A medicament concentration detection system and method

Abstract

The application discloses a kind of method medicament concentration detection system and method, comprising: medicament concentration sensor, converter and server;Wherein, medicament concentration sensor, for receiving the concentration detection request of the medicament to be detected;Based on violet light emission module emits violet light and excites fluorescent agent in the medicament to be detected, and measure the fluorescent signal in the fluorescent agent;Converter is connected with medicament concentration sensor, for receiving the fluorescent signal sent by the medicament concentration sensor, and transmission to server;Server, for receiving the fluorescent signal transmitted by converter, and the fluorescent signal is handled, and the medicament concentration detection result is determined;Greatly improve medicament concentration detection precision, realize fast, convenient on-site detection, and provide effective data support for production and research in time.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of concentration detection, and in particular to a medicament concentration detection system and method. BACKGROUND

[0002] In the field of modern medicine research, chemical production and biological science, accurately and efficiently detecting the concentration of medicaments is a key link to guarantee product quality, optimize process flow and promote scientific research. At present, the common medicament concentration detection systems on the market have many shortcomings. On the one hand, some detection systems rely on traditional detection principles, such as using chemical color reaction or simple optical absorption measurement. The detection process is easily disturbed by environmental factors, the detection precision is low, and it is difficult to meet the high-precision detection demand. The error is particularly obvious when detecting trace amount of medicament concentration. On the other hand, the existing detection systems mostly use large and complex instruments and equipment. Not only the purchase cost is high, but also the equipment is bulky and the operation is complicated. It needs professional personnel to operate and maintain, and cannot realize rapid and convenient on-site detection, which is difficult to adapt to diversified application scenarios such as field scientific research and emergency detection. In addition, the data transmission and processing mode between the components of the traditional detection system is not efficient. The signal is easy to attenuate or distort in the transmission process, which leads to the delay of the detection result and cannot provide effective data support for production and research in time. Therefore, it is an urgent problem to be solved to develop a medicament concentration detection system with high detection precision, strong portability and efficient data transmission and processing.

[0003] Therefore, there is an urgent need for a medicament concentration detection system and method to solve the above technical problems. SUMMARY

[0004] The present application aims to at least one of the above technical problems. To this end, the first aspect of the present application aims to provide a medicament concentration detection system, which greatly improves the detection precision of medicament concentration, realizes rapid and convenient on-site detection, and provides effective data support for production and research in time.

[0005] The second aspect of the present application aims to provide a medicament concentration detection method.

[0006] To achieve the above-mentioned purpose, the first aspect of the present application provides a medicament concentration detection system, comprising: a medicament concentration sensor, a converter and a server; wherein,

[0007] The medicament concentration sensor is used for receiving the concentration detection request of the medicament to be detected, exciting the fluorescent agent in the medicament to be detected by the violet light emitted by the violet light emitting module, and measuring the fluorescent signal in the fluorescent agent.

[0008] The converter is connected with the medicament concentration sensor, used for receiving the fluorescent signal sent by the medicament concentration sensor and transmitting to the server.

[0009] a server configured to receive the fluorescent signal transmitted by the converter and process the fluorescent signal to determine the drug concentration detection result.

[0010] Preferably, the drug concentration sensor comprises:

[0011] a first receiving module configured to receive a concentration detection request for a drug to be detected;

[0012] a violet light emitting module configured to analyze the concentration detection request and emit violet light to excite a fluorescent agent in the drug to be detected;

[0013] a measuring module configured to measure a fluorescent signal of the fluorescent agent in the drug to be detected;

[0014] a sending module configured to send the fluorescent signal measured by the measuring module to the converter through the needle connector.

[0015] Preferably, the converter comprises:

[0016] a second receiving module connected to the needle connector of the drug concentration sensor and configured to receive the fluorescent signal emitted by the drug concentration sensor;

[0017] a transmitting module connected to the server and configured to send the fluorescent signal received by the second receiving module to the server.

[0018] Preferably, the server comprises:

[0019] a third receiving module configured to receive the fluorescent signal transmitted by the converter;

[0020] a noise reduction module configured to reduce noise of the fluorescent signal;

[0021] a detection module configured to input the fluorescent signal after noise reduction into a pre-trained drug concentration detection model for detection to determine the drug concentration detection result.

[0022] Preferably, the noise reduction module comprises:

[0023] a first obtaining sub-module configured to obtain spectral data corresponding to the fluorescent signal;

[0024] a constructing sub-module configured to establish a Cartesian rectangular coordinate system based on the spectral data;

[0025] an abnormality identifying sub-module configured to identify abnormality of wave peaks included in the spectral data in the Cartesian rectangular coordinate system to determine a plurality of abnormal wave peaks;

[0026] a second obtaining sub-module configured to obtain the fluorescent signal corresponding to the plurality of abnormal wave peaks to obtain a plurality of abnormal fluorescent signals.

