Refrigerator bacterium detection method and system based on optical fiber ATP sensor

Abstract

The embodiment of the invention discloses a refrigerator bacterium detection method and system based on an optical fiber ATP sensor, and the method comprises the steps: mixing a sample in a to-be-detected area of a refrigerator with a reaction reagent to form a to-be-detected solution, injecting the to-be-detected solution into a microfluidic reaction tank of the optical fiber ATP sensor, enabling an ATP component and an enzyme reagent to generate a chemiluminescence reaction, and enabling an optical signal to enter a transmission optical fiber through focusing coupling; the optical signal is transmitted to a photoelectric detection unit through a transmission optical fiber to be converted into an electric signal, and the electric signal contains response dynamic information; carrying out multi-stage processing on the electric signal, and extracting a signal characteristic parameter representing the concentration of the ATP component; and finally, performing matching analysis on the signal characteristic parameters and preset data, determining an ATP component content value, querying corresponding relation data to generate a detection result containing bacteria quantity information, and outputting the detection result to a refrigerator matched display device in a visual form. The problems of long detection time, complicated operation, inconvenience in equipment, difficulty in in-situ real-time monitoring and the like in the existing refrigerator bacterium detection technology can be solved.

Description

Technical Field

[0001] This invention relates to the field of data analysis technology, specifically to a method and system for detecting bacteria in a refrigerator based on a fiber optic ATP sensor. Background Technology

[0002] As widely used refrigeration equipment in homes and businesses, the bacterial growth inside refrigerators directly affects food safety and quality. Traditional methods for detecting bacteria in refrigerators, such as microbial culture, are complex and time-consuming, often requiring hours or even days to obtain results, and demanding high standards for the operating environment and technical personnel. Furthermore, culture methods can only detect bacteria that can grow on specific culture media, easily missing some difficult-to-culture or slow-growing bacteria.

[0003] ATP (adenosine triphosphate) bioluminescence assay has seen some application in recent years as a rapid method for detecting microbial contamination. However, existing ATP detection technologies mostly employ traditional bulk photoelectric detection devices, which suffer from problems such as large size and inconvenience in portability, making it difficult to meet the needs for in-situ, real-time monitoring of bacteria inside refrigerators. Summary of the Invention

[0004] This invention provides a method and system for detecting bacteria in a refrigerator based on an optical fiber ATP sensor.

[0005] In a first aspect, embodiments of the present invention provide a method for detecting bacteria in a refrigerator based on a fiber optic ATP sensor, applied to a refrigerator bacteria detection system based on a fiber optic ATP sensor, the method comprising: The sample collected from the area to be tested inside the refrigerator is mixed with the reaction reagent to form the test solution. The test solution is injected into the microfluidic reaction cell of the fiber optic ATP sensor, so that the ATP component in the test solution reacts with the enzyme reagent fixed on the sensor surface to produce a chemiluminescent reaction. The generated light signal is focused by the optical lens integrated by the sensor, and the focused light signal is coupled into the incident end face of the transmission optical fiber. The coupled optical signal is transmitted to the photoelectric detection unit through the transmission optical fiber. The photoelectric detection unit converts the received optical signal into a corresponding electrical signal. The waveform characteristics of the electrical signal correspond to the intensity change trend of the optical signal. The electrical signal contains dynamic change information reflecting the chemiluminescence reaction process. The electrical signal is subjected to multi-stage signal processing, including signal amplification, noise suppression, and feature extraction, to obtain signal characteristic parameters characterizing the concentration of ATP components. The variation of the signal characteristic parameters is correlated with the ATP content in the solution to be detected. The signal feature parameters are matched and analyzed with preset ATP content correlation data to determine the ATP component content value corresponding to the signal feature parameters. Based on the ATP component content value, preset ATP content and bacterial quantity correspondence data are queried to generate a detection result containing bacterial quantity information. The detection result is output to the refrigerator's matching display device in a visual form.

[0006] Secondly, embodiments of the present invention provide a refrigerator bacterial detection system based on a fiber optic ATP sensor, comprising: processor; Storage device, on which computer programs are stored, When the computer program is executed by the processor, the processor implements any of the described methods for detecting bacteria in a refrigerator based on a fiber optic ATP sensor.

[0007] This invention provides a readable storage medium storing a program or instructions, which, when executed by a processor, implement the steps of the refrigerator bacteria detection method based on an optical fiber ATP sensor.

[0008] Therefore, the embodiments of the present invention can improve the problems of long detection time, complicated operation, inconvenient equipment and difficulty in in-situ real-time monitoring in existing refrigerator bacterial detection technologies, and realize rapid, accurate and convenient detection of bacteria in refrigerators. Attached Figure Description

[0009] Figure 1 This is a flowchart of a refrigerator bacteria detection method based on an optical fiber ATP sensor, provided in an embodiment of the present invention.

[0010] Figure 2 This is a schematic diagram illustrating the principle of refrigerator bacterial detection based on an optical fiber ATP sensor, as provided in an embodiment of the present invention.

[0011] Figure 3 This is a schematic diagram of the basic structure of a refrigerator bacterial detection system based on an optical fiber ATP sensor, provided in an embodiment of the present invention. Detailed Implementation

[0012] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the embodiments of the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0013] See Figure 1 As shown in the figure, this is a flowchart of a refrigerator bacteria detection method based on a fiber optic ATP sensor provided by an embodiment of the present invention. This method can be applied to a refrigerator bacteria detection system based on a fiber optic ATP sensor. Figure 1As shown, the method may include steps 110-140.

[0014] Step 110: The sample collected from the area to be tested inside the refrigerator is mixed with the reaction reagent to form a test solution. The test solution is injected into the microfluidic reaction cell of the fiber optic ATP sensor, so that the ATP component in the test solution reacts with the enzyme reagent fixed on the sensor surface to produce a chemiluminescent reaction. The generated light signal is focused by the optical lens integrated by the sensor, and the focused light signal is coupled into the incident end face of the transmission optical fiber.

