high-sensitivity ATP fluorescence detection system of the illustrated embodiment can be implemented as a computing device, such as... Figure 3 As shown, the computing device may include a storage component 31 and a processing component 32;

[0232] The storage component 31 stores one or more computer instructions, wherein the one or more computer instructions are invoked and executed by the processing component 32.

[0233] The processing component 32 is used to: determine the optimal ratio of luciferase to luciferin, and prepare a reaction mixture containing the optimal ratio of luciferase and luciferin to ensure a significantly enhanced fluorescence signal under low ATP concentration conditions; based on the reaction mixture, apply microfluidic chip technology to control the reaction environment, ensuring that the reaction mixture and the sample to be tested are fully mixed in a dark environment, while using dynamic temperature control technology to adjust the reaction temperature, improve the reaction rate and fluorescence signal stability, and obtain a mixed reaction system; use a high-sensitivity photomultiplier tube to capture the fluorescence signal generated by the mixed reaction system within a set time, and use digital lock-in amplification technology and adaptive noise cancellation algorithm to process the fluorescence signal, separate and generate a pure target fluorescence signal, and simultaneously measure the intensity of the pure target fluorescence signal; based on the intensity of the pure target fluorescence signal, use a machine learning nonlinear regression model trained with a large dataset of ATP standard solutions of different concentrations, use a gradient descent algorithm to optimize the model parameters, analyze the intensity of the pure target fluorescence signal, and calculate and generate the concentration value of ATP in the sample to be tested through model prediction, ensuring the accuracy and reliability of the detection results.

[0234] The processing component 32 may include one or more processors to execute computer instructions to complete all or part of the steps in the above-described method. Alternatively, the processing component may be implemented as one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field-programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to perform the above-described method.

[0235] Storage component 31 is configured to store various types of data to support operations at the terminal. The storage component can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0236] Of course, computing devices may also include other components, such as input / output interfaces, display components, communication components, etc.

[0237] Input / output interfaces provide interfaces between processing components and peripheral interface modules, which can be output devices, input devices, etc.

[0238] The communication components are configured to facilitate wired or wireless communication between computing devices and other devices.

[0239] The computing device can be a physical device or an elastic computing host provided by a cloud computing platform. In this case, the computing device can refer to a cloud server, and the aforementioned processing components, storage components, etc., can be basic server resources rented or purchased from the cloud computing platform.

[0240] This application also provides a computer storage medium storing a computer program, which, when executed by a computer, can perform the above-described functions. Figure 1 The embodiment shown illustrates a highly sensitive ATP fluorescence detection method.

[0241] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0242] The system embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0243] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0244] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. A highly sensitive method for detecting ATP fluorescence, characterized in that, include: Determine the optimal ratio of luciferase to luciferin, prepare a reaction mixture containing the optimal ratio of luciferase to luciferin, and ensure that a significantly enhanced fluorescence signal is generated under low ATP concentration conditions. Based on the reaction mixture, microfluidic chip technology is used to control the reaction environment, so that the reaction mixture and the sample to be tested are fully mixed in the dark environment. At the same time, dynamic temperature control technology is used to adjust the reaction temperature, improve the reaction rate and fluorescence signal stability, and obtain a mixed reaction system. A high-sensitivity photomultiplier tube is used to capture the fluorescence signal generated by the mixed reaction system within a set time. The fluorescence signal is then processed using digital lock-in amplification (DLA) and an adaptive noise cancellation algorithm to separate and generate a pure target fluorescence signal. Simultaneously, the intensity of the pure target fluorescence signal is measured. The process of processing the fluorescence signal using DLA and the adaptive noise cancellation algorithm to separate and generate the pure target fluorescence signal includes: based on the initially captured signal, using frequency-tracking DLA, analyzing the signal spectrum using a fast Fourier transform algorithm, selecting the frequency closest to the target fluorescence signal as a synchronization reference signal, and performing phase-locking processing on the initially captured signal to obtain the amplified target fluorescence signal; based on the amplified target fluorescence signal, using a deep learning-based adaptive noise cancellation algorithm, which monitors and analyzes the signal in real time through a convolutional neural network, automatically adjusts the algorithm parameters, and performs noise removal processing on the amplified target fluorescence signal to obtain the pure target fluorescence signal. Based on the intensity of the pure target fluorescence signal, a machine learning nonlinear regression model trained with a large dataset of ATP standard solutions of different concentrations is used. The model parameters are optimized using a gradient descent algorithm to analyze the intensity of the pure target fluorescence signal. The concentration of ATP in the test sample is predicted and calculated by the model, ensuring the accuracy and reliability of the detection results. The process of optimizing the model parameters using the gradient descent algorithm to analyze the intensity of the pure target fluorescence signal includes: using the trained nonlinear regression model to analyze the intensity of the pure target fluorescence signal; using the intensity of the pure target fluorescence signal as input; and using the model to predict and calculate the concentration of ATP in the test sample to obtain a preliminary concentration value.

