A test paper type recognition method, system and storage medium based on current characteristics

Abstract

This invention provides a method, system, and storage medium for identifying test strip types based on current characteristics. First, raw current signals are acquired through test strip electrodes and preprocessed to generate a current response curve. Second, a stability index for the initial time window of the curve is calculated. When the stability index is lower than a preset stability threshold, dynamic response features are extracted. Then, a first type of result is determined based on the deviation between the response time and a preset reference threshold. A second type of result is obtained by comparing the corresponding peak current with a rise slope threshold. Finally, a comprehensive confidence level is generated by calculating the deviation between the dynamic features and the threshold, which determines the final output type, triggers a retest, or reports an error. This invention overcomes the limitations of single-feature discrimination through multi-level collaboration of signal stability screening, chain threshold discrimination, and confidence level assessment, improving the accuracy and reliability of HCG and LH test strip identification in complex application environments.

Description

Technical Field

[0001] This invention relates to the field of medical test strip type identification, and more specifically, to a test strip type identification method, system, and storage medium based on current characteristics. Background Technology

[0002] In the field of medical testing equipment, especially in portable test strip devices for home use, the automatic and accurate identification of test strip types, particularly for HCG and LH test strips, two important physiological indicators, has always been a technical challenge.

[0003] First, traditional manual identification methods rely entirely on users visually observing the markings on the test strips or manually selecting the model. This severely limits the automation level of testing equipment and is prone to misjudgment due to user errors or worn markings on the test strips. Second, while current optical identification solutions have achieved automation to some extent, the image acquisition and processing modules they introduce significantly increase the system's complexity and manufacturing costs. Furthermore, this method has extremely stringent requirements regarding the positioning accuracy of the test strips, ambient lighting conditions, and the cleanliness of the marking area, making it difficult to guarantee stability in real-world home environments. Third, while electrical characteristic-based identification methods avoid some of the drawbacks of optical solutions, existing solutions are mostly limited to measuring a single static parameter of the test strip, such as basic impedance or DC current. Since the electrochemical characteristics of HCG and LH test strips are quite similar and easily affected by multiple factors such as test strip production batches, ambient temperature and humidity, and electrode contact conditions, the distinguishability of a single parameter is limited, and the identification accuracy is often unsatisfactory in complex usage scenarios. Finally, even in feature-based electrical solutions, most still employ a simple single-threshold judgment strategy. This rigid discrimination mode struggles to handle boundary samples with indistinct features and cannot quantify the reliability of the identification results, leading to low system detection accuracy and requiring improvement in both intelligence and fault tolerance. Therefore, there is an urgent need for a low-cost, high-precision test strip type identification technology based on current characteristics for portable devices. Summary of the Invention

[0004] In view of the above problems, the purpose of this invention is to provide a test strip type identification method, system and storage medium based on current characteristics. By constructing a complete closed-loop process of signal quality assessment, feature extraction, chain threshold comparison and confidence decision, it achieves the core effect of ensuring signal validity from the source and adopting a multi-level intelligent discrimination strategy, thereby fundamentally improving the accuracy and reliability of identification. Specifically, firstly, by standardizing signal excitation and refining filtering and smoothing preprocessing, the signal quality and consistency are improved from the source. Secondly, by calculating the statistical variance of the initial time window and comparing it with a stability threshold, a rapid and objective diagnosis of signal quality is achieved, effectively blocking invalid detections caused by poor contact. Thirdly, by extracting three key features—response time, peak current, and rising slope—and combining them into a feature vector, a multi-dimensional and in-depth characterization of the electrochemical response characteristics of the test strip is achieved. Simultaneously, a preliminary comparison of the response time with a preset threshold is used to pre-classify the test strip type, and subsequent comparison thresholds are dynamically selected accordingly, achieving a combination of rapid screening and accurate matching. Furthermore, by requiring that both the peak current and rising slope fall within the deviation range centered on the corresponding threshold to confirm the final type result, a dual-insurance joint judgment effect is achieved, reducing the risk of misjudgment caused by mismatch of a single feature.