[0015] In this embodiment, the fiber optic ATP sensor is first prepared for refrigerator bacterial detection. Prepare luciferase and luciferin with a purity greater than 99%, analytical grade tetraethyl orthosilicate, ethanol, deionized water, hydrochloric acid, and other reagents. Add a certain amount of tetraethyl orthosilicate to a mixed solution of ethanol and deionized water, stir until homogeneous, then add an appropriate amount of hydrochloric acid as a catalyst to adjust the pH to 3-4. Stir the reaction at room temperature for 2-3 hours to obtain a transparent sol solution. Next, add luciferase and luciferin to the sol solution in a certain proportion and continue stirring for 1-2 hours. Cut the quartz fiber to a suitable length, ultrasonically clean it sequentially with ethanol and deionized water for 15-20 minutes, then place it in a high-temperature furnace and calcine at 500-600°C for 1-2 hours to activate the surface. The pretreated fiber optic probe was immersed in the sol-gel precursor solution at a speed of 5-10 mm / s and slowly pulled up. The coating was repeated 3-5 times. After each coating, the probe was dried at room temperature for 1-2 hours. Finally, it was cured at 80-100℃ for 2-3 hours to obtain the fiber optic ATP sensor.

[0016] Then, sample collection is performed using nylon fiber sampling swabs with a diameter of 3-5 mm and surface modification. Before use, these swabs are soaked in a sterile aqueous solution containing 0.1% Triton X-100 for 1-2 hours and then air-dried. The refrigerator door is opened, and the sampling swabs are used to swab the inner walls, shelves, and food surfaces of the areas to be tested. Each swab area should be at least 5 cm², and the swabs should be applied at least 10 times. After sampling, the swabs are placed in a sterile test tube containing lysis buffer. The lysis buffer is prepared by weighing 0.5 g Triton X-100, 0.1 g NaCl, 0.05 g KCl, and 0.02 g MgCl2, adding them to 100 mL of deionized water, stirring thoroughly, and adjusting the pH to 7.0-7.5. The test tube containing the sampling swab and lysis buffer is placed in an ultrasonic cleaner and sonicated for 5-10 minutes to rupture bacterial cells and release ATP. The tube is then centrifuged at 3000-4000 rpm for 5-10 minutes. The supernatant is collected and mixed with reaction reagents to form the test solution. This test solution is injected into the microfluidic reaction cell of the fiber optic ATP sensor. Under conditions of approximately 25°C and pH 7.0-7.5, the ATP in the test solution reacts with luciferase and luciferin immobilized on the fiber optic surface to produce a bioluminescent reaction. The generated light signal is collected by a focusing lens and transmitted through the fiber optic cable. The focused light signal is then coupled into the incident end face of the transmission fiber.

[0017] Step 120: The coupled optical signal is transmitted to the photoelectric detection unit through the transmission optical fiber. The photoelectric detection unit converts the received optical signal into a corresponding electrical signal. The waveform characteristics of the electrical signal correspond to the intensity change trend of the optical signal. The electrical signal contains dynamic change information reflecting the chemiluminescence reaction process.

[0018] Step 121: Align the output end face of the transmission optical fiber with the photosensitive element of the photoelectric detection unit, so that the optical signal output from the transmission optical fiber is perpendicularly incident on the photosensitive area of ​​the photosensitive element. The photosensitive element converts the photon energy of the optical signal into charge carriers, and accumulates the charge carriers through the internal circuit to form an initial electrical signal.

[0019] In this embodiment, to ensure that the optical signal can be accurately received and converted by the photodetector unit, the emitting end face of the transmission optical fiber must be precisely aligned with the photomultiplier tube of the photodetector unit. When the optical signal is incident perpendicularly on the photosensitive area of ​​the photomultiplier tube, the photons of the optical signal will excite electrons to generate charge carriers. The internal circuit of the photomultiplier tube will accumulate these charge carriers to form an initial electrical signal. The magnitude of the initial electrical signal is related to the intensity of the optical signal. The greater the intensity of the optical signal, the more charge carriers are generated, and the greater the amplitude of the initial electrical signal. In this way, the optical signal is converted into an electrical signal by the photomultiplier tube.

[0020] Step 122: Perform pre-amplification processing on the initial electrical signal. Use a low-noise operational amplifier to increase the amplitude of the initial electrical signal to the effective input range of the subsequent processing circuit to obtain the amplified primary electrical signal.

[0021] In this embodiment, the initial electrical signal has a small amplitude and cannot be effectively processed by subsequent processing circuits. A low-noise operational amplifier (LNO) minimizes noise introduction during signal amplification. By amplifying the initial electrical signal with an LNO, its amplitude is increased to a range that subsequent processing circuits can effectively handle, resulting in an amplified primary electrical signal. For example, the initial electrical signal might have an amplitude in the microvolt range; after amplification by the LNO, the amplitude can be increased to the millivolt range, meeting the input requirements of subsequent processing circuits.

[0022] Step 123: The primary electrical signal is processed by a bandpass filter circuit to select the frequency, retaining the characteristic frequency components related to the chemiluminescence reaction and filtering out the interference frequency components caused by ambient light and circuit noise, to obtain the filtered secondary electrical signal.

[0023] In this embodiment, the primary electrical signal contains interfering frequency components such as ambient light and circuit noise, which can affect the subsequent analysis of the chemiluminescence reaction signal. The bandpass filter circuit has a frequency selection function, allowing only characteristic frequency components related to the chemiluminescence reaction to pass through, while filtering out other interfering frequency components. For example, ambient light may generate interference signals at specific frequencies, and circuit noise also has its specific frequency range. The bandpass filter circuit can identify and filter out these interference signals, retaining only the frequency components related to the chemiluminescence reaction, making the filtered secondary electrical signal purer and more accurately reflecting the information of the chemiluminescence reaction.

[0024] Step 124: Perform analog-to-digital conversion on the secondary electrical signal to convert the analog electrical signal into a digital electrical signal sequence. The sampling interval of the electrical signal sequence matches the dynamic change period of the chemiluminescence reaction. The electrical signal sequence is an electrical signal that contains dynamic change information reflecting the chemiluminescence reaction process.

[0025] In this embodiment, the analog secondary electrical signal needs to be converted into a digital electrical signal sequence. During analog-to-digital conversion, the sampling interval of the electrical signal sequence must match the dynamic change period of the chemiluminescence reaction. If the dynamic change period of the chemiluminescence reaction is several milliseconds, then the sampling interval of the electrical signal sequence should also be set to several milliseconds, thereby accurately capturing the dynamic change information during the chemiluminescence reaction process.