2. The highly sensitive ATP fluorescence detection method according to claim 1, characterized in that, The method of capturing the fluorescence signal generated by the mixed reaction system within a set time using a high-sensitivity photomultiplier tube includes: Using a high-sensitivity photomultiplier tube, based on a dynamic integral time window of fluorescence signal intensity change, and combined with an adaptive threshold algorithm, the fluorescence signal generated by the mixed reaction system is efficiently captured to obtain a preliminary captured signal; the intensity of the pure target fluorescence signal is measured, including: The pure target fluorescence signal is digitized using a built-in high-precision analog-to-digital converter. A self-calibrating algorithm is used to correct the nonlinear error in the analog-to-digital conversion process, and the intensity of the pure target fluorescence signal is accurately measured to generate the final fluorescence signal intensity data.

3. The highly sensitive ATP fluorescence detection method according to claim 2, characterized in that, The method utilizes a high-sensitivity photomultiplier tube, based on a dynamic integral time window of fluorescence signal intensity changes, combined with an adaptive threshold algorithm, to efficiently capture the fluorescence signal generated by the mixed reaction system, obtaining a preliminary captured signal, including: Using a high-sensitivity photomultiplier tube, based on the average fluorescence signal intensity determined in a preliminary experiment, an initial integration time window is set to begin the initial capture of the fluorescence signal generated by the mixed reaction system, and the intensity of the initially captured fluorescence signal is obtained. Based on the intensity of the initially captured fluorescence signal, an adaptive threshold algorithm is used to evaluate the signal intensity in real time. When the signal intensity is detected to exceed the preset threshold range, the length of the integration time window is dynamically adjusted to obtain the optimized integration time window. Based on the optimized integration time window, a high-sensitivity photomultiplier tube is used to continuously capture the fluorescence signal generated by the mixed reaction system. The integration time window is updated immediately after each capture to ensure that the best signal capture performance is maintained throughout the detection process, thereby obtaining continuously captured fluorescence signals. All the continuously captured fluorescence signals are aggregated to form a preliminary capture signal, providing high-quality raw data for subsequent processing using frequency tracking-based digital lock-in amplification technology.

4. The highly sensitive ATP fluorescence detection method according to claim 2, characterized in that, The process involves using frequency-tracking-based digital phase-locked amplification technology based on the initial captured signal, analyzing the signal spectrum using a fast Fourier transform algorithm, selecting the frequency closest to the target fluorescence signal as a synchronization reference signal, and performing phase-locking processing on the initial captured signal to obtain the amplified target fluorescence signal, including: The Fast Fourier Transform algorithm is used to perform spectral analysis on the initially captured signal to identify multiple frequency components in the signal and obtain preliminary spectral analysis results. Based on the preset frequency range and signal strength threshold, the frequency closest to the target fluorescence signal is selected from the preliminary spectrum analysis results as the synchronization reference signal, and the selected synchronization reference signal is generated. Based on the selected synchronization reference signal, a digital phase-locked amplification technique based on frequency tracking is adopted. By adjusting the phase-locked loop parameters, the output signal of the phase-locked loop is kept in phase synchronization with the selected synchronization reference signal. The initial captured signal is then subjected to phase-locking processing to generate a phase-locked signal. The phase-locked signal is amplified using the adjusted amplifier gain to ensure a significant increase in the intensity of the target fluorescence signal while reducing the influence of background noise, thereby generating an amplified target fluorescence signal.

5. The highly sensitive ATP fluorescence detection method according to claim 2, characterized in that, Based on the amplified target fluorescence signal, an adaptive noise reduction algorithm based on deep learning is used. This algorithm, employing a convolutional neural network, monitors and analyzes the signal in real time, automatically adjusts the algorithm parameters, and performs noise removal processing on the amplified target fluorescence signal to obtain a clean target fluorescence signal. This includes: Based on the amplified target fluorescence signal, an adaptive noise cancellation model is constructed using a convolutional neural network. The adaptive noise cancellation model is used to monitor and analyze signal characteristics in real time. The amplified target fluorescence signal is analyzed in real time using the adaptive noise cancellation model to identify the target component and noise component in the signal and obtain the analysis results. Based on the analysis results, the parameters of the adaptive noise cancellation model are automatically adjusted to optimize the noise cancellation effect, ensuring that the noise components in the amplified target fluorescence signal are effectively removed, and generating an optimized noise cancellation model. The optimized noise cancellation model is used to remove noise from the amplified target fluorescence signal, generating a clean target fluorescence signal and providing a clean signal source for subsequent signal intensity measurements.