[0005] The first aspect of this invention provides a method for identifying test strip types based on current characteristics, the method comprising:

[0006] The raw current signal is acquired by the test strip electrodes, and the current response curve is obtained based on the electrical signal preprocessing.

[0007] Based on a preset time window, stability indicators are calculated according to the current response curve.

[0008] Determine whether the stability index is lower than a preset stability threshold;

[0009] If not, output an invalid detection signal;

[0010] If so, then extract the dynamic response features from the current response curve;

[0011] The first type of result is obtained by comparing the response time of the dynamic response feature with a preset reference time threshold.

[0012] Based on the first type of result, the peak current threshold and the rise slope threshold are obtained;

[0013] By comparing the peak current and rise slope of the dynamic response characteristics with the peak current threshold and rise slope threshold, a second type of result is obtained;

[0014] The overall confidence level is calculated based on the deviation between the dynamic response characteristics and the corresponding threshold.

[0015] Determine whether the overall confidence level is higher than a preset first confidence threshold;

[0016] If so, output the result of type 2;

[0017] If not, a retest will be triggered or an invalid detection signal will be output.

[0018] In this scheme, the process of acquiring the raw current signal through test paper electrodes and obtaining the current response curve based on electrical signal preprocessing specifically includes:

[0019] In response to the test strip insertion signal, a preset excitation voltage signal is applied to the test strip electrodes;

[0020] The raw microcurrent signal flowing through the test strip electrodes is acquired synchronously and converted into a digital current sequence;

[0021] The digital current sequence is filtered.

[0022] The filtered digital current sequence is smoothed to generate a current response curve.

[0023] In this scheme, the calculation of stability indicators based on the current response curve within a preset time window specifically includes:

[0024] Based on a preset stabilization time window, the window current data points are divided at the beginning stage of the current response curve.

[0025] Calculate the statistical variance of the current data points within the window as a stability indicator;

[0026] The statistical variance is compared with a preset stability threshold;

[0027] If the statistical variance is lower than the stability threshold, the stability of the current signal is determined to meet the requirements, and the dynamic feature extraction step is initiated.

[0028] In this solution, extracting dynamic response features from the current response curve specifically includes:

[0029] Identify the current rise phase in the current response curve and calculate the current rise slope during the current rise phase through linear fitting.

[0030] Traverse the entire current response curve and extract the global maximum value as the peak current;

[0031] The response time is calculated as the duration required from the start of the current response until the current value reaches a specified percentage of its peak value.

[0032] The current rise slope, peak current, and response time are combined into a dynamic response feature vector.

[0033] In this scheme, the step of comparing the response time of the dynamic response characteristics with a preset reference time threshold to obtain the first type of result specifically includes:

[0034] The response time is compared with the preset LH test strip response time threshold and HCG test strip response time threshold to obtain the LH response time deviation and HCG response time deviation.

[0035] Determine whether the LH response time deviation is lower than the HCG response time deviation;

[0036] If so, then the first type result is set to type LH;

[0037] If not, the first type result is set to HCG type;

[0038] Based on the preset type threshold mapping table, the corresponding peak current threshold and rise slope threshold are obtained according to the first type result.

[0039] In this scheme, the peak current and rise slope of the dynamic response characteristics are compared with the peak current threshold and rise slope threshold to obtain a second type of result, which specifically includes:

[0040] Based on the results of the first type, the peak current deviation range and the rise slope deviation range are obtained;

[0041] Determine whether the peak current is within the peak current deviation range with the peak current threshold as the midpoint, and whether the rising slope is within the rising slope deviation range with the rising slope as the midpoint.

[0042] If both conditions are met, then the second type of result is set according to the first type of result;

[0043] If any condition is not met, the type identification is deemed to have failed.