[0026] Step 130: Perform multi-stage signal processing on the electrical signal, sequentially performing signal amplification, noise suppression, and feature extraction to obtain signal characteristic parameters characterizing the concentration of ATP components. The variation law of the signal characteristic parameters is related to the ATP content in the solution to be detected.

[0027] Step 131: Input the electrical signal into a programmable gain amplifier for dynamic gain adjustment. The amplification factor is automatically adjusted according to the initial amplitude of the electrical signal so that the amplitude of the amplified electrical signal is in the preset optimal processing range, thereby obtaining the gain-adjusted electrical signal.

[0028] In this embodiment, the programmable gain amplifier plays a crucial role in processing the electrical signal. First, a peak detection circuit monitors the amplitude of the input electrical signal in real time, capturing the maximum amplitude within a preset time window as the initial amplitude. A preset optimal processing range has clearly defined upper and lower thresholds. The initial amplitude is compared to these thresholds. If the initial amplitude is below the lower threshold, it indicates that the electrical signal amplitude is too small; in this case, the gain of the programmable gain amplifier is increased to increase the signal amplitude. If the initial amplitude is above the upper threshold, it indicates that the electrical signal amplitude is too large; in this case, the gain of the programmable gain amplifier is decreased. If the initial amplitude is within the threshold range, the current gain remains unchanged. The electrical signal is amplified according to the adjusted gain, thus limiting the amplitude variation range of the gained electrical signal to within the optimal processing range.

[0029] Step 1311: Monitor the amplitude of the input electrical signal in real time. Capture the maximum amplitude of the electrical signal within a preset time window through a peak detection circuit, and use the maximum amplitude as the initial amplitude.

[0030] In this embodiment, the peak detection circuit can monitor the amplitude changes of the input electrical signal in real time. Within a preset time window, the amplitude of the electrical signal is continuously detected and compared to find the maximum amplitude, which is then determined as the initial amplitude. The preset time window is set to acquire the characteristic information of the electrical signal within a suitable time period, avoiding the accuracy of the initial amplitude being affected by a time that is too long or too short. For example, setting a time window at a specific stage of the chemiluminescence reaction and capturing the maximum amplitude of the electrical signal within that window can more accurately reflect the intensity characteristics of the electrical signal at that stage.

[0031] Step 1312: Compare the initial amplitude with the upper and lower thresholds of the preset optimal processing range. If the initial amplitude is lower than the lower threshold, increase the gain of the programmable gain amplifier. If the initial amplitude is higher than the upper threshold, decrease the gain of the programmable gain amplifier. If the initial amplitude is within the threshold range, maintain the current gain.

[0032] In this embodiment, the preset optimal processing range has clearly defined upper and lower thresholds. These two thresholds are determined based on the requirements of subsequent signal processing and the performance of the circuit. The previously obtained initial amplitude is compared with these two thresholds. If the initial amplitude is lower than the lower threshold, it indicates that the amplitude of the electrical signal is too small, and the subsequent processing circuit may not be able to accurately process such a small signal. Therefore, it is necessary to increase the gain of the programmable gain amplifier to increase the amplitude of the electrical signal so that it can meet the requirements of subsequent processing. Conversely, if the initial amplitude is higher than the upper threshold, the amplitude of the electrical signal is too large, which may cause saturation or distortion in the subsequent processing circuit. Therefore, the gain of the programmable gain amplifier should be reduced. If the initial amplitude is within the threshold range, it indicates that the current amplitude of the electrical signal is appropriate, and it is not necessary to adjust the gain. Maintaining the current gain is sufficient to keep the electrical signal within a suitable amplitude range.

[0033] Step 1313: Amplify the electrical signal according to the adjusted gain factor to obtain a gain-adjusted electrical signal. The amplitude variation range of the gain-adjusted electrical signal is limited to the optimal processing range.

[0034] In this embodiment, after determining a suitable gain factor, the programmable gain amplifier amplifies the electrical signal according to the adjusted gain factor. Through amplification, the amplitude of the electrical signal changes accordingly. Since the gain factor has already been adjusted based on the initial amplitude, the amplitude variation range of the amplified, gain-adjusted electrical signal is limited to a preset optimal processing range. This ensures that the electrical signal will not be inaccurately detected and analyzed due to excessively small amplitude, nor will it cause saturation or distortion in the processing circuit due to excessively large amplitude, thus improving the accuracy and reliability of signal processing.

[0035] Step 132: Perform wavelet threshold denoising on the gain-adjusted electrical signal. Decompose the electrical signal into wavelet coefficients of different frequency bands through wavelet decomposition. Perform threshold truncation on the high-frequency wavelet coefficients, retain the effective coefficients that reflect the signal characteristics, and reconstruct the denoised electrical signal.

[0036] In this embodiment, the gain-adjusted electrical signal may still contain some noise interference. To improve signal quality, wavelet threshold denoising is required. Wavelet decomposition decomposes the electrical signal into wavelet coefficients of different frequency bands, which contain signal information of different scales and frequencies. High-frequency wavelet coefficients typically contain more noise components. Threshold truncation of these high-frequency wavelet coefficients involves setting a threshold, setting high-frequency wavelet coefficients below the threshold to zero, and retaining only high-frequency wavelet coefficients above the threshold, as well as other effective coefficients reflecting signal characteristics. Then, signal reconstruction is performed based on these retained effective coefficients to obtain the denoised electrical signal. Wavelet threshold denoising effectively removes noise from the electrical signal, making the denoised electrical signal more accurately reflect the true information of the chemiluminescence reaction.

[0037] Step 133: Perform time-domain feature extraction processing on the denoised electrical signal to extract the peak amplitude, rising edge rate, falling edge rate, and duration of the signal. Combine these time-domain feature parameters to form signal feature parameters characterizing the concentration of ATP components.

[0038] Step 1331: Perform waveform analysis on the denoised electrical signal, identify the maximum amplitude point in the electrical signal waveform, and record the voltage value corresponding to the maximum amplitude point as the peak amplitude.

[0039] In this embodiment, waveform analysis of the denoised electrical signal is a crucial step in extracting signal features. Detailed observation and analysis of the electrical signal waveform allows identification of the maximum amplitude point, which represents the maximum intensity of the signal throughout its variation. The voltage value corresponding to this maximum amplitude point is recorded as the peak amplitude, which is closely related to the intensity of the chemiluminescence reaction. Because the intensity of the chemiluminescence reaction is related to the concentration of ATP in the solution being tested, the peak amplitude can serve as an important characteristic parameter reflecting the ATP concentration. In actual detection, different concentrations of ATP may lead to different peak amplitudes; accurate measurement and analysis of the peak amplitude are essential for accurate detection.