[0027] a noise reduction submodule, configured to:

[0028] any abnormal fluorescence signal;

[0029] performing short-time Fourier transform on the abnormal fluorescence signal to obtain a target frequency spectrum corresponding to the abnormal fluorescence signal;

[0030] performing feature extraction on the target frequency spectrum to obtain a feature spectrum line;

[0031] obtaining energy values of each node on the feature spectrum line, and determining an energy maximum value in each node on the feature spectrum line;

[0032] querying a preset energy value-noise reduction coefficient table according to the energy maximum value to determine a target noise reduction coefficient;

[0033] performing noise reduction on the abnormal fluorescence signal based on the target noise reduction coefficient;

[0034] iterating through all abnormal fluorescence signals to complete noise reduction on the fluorescence signals.

[0035] Preferably, the abnormality identification submodule comprises:

[0036] a calculation unit, configured to:

[0037] selecting any wave crest in the Cartesian rectangular coordinate system as a target wave crest;

[0038] calculating an evaluation value corresponding to the target wave crest;

[0039] iterating through all wave crests in the Cartesian rectangular coordinate system to obtain evaluation values of a plurality of wave crests;

[0040] an abnormality identification unit, configured to filter wave crests based on the evaluation values of the plurality of wave crests to determine a plurality of abnormal wave crests.

[0041] Preferably, the calculation unit comprises:

[0042] a first obtaining subunit, configured to:

[0043] selecting any wave crest in the Cartesian rectangular coordinate system as a target wave crest;

[0044] obtaining a vertex of the target wave crest and minimum values on left and right sides of the target wave crest;

[0045] a second obtaining subunit, configured to obtain a perpendicular line of the vertex of the target wave crest and a horizontal axis of the Cartesian rectangular coordinate system as a target perpendicular line;

[0046] a first calculation subunit, configured to:

[0047] An area of a closed region composed of the vertex of the target wave crest, the target vertical line, the left minimum point of the target wave crest and the horizontal axis is calculated as a first area;

[0048] An area of a closed region composed of the vertex of the target wave crest, the target vertical line, the right minimum point of the target wave crest and the horizontal axis is calculated as a second area;

[0049] A difference value between the first area and the second area is calculated as a target area difference value;

[0050] The third obtaining subunit is configured to:

[0051] The coordinate of the vertex of the target wave crest is obtained, and a height value of the target wave crest is determined based on the coordinate of the vertex;

[0052] A distance between the minimum points on the left and right sides of the target wave crest is obtained as a first distance;

[0053] The second calculating subunit is configured to:

[0054] The product of the height value of the target wave crest and the first distance is taken as a feature value of the target wave crest;

[0055] The product of the feature value of the target wave crest and the target area difference value is taken as an evaluation value of the target wave crest.

[0056] Preferably, the anomaly identifying unit comprises:

[0057] The difference calculating subunit is configured to calculate a difference value between evaluation values of every two adjacent wave crests, and determine an evaluation difference value;

[0058] The anomaly identifying subunit is configured to compare the evaluation difference value with a preset evaluation difference value threshold, and take two wave crests with an evaluation value difference value greater than or equal to the preset evaluation difference value threshold as abnormal wave crests, to obtain a plurality of abnormal wave crests.

[0059] Preferably, the method for obtaining a medicament concentration detection model comprises:

[0060] A medicament concentration detection sample training data set is obtained;

[0061] A neural network model is trained based on the medicament concentration detection sample training data set, to obtain an initial medicament concentration detection model;

[0062] A medicament concentration detection sample test data set is obtained;

[0063] The initial medicament concentration detection model is tested based on the medicament concentration detection sample test data set, and when the testing is qualified, a trained medicament concentration detection model is obtained.

[0064] To achieve the above object, the second aspect of the present application provides a medicament concentration detection method, comprising:

[0065] The medicament concentration sensor receives the concentration detection request of the to-be-detected medicament, excites the fluorescent agent in the to-be-detected medicament based on the violet light emission module, and measures the fluorescent signal in the fluorescent agent;

[0066] The converter receives the fluorescent signal sent by the medicament concentration sensor and transmits to the server;

[0067] The server receives the fluorescent signal transmitted by the converter and processes the fluorescent signal to determine the medicament concentration detection result.

[0068] The present application provides a medicament concentration detection system and method, the medicament concentration sensor can automatically receive the concentration detection request, and independently excite the fluorescent agent and measure the fluorescent signal, which reduces the manual operation, improves the automation degree of detection, reduces the manual error, and improves the detection efficiency; the converter is responsible for transmitting the fluorescent signal measured by the sensor to the server, realizing reliable transmission of data, ensuring that the detection data can accurately and correctly reach the server for processing, providing a basis for subsequent medicament concentration calculation; the server can quickly receive the fluorescent signal and process it, and determine the medicament concentration detection result through analysis of the fluorescent signal. This centralized data processing method can utilize the powerful computing power of the server to quickly obtain accurate detection results, providing a basis for decision-making; based on the principle of exciting the fluorescent agent by violet light and measuring the fluorescent signal, the characteristics of fluorescent substances are utilized to detect the medicament concentration. This method has high sensitivity and selectivity, and can detect low-concentration medicaments. It has important significance for some scenes with accurate concentration requirements, such as drug production and environmental monitoring.