[0044] A second aspect of the present invention provides a test strip type identification system based on current characteristics, including a test strip type identification method program based on current characteristics, wherein the test strip type identification method program based on current characteristics, when executed by the processor, performs the following steps:

[0045] The raw current signal is acquired by the test strip electrodes, and the current response curve is obtained based on the electrical signal preprocessing.

[0046] Based on a preset time window, stability indicators are calculated according to the current response curve.

[0047] Determine whether the stability index is lower than a preset stability threshold;

[0048] If not, output an invalid detection signal;

[0049] If so, then extract the dynamic response features from the current response curve;

[0050] The first type of result is obtained by comparing the response time of the dynamic response feature with a preset reference time threshold.

[0051] Based on the first type of result, the peak current threshold and the rise slope threshold are obtained;

[0052] By comparing the peak current and rise slope of the dynamic response characteristics with the peak current threshold and rise slope threshold, a second type of result is obtained;

[0053] The overall confidence level is calculated based on the deviation between the dynamic response characteristics and the corresponding threshold.

[0054] Determine whether the overall confidence level is higher than a preset first confidence threshold;

[0055] If so, output the result of type 2;

[0056] If not, a retest will be triggered or an invalid detection signal will be output.

[0057] A third aspect of the present invention provides a computer-readable storage medium including a test strip type identification method program based on current characteristics, wherein when the test strip type identification method program based on current characteristics is executed by a processor, it implements the steps of the test strip type identification method based on current characteristics as described in any of the preceding claims.

[0058] This invention provides a method, system, and storage medium for identifying test strip types based on current characteristics. First, raw current signals are acquired through test strip electrodes and preprocessed to generate a current response curve. Second, a stability index for the initial time window of the curve is calculated. When the stability index is lower than a preset stability threshold, dynamic response features are extracted. Then, a first type of result is determined based on the deviation between the response time and a preset reference threshold. A second type of result is obtained by comparing the corresponding peak current with a rise slope threshold. Finally, a comprehensive confidence level is generated by calculating the deviation between the dynamic features and the threshold, which determines the final output type, triggers a retest, or reports an error. This invention overcomes the limitations of single-feature discrimination through multi-level collaboration of signal stability screening, chain threshold discrimination, and confidence level assessment, improving the accuracy and reliability of HCG and LH test strip identification in complex application environments. Attached Figure Description

[0059] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly described below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as a limitation on the scope.

[0060] Figure 1 A flowchart of a test paper type identification method based on current characteristics according to the present invention is shown;

[0061] Figure 2 A flowchart illustrating the generation of a current response curve according to an embodiment of the present invention is shown;

[0062] Figure 3 A flowchart illustrating a current response stability determination method provided by an embodiment of the present invention is shown.

[0063] Figure 4 A block diagram of a test strip type identification system based on current characteristics according to the present invention is shown. Detailed Implementation

[0064] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0065] Unless otherwise defined, all terms (including technical and scientific terms) used in embodiments of this invention shall have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. It should also be understood that terms such as those defined in a common dictionary shall be interpreted as having a meaning consistent with their meaning in the context of the relevant art, and not as being interpreted in an idealized or highly formalized sense, unless expressly defined in this embodiment of the invention.

[0066] The terms "first," "second," and similar words used in the embodiments of this invention do not indicate any order, quantity, or importance, but are merely used to distinguish different components. Terms such as "an," "a," or "the" do not indicate a quantity limitation, but rather indicate the presence of at least one. Similarly, terms such as "including" or "comprising" mean that the element or object preceding the word encompasses the elements or objects listed after the word and their equivalents, without excluding other elements or objects. Terms such as "connected" or "linked" are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. The steps preceding or following the steps in the method of the embodiments of this invention are not necessarily performed precisely in sequence. Instead, various steps can be processed in reverse order or simultaneously. Furthermore, other operations can be added to these processes, or one or more steps can be removed from these processes.