[0040] Step 1332: Determine the rising phase from the starting point to the peak amplitude point in the electrical signal waveform, calculate the time interval and amplitude change of the rising phase, and use the ratio of amplitude change to time interval as the signal rising edge change rate.

[0041] In this embodiment, identifying the rise phase from the starting point to the peak amplitude point in the denoised electrical signal waveform is fundamental to calculating the signal rise edge rate. First, the starting point and peak amplitude point of the electrical signal must be accurately determined. Then, the time interval between these two points and the amplitude change are calculated. The time interval reflects the time taken for the chemiluminescent reaction to reach its maximum intensity from the start, while the amplitude change reflects the increase in signal intensity during that time interval. The ratio of the amplitude change to the time interval is used as the signal rise edge rate, reflecting the rate of change in the initial stage of the chemiluminescent reaction. Different concentrations of ATP react with the enzyme reagent immobilized on the sensor surface, potentially leading to different signal rise edge rates. Therefore, the signal rise edge rate can serve as an important time-domain characteristic parameter to characterize the concentration of ATP in the solution being detected.

[0042] Step 1333: Determine the falling phase from the peak amplitude point to the end point of the signal in the electrical signal waveform, calculate the time interval and amplitude change of the falling phase, and use the ratio of amplitude change to time interval as the falling edge rate of the signal.

[0043] In this embodiment, determining the descent phase from the peak amplitude point to the signal termination point in the denoised electrical signal waveform is crucial for analyzing the characteristics of the chemiluminescence reaction termination phase. After accurately locating the peak amplitude point and the signal termination point, the time interval and amplitude change between these two points are calculated. The time interval represents the time elapsed from the chemiluminescence reaction's maximum intensity to its termination, while the amplitude change reflects the magnitude of the signal intensity decrease during this time interval. The ratio of the amplitude change to the time interval is used as the signal descent rate, reflecting the changes during the chemiluminescence reaction termination phase. Different concentrations of ATP may exhibit different signal descent rates at the reaction termination, making the signal descent rate also an important temporal characteristic parameter.

[0044] Step 1334: Identify the start time when the amplitude of the electrical signal waveform exceeds a preset start threshold and the end time when the amplitude is lower than a preset end threshold, and calculate the time difference between the start time and the end time as the signal duration.

[0045] In this embodiment, to accurately calculate the signal duration, it is necessary to identify the start time (where the amplitude exceeds a preset start threshold) and the end time (where the amplitude falls below a preset end threshold) in the denoised electrical signal waveform. The preset start and end thresholds are determined based on experimental and actual detection requirements and are used to define the start and end times of the chemiluminescent reaction. By identifying these two times, the time difference between them is calculated; this time difference is the signal duration. The signal duration is related to the entire chemiluminescent reaction process. Different concentrations of ATP reacting with the enzyme reagent immobilized on the sensor surface may result in different signal durations. Therefore, the signal duration can serve as an important time-domain characteristic parameter reflecting the concentration of ATP in the solution to be detected.

[0046] Step 1335: Arrange the peak amplitude, signal rise edge change rate, signal fall edge change rate, and signal duration in a preset order to form signal characteristic parameters.

[0047] In this embodiment, the four extracted time-domain feature parameters—peak amplitude, rate of change of signal rising edge, rate of change of signal falling edge, and duration of signal—are arranged in a preset order to form signal feature parameters. This preset order is determined based on the needs of actual detection and data analysis; this arrangement combines these feature parameters into an ordered whole, facilitating subsequent processing and analysis. Since these time-domain feature parameters are all related to the concentration of ATP in the solution being tested, the signal feature parameters formed by combining them can more comprehensively and accurately reflect the concentration information of ATP. In the subsequent ATP content calculation and bacterial count conversion processes, the signal feature parameters will be processed and analyzed as key data.

[0048] Step 140: Match and analyze the signal feature parameters with preset ATP content correlation data to determine the ATP component content value corresponding to the signal feature parameters. Based on the ATP component content value, query the preset ATP content and bacterial quantity correspondence data to generate a detection result containing bacterial quantity information. Output the detection result in a visual form to the refrigerator's matching display device.

[0049] Step 141: Call the preset ATP content correlation data in the storage unit. The ATP content correlation data contains multiple sets of standard signal feature parameters corresponding to known ATP content. Each set of standard signal feature parameters corresponds to a set ATP content.

[0050] In this embodiment, the storage unit pre-stores ATP content correlation data, which was obtained during the experimental phase by detecting a series of ATP standard solutions of different concentrations. Each set of data contains standard signal characteristic parameters corresponding to a known ATP content, and each set of standard signal characteristic parameters corresponds to a set ATP content. During actual detection, this ATP content correlation data in the storage unit is retrieved to provide basic data for subsequent matching analysis. By comparing and matching the signal characteristic parameters obtained from actual detection with these standard signal characteristic parameters, the ATP content in the solution to be tested can be determined.

[0051] Step 142: Calculate the similarity between the signal feature parameters and each set of standard signal feature parameters, and select the ATP content corresponding to the standard signal feature parameter with the highest similarity as the preliminary matching content value.

[0052] Step 1421: Compare the signal feature parameters with each set of standard signal feature parameters and calculate the differences in peak amplitude, rising edge rate, falling edge rate, and duration.

[0053] In this embodiment, when calculating the similarity between the signal feature parameters and each set of standard signal feature parameters, comparisons and difference calculations are performed in four dimensions. In the peak amplitude dimension, the actual detected peak amplitude is compared with the peak amplitude in each set of standard signal feature parameters, and the difference is calculated. This difference reflects the degree of difference in peak amplitude. In the signal rise time dimension, the difference between the actual signal rise time and the rate of change in the standard signal feature parameters is calculated, reflecting the difference in the rate of change during the rise phase of the chemiluminescence reaction. Similarly, in the signal fall time dimension, the difference between the two rates is calculated, reflecting the difference in the change during the final phase of the chemiluminescence reaction. In the signal duration dimension, the difference between the actual signal duration and the duration in the standard signal feature parameters is calculated, reflecting the difference in the duration of the entire chemiluminescence reaction process. By calculating the difference values ​​in these four dimensions, a comprehensive comparison of the differences between the signal feature parameters and the standard signal feature parameters can be achieved.