[0069] Other features and advantages of the present application will be described in the following description, and some will become apparent from the description, or will be understood by those skilled in the art. The purpose and other advantages of the present application can be achieved and obtained by the structure specifically pointed out in the written description and the accompanying drawings.

[0070] The technical solutions of the present application will be further described in detail below with the help of the drawings and examples. BRIEF DESCRIPTION OF DRAWINGS

[0071] The accompanying drawings are used to provide a further understanding of the present application, and constitute a part of the specification, together with the embodiments of the present application, to explain the present application, and do not constitute a limitation on the present application. In the drawings:

[0072] Figure 1 is a block diagram of a medicament concentration detection system according to an embodiment of the present application;

[0073] Figure 2 is a block diagram of a medicament concentration sensor according to an embodiment of the present application;

[0074] Figure 3 is a flow chart of a medicament concentration detection method according to an embodiment of the present application;

[0075] Figure 4 is a schematic diagram of a needle joint according to an embodiment of the present application;

[0076] Figure 5 is a schematic diagram of a connection mode according to an embodiment of the present application;

[0077] Figure 6 is a schematic diagram of a medicament concentration sensor according to an embodiment of the present application. DETAILED DESCRIPTION

[0078] The preferred embodiments of the present application will be described hereinafter with reference to the accompanying drawings, in which, it should be understood that the preferred embodiments described herein are intended for explaining and illustrating the present application, and are not intended for limiting the present application.

[0079] Embodiment 1

[0080] As shown in Figure 1 a medicament concentration detection system, comprising: a medicament concentration sensor, a converter and a server; wherein,

[0081] The medicament concentration sensor is configured to receive a request for detecting the concentration of a medicament to be detected, excite a fluorescent agent in the medicament to be detected by emitting ultraviolet light from an ultraviolet light emitting module, and measure a fluorescent signal in the fluorescent agent.

[0082] The converter is connected to the medicament concentration sensor and configured to receive the fluorescent signal sent by the medicament concentration sensor and transmit the fluorescent signal to the server.

[0083] The server is configured to receive the fluorescent signal transmitted by the converter, process the fluorescent signal, and determine a result of detecting the concentration of the medicament.

[0084] In this embodiment, the specific implementation of measuring the fluorescence signal in the fluorescent agent is to measure by 90-degree scattered light method; a specific wavelength of violet light is emitted by the violet light emitting module to irradiate the fluorescent agent in the drug to be detected; after the fluorescent agent absorbs the violet light energy, it transitions from the ground state to the excited state, and then emits fluorescence when it returns to the ground state; the detector is arranged at a position 90 degrees from the propagation direction of the excitation light. Such an arrangement is to reduce the interference of the excitation light directly scattered onto the detector, because in the 90-degree direction, the scattering light intensity of the excitation light is relatively weak, and the fluorescence signal is relatively easy to be detected; the detector collects the fluorescence emitted from the fluorescent agent. Since fluorescence is emitted in all directions, there will be a certain intensity of fluorescence propagating in the 90-degree direction, and the detector collects and measures the fluorescence in this direction.

[0085] In this embodiment, the fluorescence signal measured by the 90-degree scattered light method mainly includes the intensity, wavelength distribution, and fluorescence lifetime of the fluorescence. The fluorescence intensity can reflect the content or concentration of the fluorescent agent, and generally the fluorescence intensity is proportional to the amount of the fluorescent agent; the wavelength distribution can help determine the type of the fluorescent agent, different fluorescent agents have specific emission wavelength ranges; the fluorescence lifetime refers to the average time experienced by the fluorescent molecule from the excited state to the ground state, and it is also an important characteristic of the fluorescent agent, which can be used to distinguish different types of fluorescent agents or to study the environmental changes of the fluorescent agent.

[0086] In this embodiment, the overall waterproof of the drug concentration sensor adopts IP68 waterproof standard.

[0087] In this embodiment, the converter can be but is not limited to a USB-485 converter.

[0088] In this embodiment, the technical parameters of the drug concentration sensor are shown in Table 1:

[0089] Table 1

[0090]

[0091]

[0092] The working principle of the above technical solution is that: the medicament concentration sensor first receives a concentration detection request for the to-be-detected medicament; the violet light emitting module in the medicament concentration sensor emits violet light, which irradiates the fluorescent agent in the to-be-detected medicament. The fluorescent agent molecules absorb the energy of the violet light and transition from the ground state to the excited state. Because the excited state is unstable, the fluorescent agent molecules will quickly return to the ground state while releasing fluorescence in the form of light, realizing the process from violet light excitation to fluorescence generation; the medicament concentration sensor measures the fluorescence signal generated by the fluorescent agent. The fluorescence signal is acquired through a specific measurement method (such as 90-degree scattering light method, etc.); the converter is connected with the medicament concentration sensor and receives the fluorescence signal sent by the sensor; the server processes the fluorescence signal transmitted by the converter after receiving it; because the characteristics of the fluorescence signal have a certain relationship with the concentration of the fluorescent agent, through the analysis of the signal, the server can determine the detection result of the medicament concentration, providing data support for subsequent decision-making.