[0067] In addition, the functional modules in the various embodiments of the present invention can be integrated together to form an independent part, or each module can exist independently, or two or more modules can be integrated to form an independent part.

[0068] Figure 1 A flowchart of a test paper type identification method based on current characteristics according to the present invention is shown.

[0069] like Figure 1 As shown, the first aspect of this invention discloses a test strip type identification method based on current characteristics, the method comprising:

[0070] S102, the raw current signal is collected through the test paper electrode, and the current response curve is obtained based on the electrical signal preprocessing;

[0071] S104, based on a preset time window, calculates stability indicators according to the current response curve;

[0072] S106, determine whether the stability index is lower than the preset stability threshold;

[0073] S108, if not, output an invalid detection signal;

[0074] S110, if so, then extract the dynamic response features from the current response curve;

[0075] S112, Based on the comparison between the response time of the dynamic response feature and the preset reference time threshold, a first type of result is obtained;

[0076] S114, Based on the first type of result, obtain the peak current threshold and the rising slope threshold;

[0077] S116, compare the peak current and rise slope of the dynamic response characteristics with the peak current threshold and rise slope threshold to obtain the second type of result;

[0078] S118, Based on the deviation between the dynamic response characteristics and the corresponding threshold, the overall confidence level is calculated;

[0079] S120, determine whether the overall confidence level is higher than a preset first confidence threshold;

[0080] S122, if so, output the result of type 2;

[0081] S124, if not, trigger a retest or output an invalid detection signal.

[0082] It should be noted that in this embodiment, firstly, the raw current signal corresponding to the electrochemical reaction with the test strip is acquired through the test strip electrode, and the signal is preprocessed, including filtering and smoothing, to generate a stable current response curve. Then, data points are extracted from the initial stage of the current response curve based on a preset time window, and their statistical variance is calculated as a stability index. Whether to continue detection is determined by whether this stability index is lower than a preset stability threshold. If the stability index is not lower than the threshold, an invalid detection signal is output; if it is lower than the threshold, the dynamic response feature extraction stage begins. In the feature extraction stage, dynamic response features, including response time, peak current, and current rise slope, are extracted from the current response curve. Then, based on the comparison between the response time and the preset reference time thresholds for LH and HCG test strips, a first type of result is generated, and the first type of result is mapped to the corresponding peak current threshold and rise slope threshold. The peak current and rise slope in the dynamic response features are further compared with the mapped thresholds to generate a second type of result. Finally, the overall confidence level is calculated based on the deviation between the dynamic response feature values ​​and the corresponding thresholds. By determining whether the overall confidence level is higher than the first confidence threshold, the system decides whether to output a second type of result, trigger a retest, or output an invalid detection signal. This embodiment effectively eliminates misjudgments caused by signal instability by organically combining multi-level threshold judgment and confidence assessment, and achieves intelligent handling of uncertainties through the confidence mechanism, thereby improving the accuracy of test strip type identification and the robustness of the system.

[0083] Figure 2 A flowchart illustrating the generation of a current response curve provided by an embodiment of the present invention is shown.

[0084] According to embodiments of the present invention, such as Figure 2 As shown, the process of acquiring the raw current signal through the test strip electrodes and obtaining the current response curve based on the electrical signal preprocessing specifically includes:

[0085] S202, in response to the test strip insertion signal, applies a preset excitation voltage signal to the test strip electrodes;

[0086] S204 synchronously acquires the raw microcurrent signal flowing through the test strip electrode and converts it into a digital current sequence;

[0087] S206, The digital current sequence is filtered;

[0088] S208 smooths the filtered digital current sequence and generates a current response curve.