[0054] Step 1422: Normalize the difference values ​​for each dimension, converting them into standardized difference values ​​with the same units. The magnitude of the standardized difference values ​​reflects the degree of similarity in that dimension.

[0055] In this embodiment, since the difference values ​​calculated from different dimensions may have different dimensions, it is necessary to normalize the difference values ​​for each dimension in order to fairly compare the similarity of each dimension. Normalization converts the difference values ​​of each dimension into standardized difference values ​​with the same dimension. For example, the difference value of peak amplitude may be in voltage units, and the difference value of signal rise time rate may be in voltage change rate units. Normalization can unify these difference values ​​with different dimensions into the same dimensional system. The magnitude of the standardized difference value directly reflects the degree of similarity in that dimension; the smaller the standardized difference value, the more similar the actually detected signal feature parameters are to the standard signal feature parameters in that dimension.

[0056] Step 1423: The standardized difference values ​​of each dimension are weighted and summed according to a preset weight ratio to obtain a comprehensive difference value. The smaller the comprehensive difference value, the higher the similarity of the two sets of signal feature parameters.

[0057] In this embodiment, after obtaining the standardized difference values ​​for each dimension, they need to be weighted and summed according to a preset weight ratio to obtain a comprehensive difference value. The preset weight ratio is determined based on the importance of each dimension in reflecting ATP component concentration. For example, peak amplitude may be more important in reflecting ATP component concentration, so its corresponding weight ratio may be relatively high. By multiplying the standardized difference values ​​of each dimension by their corresponding weight ratios and then summing them, the resulting comprehensive difference value can comprehensively consider the information from each dimension and more fully reflect the similarity between the two sets of signal feature parameters. The smaller the comprehensive difference value, the smaller the difference between the actual detected signal feature parameters and the standard signal feature parameters in each dimension, that is, the higher the similarity between the two sets of signal feature parameters.

[0058] Step 1424: Iterate through the comprehensive difference values ​​corresponding to all standard signal feature parameters, select the set of standard signal feature parameters with the smallest comprehensive difference value, and determine the ATP content associated with this set of standard signal feature parameters as the preliminary matching content value.

[0059] In this embodiment, after calculating the comprehensive difference values ​​corresponding to all standard signal feature parameters, it is necessary to iterate through these comprehensive difference values. The set of standard signal feature parameters with the smallest comprehensive difference value is identified through comparison. Since a smaller comprehensive difference value indicates a higher similarity between the two sets of signal feature parameters, this set of standard signal feature parameters with the smallest comprehensive difference value is most similar to the actually detected signal feature parameters. The ATP content associated with this set of standard signal feature parameters is determined as the preliminary matching content value, which is based on the similarity of the signal feature parameters.

[0060] Step 143: Correct the preliminary matching content value by combining the weight coefficients of each time-domain feature in the signal feature parameters to adjust the preliminary matching content value and obtain the target ATP component content value.

[0061] Step 1431: Obtain the preset weight coefficients of each time-domain feature, wherein the weight coefficients reflect the importance of each time-domain feature in characterizing ATP content.

[0062] In this embodiment, the pre-defined weighting coefficients for each time-domain feature were determined during the experimental and data analysis phases. Different time-domain features have different levels of importance in characterizing ATP content. For example, peak amplitude may be more direct and sensitive to ATP content, and its weighting coefficient may be relatively high; while signal duration may be relatively less important, and its weighting coefficient may be lower. Obtaining these weighting coefficients is the basis for correction processing. Subsequently, correction factors for each time-domain feature will be calculated based on these weighting coefficients, thereby making a more accurate adjustment to the initial matched content value.

[0063] Step 1432: Calculate the deviation ratio between each time-domain feature of the signal feature parameters and the time-domain feature of the corresponding standard signal feature parameters.

[0064] In this embodiment, to correct the initial matching content value, it is necessary to calculate the deviation ratio between each time-domain feature of the signal feature parameters and the corresponding time-domain feature parameters of the standard signal. This calculation is performed for four time-domain features: peak amplitude, signal rise rate, signal fall rate, and signal duration. For example, for peak amplitude, the difference between the actual detected peak amplitude and the peak amplitude in the standard signal feature parameters is calculated, and then this difference is divided by the standard peak amplitude to obtain the deviation ratio. Similarly, the deviation ratios for other time-domain features are calculated; these deviation ratios reflect the degree of difference between the actual detection results and the standard situation in each time-domain feature.

[0065] Step 1433: Multiply the deviation ratio of each time-domain feature by the corresponding weighting coefficient to obtain the correction factor of each time-domain feature. The correction factor reflects the degree of influence of the deviation of the time-domain feature on the ATP content matching result.

[0066] In this embodiment, after obtaining the deviation ratio and weighting coefficient of each time-domain feature, the two are multiplied to obtain the correction factor for each time-domain feature. The weighting coefficient reflects the importance of each time-domain feature in characterizing ATP content, while the deviation ratio reflects the degree of difference between the actual detection results and the standard conditions in each time-domain feature. The correction factor obtained by multiplying the two can comprehensively consider these two factors and accurately reflect the influence of the deviation of the time-domain feature on the ATP content matching result. For example, if the weighting coefficient of a certain time-domain feature is high and its deviation ratio is large, then the correction factor of that time-domain feature will be large, indicating that the deviation of that time-domain feature has a greater impact on the ATP content matching result.

[0067] Step 1434: Sum the correction factors of all time-domain features to obtain the total correction coefficient, and weight the preliminary matching content value based on the total correction coefficient to obtain the weighted adjusted target ATP component content value.

[0068] In this embodiment, after obtaining the correction factors for each time-domain feature, the correction factors for all time-domain features are summed to obtain the total correction coefficient. The total correction coefficient integrates the influence of all time-domain feature deviations on the ATP content matching result. Based on the total correction coefficient, the initial matched content value is weighted and adjusted. Specifically, the initial matched content value is multiplied by the total correction coefficient to obtain the weighted adjusted target ATP component content value. Through this correction process, the deviation of each time-domain feature can be taken into account, and the initial matched content value can be adjusted more accurately, so that the target ATP component content value can better reflect the actual ATP component content in the solution to be tested.