[0093] The beneficial effects of the above technical solution are: the medicament concentration sensor can automatically receive the concentration detection request and independently excite the fluorescent agent and measure the fluorescence signal, reducing manual operation, improving the automation degree of detection, reducing human error, and improving detection efficiency; the converter is responsible for transmitting the fluorescence signal measured by the sensor to the server, realizing reliable transmission of data and ensuring that the detection data can accurately and correctly reach the server for processing, providing a basis for subsequent medicament concentration calculation; the server can quickly receive and process the fluorescence signal to determine the detection result of the medicament concentration. This centralized data processing method can utilize the powerful computing power of the server to quickly obtain accurate detection results, providing a basis for decision-making; based on the principle of exciting the fluorescent agent with violet light and measuring the fluorescence signal, the characteristics of fluorescent substances are utilized to detect the concentration of medicaments. This method has high sensitivity and selectivity and can detect low-concentration medicaments, which is of great significance for some scenes that require accurate concentration, such as drug production and environmental monitoring.

[0094] Embodiment 2

[0095] As shown in Figure 2 , the medicament concentration sensor comprises:

[0096] A first receiving module for receiving a concentration detection request for a to-be-detected medicament;

[0097] A violet light emitting module for analyzing the concentration detection request and emitting violet light to excite the fluorescent agent in the to-be-detected medicament;

[0098] A measurement module for measuring the fluorescence signal of the fluorescent agent in the to-be-detected medicament;

[0099] The sending module is configured to send the fluorescent signal measured by the measuring module to the converter through the needle connector.

[0100] In this embodiment, the needle connector includes five wires, red (power +), blue (power -), yellow (485-A), green (485-B), and black (4-20 mA positive).

[0101] The connection mode of the needle connector and the converter is as follows:

[0102] Red wire (power +): Connects the V+ or positive terminal of the USB-485 converter to provide the device operating voltage (usually 9-36V). Blue wire (power -): Connects the GND or negative terminal of the USB-485 converter to form a closed loop. RS-485 signal line connection: Yellow wire (485-A): Connects the A or DATA+ terminal of the USB-485 converter to transmit the positive pole of the differential signal. Green wire (485-B): Connects the B or DATA- terminal of the USB-485 converter to transmit the negative pole of the differential signal. Black wire (4-20 mA positive): If the USB-485 converter supports analog signal input, it needs to be connected to the corresponding terminal (such as AI+).

[0103] The beneficial effects of the above technical solution are: the first receiving module is responsible for receiving the concentration detection request of the to-be-detected medicament, can accurately capture the detection instruction, provides an explicit trigger signal for subsequent detection operation, ensures that the detection process is orderly conducted after receiving the accurate instruction, and avoids the situation of false or missed detection request; the violet light emitting module emits violet light after analyzing the concentration detection request, this targeted excitation method can accurately make the fluorescent agent in the to-be-detected medicament absorb energy and produce fluorescence; the measuring module focuses on measuring the fluorescent signal of the fluorescent agent in the to-be-detected medicament; and the sending module sends the fluorescent signal measured by the measuring module to the converter through the needle connector.

[0104] Embodiment 3

[0105] The converter comprises:

[0106] The second receiving module is connected with the needle connector included in the medicament concentration sensor, and is configured to receive the fluorescent signal emitted by the medicament concentration sensor.

[0107] The transmission module is connected with the server, and is configured to send the fluorescent signal received by the second receiving module to the server.

[0108] Embodiment 4

[0109] The server comprises:

[0110] The third receiving module is configured to receive the fluorescent signal transmitted by the converter.

[0111] a noise reduction module, configured to reduce noise of the fluorescence signal;

[0112] a detection module, configured to input the fluorescence signal after noise reduction into a pre-trained medicament concentration detection model for detection to determine a medicament concentration detection result.

[0113] The beneficial effects of the above technical solution are as follows: the third receiving module is specially responsible for receiving the fluorescence signal transmitted by the converter; the noise reduction module reduces noise of the fluorescence signal. Since the signal may be interfered by various factors during transmission, noise may be generated, which affects the quality of the signal and subsequent analysis results. Through noise reduction, the purity of the fluorescence signal can be improved, the influence of noise on the detection result can be reduced, the signal characteristics are more clear, and thus the accuracy and reliability of detection are improved; the detection module inputs the fluorescence signal after noise reduction into the pre-trained medicament concentration detection model for detection. The trained model can accurately infer the medicament concentration according to the characteristics of the fluorescence signal. The pre-trained model is learned and optimized by a large amount of data, has high accuracy and generalization ability, can quickly and accurately determine the medicament concentration detection result, improves the detection efficiency and precision, provides reliable concentration detection data for actual application, and meets the demand for accurate detection of medicament concentration.