6. The highly sensitive ATP fluorescence detection method according to claim 5, characterized in that, Based on the amplified target fluorescence signal, an adaptive noise reduction algorithm based on deep learning is used. This algorithm, employing a convolutional neural network, monitors and analyzes the signal in real time, automatically adjusts the algorithm parameters, and performs noise removal processing on the amplified target fluorescence signal to obtain a clean target fluorescence signal. This includes: Based on the amplified target fluorescence signal S amp An adaptive noise cancellation model M is constructed using a convolutional neural network. CNN The adaptive noise cancellation model M CNN Used for real-time monitoring and analysis of signal characteristics; Using the adaptive noise cancellation model M CNN The amplified target fluorescence signal S amp Perform real-time analysis to identify target components in the signal. Based on the analysis result R, the adaptive noise cancellation model M is automatically adjusted. CNN The parameters are optimized to improve noise cancellation and ensure the amplified target fluorescence signal S. amp The noise component N is effectively removed to generate an optimized noise cancellation model M′. CNN ; The optimized noise cancellation model M′ is calculated using the following formula. CNN : M′ CNN =M CNN +W·ΔP Where ΔP is the parameter increment automatically adjusted based on the analysis result R, and W is the weighting coefficient, which is expressed as: Where α is the sensitivity coefficient, S threshold It is a preset target component intensity threshold. This formula ensures that when the target component intensity S... target Approaching or exceeding the threshold S threshold When the weighting coefficient W approaches 1, the adjustment of the parameters is increased. Using the optimized noise cancellation model M′ CNN The amplified target fluorescence signal S amp Noise removal is performed to generate a pure target fluorescence signal S. clean This provides a clean signal source for subsequent signal strength measurements; The noise removal formula is calculated using the following formula: S clean =S amp -β·N Where β is the noise removal coefficient, which is expressed as: Where γ is the exponential factor, this formula ensures that when the target component intensity S... target When the noise component N is relatively strong, the noise removal coefficient β is close to 1, thus removing noise more effectively.

7. The highly sensitive ATP fluorescence detection method according to claim 2, characterized in that, The process involves using a built-in high-precision analog-to-digital converter to digitize the purified target fluorescence signal, employing a self-calibrating algorithm to correct nonlinear errors during the analog-to-digital conversion, accurately measuring the intensity of the purified target fluorescence signal, and generating final fluorescence signal intensity data. This includes: The pure target fluorescence signal is digitally processed by the built-in high-precision analog-to-digital converter, converting the analog signal into a digital signal to obtain a digital pure target fluorescence signal. Based on the digitally purified target fluorescence signal, a self-calibration algorithm is used to analyze the nonlinear errors that may exist in the analog-to-digital conversion process, including quantization error and gain error, and the error analysis results are obtained. Based on the error analysis results, the parameters of the self-calibration algorithm are adjusted to correct the nonlinear error in the analog-to-digital conversion process, optimize the quality of the digital signal, and generate the calibrated digital signal. Using the corrected digital signal, the intensity of the pure target fluorescence signal is accurately measured to generate the final fluorescence signal intensity data, ensuring the accuracy and reliability of the measurement results.

8. The highly sensitive ATP fluorescence detection method according to claim 1, characterized in that, The method involves analyzing the intensity of the pure target fluorescence signal using a machine learning nonlinear regression model trained on a large dataset of ATP standard solutions with varying concentrations. The model parameters are optimized using a gradient descent algorithm. The model predicts and calculates the concentration of ATP in the sample to be tested, ensuring the accuracy and reliability of the detection results. This includes: Using a dataset of a large number of ATP standard solutions of different concentrations, which includes ATP standard solutions of known concentrations and their corresponding fluorescence signal intensities, the dataset is preprocessed to obtain a preprocessed dataset. Based on the preprocessed dataset, a machine learning nonlinear regression model is constructed, and the gradient descent algorithm is used to optimize the model parameters to ensure that the model can accurately reflect the nonlinear relationship between fluorescence signal intensity and ATP concentration, thereby generating a trained nonlinear regression model. Using the trained nonlinear regression model, the intensity of the pure target fluorescence signal is analyzed. Taking the intensity of the pure target fluorescence signal as input, the concentration of ATP in the sample to be tested is calculated by the model to obtain a preliminary concentration value. By iteratively optimizing the model parameters, the accuracy and reliability of the model predictions are verified, and the concentration value of ATP in the sample to be tested is generated to ensure the accuracy and reliability of the detection results.