[0069] Step 144: Based on the ATP content value, search in the preset ATP content and bacterial quantity correspondence data to determine the bacterial quantity range corresponding to the ATP content value.

[0070] Step 1441: Determine the target ATP content range into which the ATP component content value falls, and find the corresponding bacterial quantity range in the associated data table for the target ATP content range.

[0071] In this embodiment, the preset correlation data between ATP content and bacterial count exists in the form of a relational data table, which contains multiple ATP content intervals, each corresponding to a set range of bacterial counts. Based on the calculated target ATP content value, the target ATP content interval into which it falls is searched in the relational data table. By comparing the target ATP content value with the upper and lower limits of each interval, the interval to which it belongs is determined. Then, the bacterial count range corresponding to the target ATP content interval in the relational data table is searched. This bacterial count range is predetermined through experiments and reflects the bacterial count in the refrigerator corresponding to the ATP content interval. Through this search method, the bacterial count range corresponding to the target ATP content value can be preliminarily determined.

[0072] Step 1442: If the ATP content value is equal to the boundary value of the target ATP content range, then merge the bacterial count ranges corresponding to the two adjacent ranges to which the boundary value belongs to form a combined bacterial count range containing the two ranges.

[0073] In this embodiment, when the target ATP content value is exactly equal to the boundary value of the target ATP content range, the bacterial count ranges corresponding to the two adjacent intervals to which the boundary value belongs are merged to more accurately determine the bacterial count range. Because the boundary value is located at the intersection of two intervals and may be related to the conditions of both intervals simultaneously, merging the bacterial count ranges corresponding to the two intervals yields a more comprehensive combined bacterial count range. This avoids inaccurate estimation of the bacterial count range due to the limitations of a single interval, making the detection results more reliable.

[0074] Step 1443: Verify the range of the combined bacterial count. If the range of the combined bacterial count exceeds the preset verification interval, it is marked as an abnormal result; otherwise, the range of the combined bacterial count is considered a valid result.

[0075] In this embodiment, to ensure the accuracy and reliability of the detection results, the range of combined bacterial counts needs to be validated. A validation interval is preset, which is a reasonable range of bacterial counts determined based on experiments and practical experience. The combined bacterial count range is compared with the validation interval. If the combined bacterial count range exceeds the preset validation interval, it indicates that the detection result may be abnormal, possibly due to errors in the detection process, such as improper sampling or signal processing errors. In this case, the result is marked as abnormal, and further inspection of the detection process and data is required. If the combined bacterial count range is within the preset validation interval, the result is considered valid and can be used as the final bacterial count range for subsequent detection result output.

[0076] Step 145: Integrate the bacterial count range and the corresponding ATP content values ​​to form the test results, and control the display device of the refrigerator to display the test results in a combination of text and graphics. The graphic display includes a trend chart reflecting the relative level of bacterial count.

[0077] Step 1451: Convert the bacterial count range into textual description information, which includes the upper and lower limits of the count range and a description of the sanitary conditions corresponding to the range.

[0078] In this embodiment, to allow users to understand the test results more intuitively, the determined bacterial count range is converted into textual descriptions. These descriptions not only include the upper and lower limits of the bacterial count range but also provide an explanation of the hygiene conditions corresponding to that range. For example, if the bacterial count range is in a lower range, the description might read, "Bacterial count between X and Y, hygiene conditions are good, the refrigerator can be used with confidence"; if the bacterial count range is in a higher range, the description might read, "Bacterial count between Z and W, hygiene conditions are poor, timely cleaning of the refrigerator is recommended." Through these textual descriptions, users can quickly understand the general bacterial situation and hygiene conditions inside the refrigerator, providing a reference for taking appropriate measures.

[0079] Step 1452: Organize the ATP content value and bacterial count range according to a preset data format to form structured detection result data. The detection result data includes detection time information, sensor number information, ATP content value and bacterial count range.

[0080] In this embodiment, to facilitate data management and storage, the target ATP content and bacterial count range are organized according to a preset data format to form structured detection result data. The preset data format is determined based on actual needs and data analysis requirements. The detection result data includes detection time information, sensor number information, target ATP content value, and bacterial count range. The detection time information records the specific time the detection occurred, the sensor number information identifies the fiber optic ATP sensor used for the detection, and the target ATP content and bacterial count range are the core detection data. Organizing this information into structured data facilitates subsequent data querying, statistics, and analysis, and also facilitates data interaction with other systems.

[0081] Step 1453: Invoke the display control module to display the text description information as scrolling text in the text display area of ​​the display device, and simultaneously generate a bar chart reflecting the relative level of the number of bacteria. The height ratio of the bar chart corresponds to the upper limit of the range of the number of bacteria.

[0082] In this embodiment, to display the test results in a visual format, the display control module is invoked. The previously converted text description information is displayed as scrolling text in the text display area of ​​the display device. Scrolling text allows for the display of more information within a limited display area, making it easier for users to view detailed test information. Simultaneously, a bar chart reflecting the relative level of bacterial counts is generated, with the height of the bars corresponding to the upper limit of the bacterial count range. For example, if the upper limit of the bacterial count range is large, the bar chart will be taller, allowing users to intuitively compare the relative levels of bacterial counts in different test results. By combining text and graphics, the bacterial test results inside the refrigerator can be presented to users more comprehensively and intuitively.

[0083] Step 1454: Use different colors to indicate the relative level of bacterial count in the bar chart, use color changes to show whether the bacterial count is within the normal range, and label the corresponding ATP content value next to the chart.

[0084] In this embodiment, to provide users with a more intuitive understanding of the relative levels of bacteria and the overall hygiene, different colors are used to indicate the relative levels of bacteria in the generated bar chart. A normal range is pre-defined; when the bacterial count is within the normal range, the bar chart may display green, indicating good hygiene; when the bacterial count exceeds the normal range, the bar chart displays red, indicating poor hygiene and requiring immediate attention. Through color changes, users can quickly determine whether the bacterial count in the refrigerator is within the normal range. Simultaneously, the corresponding target ATP content value is labeled next to the bar chart, allowing users to correlate bacterial count with ATP content and gain a deeper understanding of the relationship between bacterial conditions and ATP levels in the refrigerator. This approach further enhances the intuitiveness and readability of the test results presentation.