[0114] Embodiment 5

[0115] The noise reduction module comprises:

[0116] a first acquisition sub-module, configured to acquire spectral data corresponding to the fluorescence signal;

[0117] a construction sub-module, configured to establish a Cartesian coordinate system based on the spectral data;

[0118] an abnormality identification sub-module, configured to identify abnormalities of wave peaks included in the spectral data in the Cartesian coordinate system to determine a plurality of abnormal wave peaks;

[0119] a second acquisition sub-module, configured to acquire fluorescence signals corresponding to the plurality of abnormal wave peaks to obtain a plurality of abnormal fluorescence signals;

[0120] a noise reduction sub-module, configured to:

[0121] select any abnormal fluorescence signal;

[0122] perform short-time Fourier transform on the abnormal fluorescence signal to obtain a target frequency spectrum corresponding to the abnormal fluorescence signal;

[0123] perform feature extraction on the target frequency spectrum to obtain a feature spectrum line;

[0124] acquire energy values of each node on the feature spectrum line to determine a maximum energy value in each node on the feature spectrum line.

[0125] According to the energy maximum value, a preset energy value-noise reduction coefficient table is queried to determine a target noise reduction coefficient;

[0126] Based on the target noise reduction coefficient, the abnormal fluorescence signal is denoised;

[0127] All abnormal fluorescence signals are traversed to complete the denoising of the fluorescence signal.

[0128] The working principle and beneficial effects of the above technical solution are: the first acquisition submodule acquires spectral data corresponding to the fluorescence signal. The spectral data can reflect the intensity distribution of the fluorescence signal at different wavelengths, and the wave peaks correspond to the strong response of the fluorescence signal at a specific wavelength. These wave peaks contain key feature information of the fluorescence signal and are the basis for subsequent analysis and processing; the construction submodule establishes a Cartesian coordinate system based on the spectral data. By mapping the spectral data into the Cartesian coordinate system, the distribution of the wave peaks can be more intuitively displayed, facilitating the analysis and processing of the wave peaks, and providing a visual and quantifiable space for subsequent anomaly identification; the anomaly identification submodule identifies anomalies in the wave peaks in the Cartesian coordinate system. By preset anomaly identification rules or algorithms, the characteristics of the wave peaks such as position, intensity, and width are analyzed and compared with the characteristics of normal wave peaks to determine which wave peaks are abnormal. These abnormal wave peaks may be caused by noise, interference, or other non-drug concentration related factors and need to be further processed; the second acquisition submodule acquires the fluorescence signal corresponding to the abnormal wave peak to obtain a plurality of abnormal fluorescence signals. Each abnormal wave peak corresponds to a specific fluorescence signal, and these abnormal fluorescence signals are the objects that need to be denoised. By extracting them separately, the noise can be processed more targetedly without affecting the normal fluorescence signal part; the denoising submodule processes each abnormal fluorescence signal in turn. First, a short-time Fourier transform is performed on an arbitrary abnormal fluorescence signal. The short-time Fourier transform can localize the analysis of the signal in the time and frequency domains to obtain a target frequency spectrum corresponding to the abnormal fluorescence signal, which shows the energy distribution of the signal at different frequencies over time. Then, feature extraction is performed on the target frequency spectrum to obtain a feature spectrum line. By analyzing the energy distribution characteristics in the frequency spectrum, the key feature information of the abnormal fluorescence signal is extracted to form a feature spectrum line, which contains information related to noise and the signal itself. Next, the energy values of each node on the feature spectrum line are obtained, and the maximum energy value is determined. The maximum energy value reflects the strongest energy response of the abnormal fluorescence signal at a certain specific frequency or time period, which is closely related to the characteristics of the noise. Then, the maximum energy value is used to query a preset energy value-denoising coefficient table. The table is established based on a large amount of experimental or empirical data and records the denoising coefficients corresponding to different energy values. By querying the table, the target denoising coefficient for the current abnormal fluorescence signal can be determined. Finally, the abnormal fluorescence signal is denoised based on the target denoising coefficient. According to the determined denoising coefficient, the abnormal fluorescence signal is processed accordingly, such as filtering, attenuation, and other methods to reduce the influence of noise and make the signal purer; all abnormal fluorescence signals are traversed, and each abnormal fluorescence signal is denoised according to the above steps to finally complete the denoising of the entire fluorescence signal.

[0129] Example 6

[0130] Anomaly identification sub-module, comprising:

[0131] A computing unit for:

[0132] Taking any one of the wave peaks in the Cartesian rectangular coordinate system as a target wave peak;

[0133] Calculating the evaluation value corresponding to the target wave peak;

[0134] Traversing all wave peaks in the Cartesian rectangular coordinate system to obtain evaluation values of a plurality of wave peaks;

[0135] An anomaly identification unit for screening the wave peaks based on the evaluation values of the plurality of wave peaks to determine a plurality of abnormal wave peaks.

[0136] The beneficial effects of the above technical solution are: the computing unit quantifies the characteristics of the wave peaks by calculating the evaluation value for each wave peak. The characteristics of the wave peaks such as height, width, shape, etc. are comprehensively reflected in the evaluation value, which makes the analysis of the wave peaks more objective and accurate; in the process of calculating the evaluation value, multiple attributes of the wave peaks can be considered comprehensively; different attributes may have different contributions to the judgment of whether the wave peak is abnormal, and by integrating these factors into the evaluation value through a reasonable calculation method, the abnormality possibility of the wave peak can be more comprehensively evaluated, avoiding the limitations of single factor judgment, thereby improving the identification accuracy of abnormal signals such as noise; the anomaly identification unit screens the wave peaks based on the evaluation values, which can accurately find out the wave peaks representing abnormal signals. In the subsequent noise reduction process, only the fluorescent signals corresponding to these abnormal wave peaks can be processed.