9. The highly sensitive ATP fluorescence detection method according to claim 8, characterized in that, The method involves analyzing the intensity of the pure target fluorescence signal using a machine learning nonlinear regression model trained on a large dataset of ATP standard solutions with varying concentrations. The model parameters are optimized using a gradient descent algorithm. The model predicts and calculates the concentration of ATP in the sample to be tested, ensuring the accuracy and reliability of the detection results. This includes: Using a large dataset of ATP standard solutions of varying concentrations, the dataset includes ATP standard solutions C of known concentrations. i and its corresponding fluorescence signal intensity I i The dataset is preprocessed to obtain the preprocessed dataset (C). i I i ); Based on the preprocessed dataset (C) i I i A machine learning nonlinear regression model f(I; θ) is constructed, and the model parameters are optimized using the gradient descent algorithm to ensure that the model can accurately reflect the nonlinear relationship between fluorescence signal intensity I and ATP concentration C, thus generating a trained nonlinear regression model f. trained (I;θ); The parameter update formula for the gradient descent algorithm is as follows: Where, θ j It is the j-th parameter in the model parameter vector θ, where θ = (θ0, θ1, ..., θj). n () represents the model's parameter vector, α is the learning rate, m is the number of samples in the dataset, and I ij It is the j-th feature value of the i-th sample; Using the trained nonlinear regression model f trained (I; θ), representing the intensity I of the pure target fluorescence signal. clean The intensity I of the pure target fluorescence signal was analyzed. clean As input, the concentration C of ATP in the sample to be tested is calculated through model prediction. pred The preliminary concentration value C was obtained. pred ; The prediction formula is as follows: C pred =f trained (I clean ;θ) By iteratively optimizing the model parameters, the accuracy and reliability of the model predictions were verified, and the concentration value C of ATP in the sample to be tested was generated. final This ensures the accuracy and reliability of the test results; The optimized formula is as follows: C final =f trained (I clean ;θ opt ) Where, θ opt It is the optimized model parameter vector; C i It is a standard ATP solution of known concentration, I i It corresponds to the fluorescence signal intensity, (C i I i ) is the preprocessed dataset, θ = (θ0, θ1, ..., θ) n ) is the model's parameter vector, α is the learning rate, m is the number of samples in the dataset, and I ij Let f(I;θ) be the j-th feature value of the i-th sample, and f(I;θ) be a nonlinear regression model. trained (I; θ) is a trained nonlinear regression model, where I... clean It is the intensity of the pure target fluorescence signal, C pred This is the preliminary concentration value, C final It is the concentration of ATP in the final generated test sample, θ opt It is the optimized model parameter vector.

10. A highly sensitive ATP fluorescence detection system, characterized in that, include: The configuration module is used to determine the optimal ratio of luciferase to luciferin, and to configure a reaction mixture containing the optimal ratio of luciferase to luciferin to ensure that a significantly enhanced fluorescence signal is generated under low ATP concentration conditions. The mixing module is used to control the reaction environment based on the reaction mixture using microfluidic chip technology, so that the reaction mixture and the sample to be tested are fully mixed in the dark environment. At the same time, dynamic temperature control technology is used to adjust the reaction temperature, improve the reaction rate and fluorescence signal stability, and obtain a mixed reaction system. The separation and measurement module is used to capture the fluorescence signal generated by the mixed reaction system within a set time using a high-sensitivity photomultiplier tube, and to process the fluorescence signal using digital lock-in amplification technology and an adaptive noise cancellation algorithm to separate and generate a pure target fluorescence signal, while simultaneously measuring the intensity of the pure target fluorescence signal. The process of processing the fluorescence signal using digital lock-in amplification technology and an adaptive noise cancellation algorithm to separate and generate a pure target fluorescence signal includes: based on the initially captured signal, using frequency tracking-based digital lock-in amplification technology, analyzing the signal spectrum using a fast Fourier transform algorithm, selecting the frequency closest to the target fluorescence signal as a synchronization reference signal, and performing phase-locking processing on the initially captured signal to obtain an amplified target fluorescence signal; based on the amplified target fluorescence signal, using a deep learning-based adaptive noise cancellation algorithm, monitoring and analyzing the signal in real time through a convolutional neural network, automatically adjusting the algorithm parameters, and performing noise removal processing on the amplified target fluorescence signal to obtain a pure target fluorescence signal. The analysis and calculation module is used to analyze the intensity of the pure target fluorescence signal based on the intensity of the pure target fluorescence signal. It employs a machine learning nonlinear regression model trained on a large dataset of ATP standard solutions of different concentrations, optimizes the model parameters using a gradient descent algorithm, and predicts and calculates the concentration of ATP in the sample to be tested, ensuring the accuracy and reliability of the detection results. The process of optimizing the model parameters using the gradient descent algorithm to analyze the intensity of the pure target fluorescence signal includes: using the trained nonlinear regression model to analyze the intensity of the pure target fluorescence signal; using the intensity of the pure target fluorescence signal as input; and predicting and calculating the concentration of ATP in the sample to be tested to obtain a preliminary concentration value.

Images