[0089] It should be noted that in this embodiment, when the test strip insertion operation is detected by a mechanical or electronic sensor, an excitation voltage signal with a precisely preset amplitude and waveform is automatically applied to the electrode in contact with the test strip to stimulate the test strip to produce a characteristic electrochemical response. Simultaneously, a high-precision current acquisition module is activated to continuously capture the weak raw current signal flowing through the test strip electrode and converts it into a discrete digital current sequence using an analog-to-digital converter. Subsequently, this digital current sequence undergoes digital filtering processing, primarily low-pass filtering, to effectively remove high-frequency noise and power frequency interference from the signal. The filtered signal is then further smoothed using algorithms based on moving averages or spline interpolation to correct glitches in the data and optimize the curve's consistency. Finally, a smooth and stable current response curve is generated. This embodiment, through standardized signal excitation and refined preprocessing, ensures the quality and consistency of the raw current signal, laying a reliable signal foundation for all subsequent analyses and judgments.

[0090] Figure 3 A flowchart illustrating the determination of current response stability provided by an embodiment of the present invention is shown.

[0091] According to embodiments of the present invention, such as Figure 3 As shown, the calculation of stability indicators based on the current response curve within a preset time window specifically includes:

[0092] S302, based on a preset stabilization time window, divides the window current data points in the initial stage of the current response curve;

[0093] S304 calculates the statistical variance of the current data points within the window as a stability indicator;

[0094] S306, compare the statistical variance with a preset stability threshold;

[0095] S308, if the statistical variance is lower than the stability threshold, the stability of the current signal is determined to meet the requirements, and the dynamic feature extraction step is initiated.

[0096] It should be noted that in this embodiment, after obtaining the current response curve, a stabilization time window of a certain length is first defined in the initial stage of the curve. The selection of this window aims to cover the critical stage from the initial transition of the signal to its tendency to stabilize. Subsequently, the statistical variance of all current data points within this window is calculated, and this value is used as the core indicator for quantifying signal stability. Next, the calculated statistical variance is compared with a stability threshold that has been verified by numerous experiments. If the statistical variance is lower than the stability threshold, the quality of the current signal is determined to meet the reliability requirements, and the subsequent dynamic response feature extraction process begins. If it does not meet the requirements, the process is immediately interrupted and an error is reported. This embodiment focuses on the stability of the signal in the initial stage and uses statistical variance as an objective indicator for quantitative evaluation, thereby achieving rapid and effective diagnosis of the quality of detection conditions and avoiding invalid feature extraction and type identification under unstable signal conditions.

[0097] According to an embodiment of the present invention, extracting dynamic response features from the current response curve specifically includes:

[0098] Identify the current rise phase in the current response curve and calculate the current rise slope during the current rise phase through linear fitting.

[0099] Traverse the entire current response curve and extract the global maximum value as the peak current;

[0100] The response time is calculated as the duration required from the start of the current response until the current value reaches a specified percentage of its peak value.

[0101] The current rise slope, peak current, and response time are combined into a dynamic response feature vector.

[0102] It should be noted that in this embodiment, morphological analysis is first performed on the preprocessed current response curve to accurately identify the current rise phase characterizing the electrochemical reaction rate, and the slope value of this phase, i.e., the current rise slope, is calculated using a linear fitting algorithm. Subsequently, the data points of the entire current response curve are traversed to find its global maximum value and determine it as the peak current, which reflects the maximum current intensity of the reaction. Simultaneously, the time interval from the start of the current response to the current value rising to a specific preset proportion (e.g., 80%) of the peak current is calculated and defined as the response time, which characterizes the speed at which the reaction reaches near saturation. Finally, the extracted current rise slope, peak current, and response time—the three key parameters—are combined to form a comprehensive dynamic response feature vector. This embodiment constructs a feature model characterizing the differences between test strip types by extracting multi-dimensional features closely related to the chemical reaction kinetics of the test strip.

[0103] According to an embodiment of the present invention, the step of comparing the response time of the dynamic response feature with a preset reference time threshold to obtain a first type of result specifically includes:

[0104] The response time is compared with the preset LH test strip response time threshold and HCG test strip response time threshold to obtain the LH response time deviation and HCG response time deviation.