[0137] Embodiment 7

[0138] A computing unit, comprising:

[0139] A first obtaining sub-unit for:

[0140] Taking any one of the wave peaks in the Cartesian rectangular coordinate system as a target wave peak;

[0141] Obtaining the vertex of the target wave peak and the minimum values on the left and right sides of the target wave peak;

[0142] A second obtaining sub-unit for obtaining the perpendicular line of the vertex of the target wave peak and the horizontal axis of the Cartesian rectangular coordinate system as a target perpendicular line;

[0143] A first calculating sub-unit for:

[0144] Calculating the area of the closed region composed of the vertex of the target wave peak, the target perpendicular line, and the minimum value point on the left side of the target wave peak and the horizontal axis as a first area;

[0145] An area of a closed region composed of the vertex of the target wave peak, the target vertical line, the right minimum point of the target wave peak and the horizontal axis is calculated as a second area;

[0146] A difference value between the first area and the second area is calculated as a target area difference value;

[0147] The third obtaining sub-unit is configured to:

[0148] A coordinate of the vertex of the target wave peak is obtained, and a height value of the target wave peak is determined based on the coordinate of the vertex;

[0149] A distance between the minimum points on the left and right sides of the target wave peak is obtained as a first distance;

[0150] The second calculating sub-unit is configured to:

[0151] A product of the height value of the target wave peak and the first distance is taken as a characteristic value of the target wave peak;

[0152] A product of the characteristic value of the target wave peak and the target area difference value is taken as an evaluation value of the target wave peak.

[0153] The working principle and beneficial effects of the technical solution are as follows: the first acquisition subunit randomly selects a wave crest as a target wave crest in the Cartesian rectangular coordinate system; the vertex of the target wave crest and the minimum values on the left and right sides of the target wave crest are further acquired. The wave crest vertex is the highest point of the wave crest, and the minimum values on the left and right sides define the range of the wave crest, and these points contain the key information of the shape of the wave crest; the second acquisition subunit acquires the perpendicular line of the horizontal axis of the Cartesian rectangular coordinate system, that is, the target perpendicular line, of the vertex of the target wave crest. This perpendicular line is used for subsequent area calculation, and it divides the left and right side regions of the wave crest; the first calculation subunit calculates the areas of two closed regions respectively. The first closed region is surrounded by the vertex of the target wave crest, the target perpendicular line, the minimum value point on the left side of the target wave crest, and the horizontal axis, and the area is recorded as the first area; the second closed region is surrounded by the vertex of the target wave crest, the target perpendicular line, the minimum value point on the right side of the target wave crest, and the horizontal axis, and the area is recorded as the second area. The two region areas reflect the shape and size difference of the left and right sides of the wave crest; the difference between the first area and the second area is calculated to obtain the target area difference. This difference reflects the degree of asymmetry of the left and right sides of the wave crest, and the greater the degree of asymmetry, the greater the difference; the third acquisition subunit acquires the coordinates of the vertex of the target wave crest, and determines the height value of the target wave crest according to the coordinates. The wave crest height reflects the intensity of the wave crest; the distance between the minimum value points on the left and right sides of the target wave crest, that is, the first distance, is acquired. This distance reflects the width of the wave crest; the second calculation subunit multiplies the height value of the target wave crest by the first distance to obtain the characteristic value of the target wave crest. The characteristic value integrates the height and width information of the wave crest; the characteristic value of the target wave crest is multiplied by the target area difference to obtain the evaluation value of the target wave crest. The evaluation value integrates the height, width, and degree of asymmetry of the left and right sides of the wave crest, and can be used for subsequent judgment of whether the wave crest is abnormal.

[0154] Embodiment 8

[0155] The abnormality recognition unit comprises:

[0156] The difference calculation subunit is configured to calculate the difference between the evaluation values of every two adjacent wave crests, and determine an evaluation difference.

[0157] The abnormality recognition subunit is configured to compare the evaluation difference with a preset evaluation difference threshold, and regard two wave crests with an evaluation value difference greater than or equal to the preset evaluation difference threshold as abnormal wave crests, to obtain a plurality of abnormal wave crests.

[0158] In this embodiment, the preset evaluation difference threshold is pre-set based on industry experience.

[0159] The beneficial effects of the above technical solutions are: the difference calculation sub-unit highlights the feature difference between adjacent wave crests by calculating the difference between the evaluation values of the adjacent wave crests. Under normal circumstances, the evaluation values of adjacent wave crests are usually close to each other because they may represent similar signal characteristics. When there is an abnormal wave crest, its evaluation value will deviate greatly from that of the adjacent wave crest. This calculation method presents this deviation in the form of a specific evaluation difference, which helps to more clearly identify abnormal wave crests; the abnormality identification sub-unit compares the evaluation difference with a preset evaluation difference threshold. The preset threshold is set according to a large amount of experimental data or practical application experience, and it provides a clear standard for judging whether a wave crest is abnormal. When the evaluation difference is greater than or equal to the threshold, it means that there is a significant difference between the two adjacent wave crests, and one or both of the wave crests is likely to be an abnormal wave crest. This threshold-based judgment method is simple and direct, and can effectively filter out abnormal wave crests from numerous wave crests; abnormal wave crests are often caused by noise, interference or other non-drug concentration related factors. By identifying and processing these abnormal wave crests in a timely manner, the interference of noise on the detection results of the drug concentration can be effectively reduced, and the accuracy and reliability of the detection can be improved.