[0085] In summary, please refer to the following: Figure 1 and Figure 2 By applying this embodiment, the problems of long detection time, complicated operation, inconvenient equipment and difficulty in real-time on-site monitoring in existing refrigerator bacterial detection technologies can be improved, so as to achieve rapid, accurate and convenient detection of bacteria in refrigerators.

[0086] See Figure 3 As shown in the figure, this is a schematic diagram of the basic structure of a refrigerator bacterial detection system 200 based on a fiber optic ATP sensor provided in an embodiment of the present invention. The refrigerator bacterial detection system 200 based on a fiber optic ATP sensor includes: Processor 201; Storage device 202, on which computer program 2020 is stored; When the computer program 2020 is executed by the processor 201, the processor 201 implements any of the described methods for detecting bacteria in a refrigerator based on a fiber optic ATP sensor.

[0087] Based on the above, a readable storage medium is provided, on which a program or instructions are stored, and when the program or instructions are executed by a processor, the steps of the above method are implemented.

[0088] It should be noted that the various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the systems or apparatus disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the descriptions are relatively simple, and relevant parts can be referred to the method section.

Claims

1. A method for detecting bacteria in a refrigerator based on a fiber optic ATP sensor, characterized in that, The method includes: The sample collected from the area to be tested inside the refrigerator is mixed with the reaction reagent to form the test solution. The test solution is injected into the microfluidic reaction cell of the fiber optic ATP sensor, so that the ATP component in the test solution reacts with the enzyme reagent fixed on the sensor surface to produce a chemiluminescent reaction. The generated light signal is focused by the optical lens integrated by the sensor, and the focused light signal is coupled into the incident end face of the transmission optical fiber. The coupled optical signal is transmitted to the photoelectric detection unit through the transmission optical fiber. The photoelectric detection unit converts the received optical signal into a corresponding electrical signal. The waveform characteristics of the electrical signal correspond to the intensity change trend of the optical signal. The electrical signal contains dynamic change information reflecting the chemiluminescence reaction process. The electrical signal is subjected to multi-stage signal processing, including signal amplification, noise suppression, and feature extraction, to obtain signal characteristic parameters characterizing the concentration of ATP components. The variation of the signal characteristic parameters is correlated with the ATP content in the solution to be detected. The signal feature parameters are matched and analyzed with preset ATP content correlation data to determine the ATP component content value corresponding to the signal feature parameters. Based on the ATP component content value, preset ATP content and bacterial quantity correspondence data are queried to generate a detection result containing bacterial quantity information. The detection result is output to the refrigerator's matching display device in a visual form.

2. The method as described in claim 1, characterized in that, The coupled optical signal is transmitted to the photoelectric detection unit via the transmission optical fiber. The photoelectric detection unit converts the received optical signal into a corresponding electrical signal. The waveform characteristics of the electrical signal correspond to the intensity change trend of the optical signal. The electrical signal contains dynamic change information reflecting the chemiluminescence reaction process, including: Align the output end face of the transmission optical fiber with the photosensitive element of the photoelectric detection unit so that the optical signal output from the transmission optical fiber is incident perpendicularly onto the photosensitive area of ​​the photosensitive element. The photosensitive element converts the photon energy of the optical signal into charge carriers, and accumulates the charge carriers through internal circuitry to form an initial electrical signal. The initial electrical signal is pre-amplified by using a low-noise operational amplifier to increase its amplitude to the effective input range of the subsequent processing circuit, thus obtaining the amplified primary electrical signal. The primary electrical signal is processed by a bandpass filter circuit to select the frequency, retaining the characteristic frequency components related to the chemiluminescence reaction and filtering out the interference frequency components caused by ambient light and circuit noise, thus obtaining the filtered secondary electrical signal. The secondary electrical signal is subjected to analog-to-digital conversion to convert the analog electrical signal into a digital electrical signal sequence. The sampling interval of the electrical signal sequence matches the dynamic change period of the chemiluminescence reaction. The electrical signal sequence is an electrical signal containing information reflecting the dynamic change of the chemiluminescence reaction process.

3. The method as described in claim 1, characterized in that, The electrical signal undergoes multi-stage signal processing, sequentially performing signal amplification, noise suppression, and feature extraction to obtain signal characteristic parameters characterizing the concentration of ATP components. The variation of these signal characteristic parameters is correlated with the ATP content in the solution to be detected, including: The electrical signal is input into a programmable gain amplifier for dynamic gain adjustment. The amplification factor is automatically adjusted according to the initial amplitude of the electrical signal so that the amplitude of the amplified electrical signal is within a preset optimal processing range, thus obtaining the gain-adjusted electrical signal. The gain-adjusted electrical signal is subjected to wavelet threshold denoising. The electrical signal is decomposed into wavelet coefficients of different frequency bands by wavelet decomposition. The high-frequency wavelet coefficients are truncated by thresholding to retain the effective coefficients that reflect the signal characteristics. The denoised electrical signal is then reconstructed. The denoised electrical signal is subjected to time-domain feature extraction processing to extract the peak amplitude, rising edge rate, falling edge rate, and duration of the signal. These time-domain feature parameters are combined to form signal feature parameters characterizing the concentration of ATP components.

4. The method as described in claim 3, characterized in that, The step of inputting the electrical signal into a programmable gain amplifier for dynamic gain adjustment, automatically adjusting the amplification factor according to the initial amplitude of the electrical signal, so that the amplitude of the amplified electrical signal is within a preset optimal processing range, to obtain a gain-adjusted electrical signal, includes: The input electrical signal is monitored in real time. The maximum amplitude of the electrical signal within a preset time window is captured by the peak detection circuit, and the maximum amplitude is used as the initial amplitude. The initial amplitude is compared with the upper and lower thresholds of the preset optimal processing range. If the initial amplitude is lower than the lower threshold, the gain of the programmable gain amplifier is increased. If the initial amplitude is higher than the upper threshold, the gain of the programmable gain amplifier is decreased. If the initial amplitude is within the threshold range, the current gain is maintained. The electrical signal is amplified according to the adjusted gain factor to obtain a gain-adjusted electrical signal. The amplitude variation range of the gain-adjusted electrical signal is limited to the optimal processing range.