[0105] Determine whether the LH response time deviation is lower than the HCG response time deviation;

[0106] If so, then the first type result is set to type LH;

[0107] If not, the first type result is set to HCG type;

[0108] Based on the preset type threshold mapping table, the corresponding peak current threshold and rise slope threshold are obtained according to the first type result.

[0109] It should be noted that in this embodiment, after obtaining the response time characteristic value characterizing the reaction kinetics, it is first compared with the preset standard response time thresholds for LH test strips and HCG test strips to obtain the LH response time deviation and HCG response time deviation, respectively. Then, a judgment logic is established by comparing the magnitudes of these two deviations. If the LH response time deviation is less than the HCG response time deviation, the test strip reaction characteristic is determined to be closer to the typical response mode of the LH test strip, and the first type result is initialized to the LH type. Otherwise, it is initialized to the HCG type. Based on this, the type-threshold mapping database pre-stored in the device is accessed, and the peak current threshold and rise slope threshold parameters matching the determined first type result are automatically indexed and called. This embodiment achieves rapid pre-classification of test strip types through preliminary screening of response time; it not only improves the retrieval efficiency of subsequent feature comparison but also provides a targeted comparison benchmark for subsequent accurate discrimination by establishing a dynamic mapping relationship between type and threshold, enhancing the adaptability to different test strips.

[0110] According to an embodiment of the present invention, by comparing the peak current and rise slope of the dynamic response characteristics with the peak current threshold and rise slope threshold, a second type of result is obtained, specifically including:

[0111] Based on the results of the first type, the peak current deviation range and the rise slope deviation range are obtained;

[0112] Determine whether the peak current is within the peak current deviation range with the peak current threshold as the midpoint, and whether the rising slope is within the rising slope deviation range with the rising slope as the midpoint.

[0113] If both conditions are met, then the second type of result is set according to the first type of result;

[0114] If any condition is not met, the type identification is deemed to have failed.

[0115] It should be noted that in this embodiment, after obtaining the first type of result and its corresponding threshold parameter, the allowable deviation range of peak current and rising slope is first determined based on the statistical characteristics of historical data for this type, forming a reasonable fluctuation range centered on the threshold. Subsequently, a dual condition judgment is performed: on the one hand, it is checked whether the actually measured peak current falls within the preset deviation range centered on the peak current threshold; on the other hand, it is simultaneously checked whether the current rising slope is within the allowable deviation range of the corresponding rising slope threshold. Only when both of these key feature parameters simultaneously meet their respective range requirements is the first type of result directly confirmed as the second type of result. If any feature value exceeds its corresponding reasonable fluctuation range, the identification process is immediately interrupted and the type identification failure status is returned. This embodiment, by establishing a dual feature verification mechanism, fully utilizes the directional guidance provided by the pre-classification results and avoids misjudgment caused by mismatch of a single feature through strict interval verification, significantly improving the reliability of the type discrimination result.

[0116] It is worth mentioning that the comprehensive confidence level calculated based on the deviation between the dynamic response characteristics and the corresponding threshold specifically includes:

[0117] Calculate the deviation between the peak current and the peak current threshold to obtain the first confidence component;

[0118] The deviation of the rising slope from the rising slope threshold is calculated to obtain the second confidence component;

[0119] The first confidence component and the second confidence component are weighted and fused to obtain the comprehensive confidence score;

[0120] If the overall confidence level is higher than the first confidence level threshold, then output the second type of result;

[0121] If the overall confidence level is between the first confidence threshold and the second confidence threshold, a retest mechanism is triggered;

[0122] If the overall confidence level is lower than the second confidence level threshold, the output type is an unknown signal.