[0160] Embodiment 9

[0161] The medicament concentration detection model comprises:

[0162] obtaining a medicament concentration detection sample training data set;

[0163] training a neural network model based on the medicament concentration detection sample training data set to obtain an initial medicament concentration detection model;

[0164] obtaining a medicament concentration detection sample test data set;

[0165] testing the initial medicament concentration detection model based on the medicament concentration detection sample test data set, and obtaining a trained medicament concentration detection model when the test is qualified.

[0166] The beneficial effects of the above technical solutions are: by obtaining a medicament concentration detection sample training data set, the neural network model can be trained based on actual data. This training method based on real data enables the model to learn the intrinsic relationship between the medicament concentration and the related fluorescence signal, thereby improving the accuracy and reliability of the model in detecting actual medicament concentration; using a neural network model for medicament concentration detection, the neural network has strong non-linear fitting capability and can capture the complex non-linear relationship between the medicament concentration and the fluorescence signal. Compared with traditional detection methods, the neural network model can handle more complex signal characteristics and improve the precision and sensitivity of the detection.

[0167] As Figure 3As shown, to achieve the above object, the second aspect embodiment of the present application proposes a medicament concentration detection method, comprising steps S1-S3:

[0168] S1: the medicament concentration sensor receives the concentration detection request of the to-be-detected medicament, excites the fluorescent agent in the to-be-detected medicament based on the violet light emission module, and measures the fluorescent signal in the fluorescent agent;

[0169] S2: the converter receives the fluorescent signal sent by the medicament concentration sensor and transmits to the server;

[0170] S3: the server receives the fluorescent signal transmitted by the converter, processes the fluorescent signal, and determines the medicament concentration detection result.

[0171] The beneficial effects of the above technical solution are: the medicament concentration sensor can automatically receive the concentration detection request, independently excite the fluorescent agent and measure the fluorescent signal, reduce manual operation, improve the automation degree of detection, reduce manual error, and improve the detection efficiency; the converter is responsible for transmitting the fluorescent signal measured by the sensor to the server, realizing reliable transmission of data, ensuring that the detection data can accurately and correctly reach the server for processing, providing a basis for subsequent medicament concentration calculation; the server can quickly receive and process the fluorescent signal, and determine the medicament concentration detection result through analysis of the fluorescent signal. This centralized data processing method can utilize the powerful computing power of the server to quickly obtain accurate detection results, providing a basis for decision-making; based on the principle of exciting the fluorescent agent by violet light and measuring the fluorescent signal, the characteristics of fluorescent substances are utilized to detect the medicament concentration. This method has high sensitivity and selectivity, and can detect low-concentration medicaments. It is of great significance for some scenes that require accurate concentration, such as drug production and environmental monitoring.

[0172] As shown in the figure, Figure 4 The pin joint includes 5 wires, red (power +), blue (power -), yellow (485-A), green (485-B), and black (4-20 mA positive).

[0173] As shown in the figure, Figure 5 When connecting the computer, the terminals of the USB-485 converter are connected with the red (power +), blue (power -), yellow (485-A), and green (485-B) of the medicament concentration sensor, respectively, and are connected to the computer. Open the application software and operate.

[0174] When the mobile phone is connected, the MICRO USB or TYPE C adapter is connected with the USB-485, and then the terminals of the USB-485 converter are connected with the red (power +), blue (power -), yellow (485-A) and green (485-B) of the drug concentration sensor, and the mobile phone is connected, the APP is opened, and the operation is performed.

[0175] Obviously, various modifications and changes can be made to the present application without departing from the spirit and scope thereof. Accordingly, it is intended that all such modifications and changes be included within the scope of the application as defined by the following claims and their equivalents.