5. The method as described in claim 3, characterized in that, The denoised electrical signal undergoes time-domain feature extraction processing to extract the peak amplitude, rise rate, fall rate, and duration of the signal. These time-domain feature parameters are combined to form signal feature parameters characterizing the concentration of ATP components, including: The denoised electrical signal is subjected to waveform analysis to identify the maximum amplitude point in the electrical signal waveform, and the voltage value corresponding to the maximum amplitude point is recorded as the peak amplitude. In the electrical signal waveform, the rising phase from the starting point to the peak amplitude point is determined, the time interval and amplitude change of the rising phase are calculated, and the ratio of amplitude change to time interval is taken as the signal rising edge change rate. In the electrical signal waveform, determine the falling phase from the peak amplitude point to the signal end point, calculate the time interval and amplitude change of the falling phase, and use the ratio of amplitude change to time interval as the falling edge rate of the signal. Identify the start time when the amplitude of the electrical signal waveform exceeds a preset start threshold and the end time when the amplitude is lower than a preset end threshold, and calculate the time difference between the start time and the end time as the signal duration. The peak amplitude, the rate of change of the signal rising edge, the rate of change of the signal falling edge, and the duration of the signal are arranged in a preset order to form signal characteristic parameters.

6. The method as described in claim 1, characterized in that, The process of matching and analyzing the signal feature parameters with preset ATP content correlation data to determine the ATP component content value corresponding to the signal feature parameters, querying preset ATP content and bacterial count correlation data based on the ATP component content value, generating a detection result containing bacterial count information, and outputting the detection result in a visual form to the refrigerator's display device includes: The system retrieves preset ATP content correlation data from the storage unit. The ATP content correlation data includes multiple sets of standard signal feature parameters corresponding to known ATP contents, and each set of standard signal feature parameters corresponds to a set ATP content. Calculate the similarity between the signal feature parameters and each set of standard signal feature parameters, and select the ATP content corresponding to the standard signal feature parameter with the highest similarity as the preliminary matching content value; The preliminary matching content value is corrected by combining the weighting coefficients of each time-domain feature in the signal feature parameters and weighting the preliminary matching content value to obtain the target ATP component content value. Based on the ATP content value, a search is conducted in a preset ATP content and bacterial count correspondence database to determine the range of bacterial counts corresponding to the ATP content value; The bacterial count range and the corresponding ATP content values ​​are integrated to form the test results. The display device of the refrigerator is controlled to display the test results in a combination of text and graphics. The graphic display includes a trend chart reflecting the relative level of bacterial count.

7. The method as described in claim 6, characterized in that, The calculation of the similarity between the signal feature parameters and each set of standard signal feature parameters, and the selection of the ATP content corresponding to the standard signal feature parameter with the highest similarity as the preliminary matching content value, includes: The signal feature parameters are compared with each set of standard signal feature parameters, and the differences in peak amplitude, rising edge rate, falling edge rate, and duration are calculated. The difference values ​​for each dimension are normalized and converted into standardized difference values ​​with the same units. The magnitude of the standardized difference value reflects the degree of similarity in that dimension. The standardized difference values ​​of each dimension are weighted and summed according to a preset weight ratio to obtain a comprehensive difference value. The smaller the comprehensive difference value, the higher the similarity of the feature parameters of the two sets of signals. Iterate through all the comprehensive difference values ​​corresponding to the standard signal feature parameters, select the set of standard signal feature parameters with the smallest comprehensive difference value, and determine the ATP content associated with this set of standard signal feature parameters as the preliminary matching content value.

8. The method as described in claim 6, characterized in that, The step of correcting the preliminary matched content value, by incorporating the weighting coefficients of each time-domain feature in the signal feature parameters, and weighting the preliminary matched content value to obtain the target ATP component content value, includes: Obtain the preset weight coefficients for each time-domain feature, wherein the weight coefficients reflect the importance of each time-domain feature in characterizing ATP content; Calculate the deviation ratio between each time-domain feature of the signal feature parameters and the time-domain feature of the corresponding standard signal feature parameters; The correction factor for each time-domain feature is obtained by multiplying the deviation ratio of each time-domain feature by the corresponding weighting coefficient. The correction factor reflects the degree of influence of the deviation of the time-domain feature on the ATP content matching result. The correction factors for all time-domain features are summed to obtain the total correction coefficient. The preliminary matched content value is then weighted based on the total correction coefficient to obtain the weighted adjusted target ATP component content value.

9. The method as described in claim 6, characterized in that, The preset ATP content and bacterial count correspondence data is an experimentally established correlation data table. Each ATP content range in this table corresponds to a set bacterial count range. The step of searching within the preset ATP content and bacterial count correspondence data based on the ATP content value to determine the bacterial count range corresponding to that ATP content value includes: Determine the target ATP content range into which the ATP component content value falls, and find the corresponding bacterial count range in the associated data table for the target ATP content range; If the ATP content value is equal to the boundary value of the target ATP content range, then the bacterial count ranges corresponding to the two adjacent ranges to which the boundary value belongs are merged to form a combined bacterial count range that includes the two ranges. The range of the combined bacterial count is verified. If the range of the combined bacterial count exceeds the preset verification interval, it is marked as an abnormal result; otherwise, the range of the combined bacterial count is considered a valid result. The test results are generated by integrating the bacterial count range and the corresponding ATP content values. The refrigerator's display device is then used to display the test results using a combination of text and graphics. The graphical display includes trend charts reflecting the relative level of bacterial counts, including: The bacterial count range is converted into textual description information, which includes the upper and lower limits of the count range and a description of the sanitary conditions corresponding to that range. The ATP content values ​​and bacterial count ranges are organized according to a preset data format to form structured detection result data, which includes detection time information, sensor number information, ATP content values, and bacterial count ranges. The display control module is invoked to display the text description information as scrolling text in the text display area of ​​the display device, and a bar chart reflecting the relative level of the number of bacteria is generated simultaneously. The height ratio of the bar chart corresponds to the upper limit of the range of the number of bacteria. Different colors are used in the bar chart to indicate the relative level of bacterial count. Color changes indicate whether the bacterial count is within the normal range, and the corresponding ATP content value is marked next to the chart.

10. A refrigerator bacterial detection system based on an optical fiber ATP sensor, characterized in that, include: processor; A storage device having a computer program stored thereon, which, when executed by the processor, causes the processor to implement the refrigerator bacteria detection method based on a fiber optic ATP sensor as described in any one of claims 1-9.

Images