Claims

1. A drug concentration detection system, characterized in that, include: Drug concentration sensor, converter, and server; among which, A drug concentration sensor is used to receive a concentration detection request for the drug to be tested; it uses a violet light emission module to emit violet light to excite the fluorescent agent in the drug to be tested and measures the fluorescence signal in the fluorescent agent. A converter, connected to a drug concentration sensor, is used to receive the fluorescence signal sent by the drug concentration sensor and transmit it to a server; The server is used to receive the fluorescence signal transmitted by the converter, process the fluorescence signal, and determine the drug concentration detection result. The server includes: The third receiving module is used to receive the fluorescence signal transmitted by the converter; A noise reduction module is used to reduce the noise of the fluorescence signal; The detection module is used to input the noise-reduced fluorescence signal into a pre-trained drug concentration detection model for detection and to determine the drug concentration detection result. The noise reduction module includes: The first acquisition submodule is used to acquire the spectral data corresponding to the fluorescence signal; Construct a submodule for establishing a Cartesian coordinate system based on the spectral data; The anomaly identification submodule is used to identify anomalies in the peaks included in the spectral data in the Cartesian coordinate system and determine several abnormal peaks. The second acquisition submodule is used to acquire the fluorescence signals corresponding to several abnormal peaks, and obtain several abnormal fluorescence signals; The noise reduction submodule is used for: Select any abnormal fluorescence signal; Perform a short-time Fourier transform on the abnormal fluorescence signal to obtain the target spectrum corresponding to the abnormal fluorescence signal; Feature extraction is performed on the target spectrogram to obtain feature spectral lines; Obtain the energy value of each node on the characteristic spectral line, and determine the maximum energy value among each node on the characteristic spectral line; Based on the maximum energy value, query the preset energy value-noise reduction coefficient table to determine the target noise reduction coefficient; The abnormal fluorescence signal is denoised based on the target denoising coefficient; The process involves iterating through all abnormal fluorescence signals and performing noise reduction on those signals.

2. The drug concentration detection system as described in claim 1, characterized in that, The drug concentration sensor includes: The first receiving module is used to receive a concentration detection request for the drug to be tested; The ultraviolet light emission module is used to analyze the concentration detection request and emit ultraviolet light to excite the fluorescent agent in the drug to be detected; The measurement module is used to measure the fluorescence signal of the fluorescent agent in the drug to be tested; The transmitting module is used to send the fluorescence signal measured by the measurement module to the converter through a pin connector.

3. The drug concentration detection system as described in claim 1, characterized in that, Converter, including: The second receiving module is connected to the needle connector included in the drug concentration sensor and is used to receive the fluorescence signal emitted by the drug concentration sensor. The transmission module, connected to the server, is used to send the fluorescence signal received by the second receiving module to the server.

4. The drug concentration detection system as described in claim 1, characterized in that, The anomaly detection submodule includes: Computational unit, used for: In the Cartesian coordinate system, any wave peak is selected as the target wave peak; Calculate the evaluation value corresponding to the target peak; By traversing all the peaks in the Cartesian coordinate system, we obtain the evaluation values ​​of several peaks. The anomaly identification unit is used to filter peaks based on the evaluation values ​​of several peaks and identify several abnormal peaks.

5. The drug concentration detection system as described in claim 4, characterized in that, The computing unit includes: The first acquisition subunit is used for: In the Cartesian coordinate system, any wave peak is selected as the target wave peak; Obtain the vertex of the target peak and the minimum values ​​on the left and right sides of the target peak; The second acquisition subunit is used to acquire the perpendicular line between the vertex of the target wave crest and the horizontal axis of the Cartesian coordinate system, as the target perpendicular line; The first computational subunit is used for: Calculate the area of ​​the closed region formed by the vertex of the target peak, the target vertical line, the left minimum point of the target peak, and the horizontal axis, and use it as the area of ​​the first region; Calculate the area of ​​the closed region formed by the vertex of the target peak, the target vertical line, the right minimum point of the target peak, and the horizontal axis, and use it as the area of ​​the second region; Calculate the difference between the area of ​​the first region and the area of ​​the second region to obtain the target area difference; The third acquisition subunit is used for: Obtain the coordinates of the vertex of the target wave crest, and determine the height of the target wave crest based on the vertex coordinates; Obtain the distance between the minimum points on the left and right sides of the target wave crest to get the first distance; The second computational subunit is used for: The product of the height of the target peak and the first distance is used as the characteristic value of the target peak. The product of the characteristic value of the target peak and the difference in the target area is used as the evaluation value of the target peak.

6. The drug concentration detection system as described in claim 4, characterized in that, Anomaly detection unit, including: The difference calculation subunit is used to calculate the difference in evaluation values ​​between every two adjacent peaks and determine the evaluation difference. An anomaly identification subunit is used to compare the evaluation difference with a preset evaluation difference threshold, and to identify two peaks whose evaluation difference is greater than or equal to the preset evaluation difference threshold as abnormal peaks, thereby obtaining a number of abnormal peaks.

7. The drug concentration detection system as described in claim 1, characterized in that, Methods for obtaining drug concentration detection models include: Obtain the training dataset for drug concentration detection samples; The neural network model was trained based on the sample training dataset for drug concentration detection to obtain the initial drug concentration detection model; Obtain the test dataset for drug concentration detection samples; The initial drug concentration detection model was tested using a sample test dataset. When the test was passed, a well-trained drug concentration detection model was obtained.

8. The method for detecting drug concentration using the drug concentration detection system according to any one of claims 1-7, characterized in that, include: After receiving a concentration detection request for the drug to be tested, the drug concentration sensor emits ultraviolet light based on the ultraviolet light emission module to excite the fluorescent agent in the drug to be tested, and measures the fluorescence signal in the fluorescent agent; The converter receives the fluorescence signal sent from the drug concentration sensor and transmits it to the server; The server receives the fluorescence signal transmitted by the converter, processes the fluorescence signal, and determines the drug concentration detection result.

Images