[0123] It should be noted that in this embodiment, the normalized deviation between the measured peak current and the corresponding threshold is calculated as the first confidence component, and the normalized deviation between the measured rising slope and the corresponding threshold is calculated as the second confidence component. The two components are linearly weighted and fused using preset weighting coefficients to generate a comprehensive confidence score with a normalized value range. Based on the higher first confidence threshold and the lower second confidence threshold, three judgment intervals are formed: when the comprehensive confidence is higher than the first threshold, the second type of result is directly output as the final judgment; when the comprehensive confidence is between the two thresholds, it is determined to be in a critical state and a retesting mechanism is automatically initiated to obtain new detection data; when the comprehensive confidence is lower than the second threshold, the signal quality or feature matching is determined to be too low, and an unknown type signal is output. This embodiment, by establishing a quantitative confidence evaluation index and a hierarchical decision-making mechanism, intelligently distinguishes between confirmed results, boundary cases, and invalid detections, ensuring the efficiency of direct output of high-confidence results while providing a review opportunity for critical cases, thus improving the rationality of decision-making in practical applications.

[0124] It is worth mentioning that it also includes:

[0125] After the test strip is inserted but before the detection excitation is applied, the background current signal of the test strip electrode is collected;

[0126] The background current signal is matched with a pre-stored background feature library to identify the production batch of the test strip;

[0127] Based on the identified production batch, select the corresponding LH response time threshold, HCG response time threshold, peak current threshold, and rise slope threshold from the threshold group mapping table.

[0128] After completing the test strip type identification, the dynamic response characteristics and the final judgment result are fed back to the background feature database.

[0129] It should be noted that in this embodiment, before applying the detection excitation signal, the background current signal of the test strip electrodes in a resting state is first collected to obtain raw information reflecting the characteristics of the test strip substrate. This background current signal is then matched with a pre-stored background feature database to identify the specific production batch or material specification of the test strip through feature comparison. Based on the identified production batch identifier, the LH response time threshold, HCG response time threshold, peak current threshold, and rise slope threshold suitable for that batch of test strips are precisely selected from a mapping table storing multiple threshold parameter groups. After completing the entire process of test strip type identification, the dynamic response feature data obtained in this detection and the final judgment result are fed back to the background feature database as new samples for iterative optimization of the threshold selection strategy. This embodiment effectively eliminates systematic errors caused by differences in test strip production batches through pre-identification of the background and dynamic threshold matching; simultaneously, the continuous learning mechanism enables the system to have self-optimization capabilities, thereby significantly improving the adaptability of the detection system to test strips from different sources and its long-term accuracy.

[0130] It is worth mentioning that it also includes:

[0131] A composite excitation signal comprising multiple frequency components is applied to the test strip electrodes;

[0132] Current response signals under different excitation frequencies were collected to obtain multiple sets of current time-series curves;

[0133] Dynamic response feature extraction is performed on each set of current time-series curves to obtain multiple sets of feature vectors;

[0134] Multiple feature vectors are fused at the feature level to generate enhanced feature vectors;

[0135] The dynamic response features are modified based on the enhanced feature vector.

[0136] It should be noted that in this embodiment, firstly, a composite periodic electrical signal containing multiple components of different frequencies is applied to the test strip electrodes as a detection excitation. Subsequently, the current response signals generated under each frequency component excitation are simultaneously acquired and separated, obtaining multiple sets of current time-series curves that are parallel in the time domain. Secondly, a complete dynamic response feature extraction process is independently performed on the current time-series curves corresponding to each frequency, resulting in multiple feature vectors containing response time, peak current, and rise slope. Simultaneously, a feature-level fusion algorithm is used to integrate the multiple feature vectors from different frequency responses into a high-dimensional enhanced feature vector, which comprehensively reflects the response characteristics of the test strip under different electrical excitation conditions. Finally, this enhanced feature vector is used to correct and supplement the dynamic response features obtained under the original single-excitation condition. This embodiment obtains richer electrochemical response information through multi-frequency excitation, effectively overcoming the limitation of features being easily interfered with under single detection conditions, and improving anti-interference capability and recognition accuracy in complex usage environments.

[0137] Figure 4 A block diagram of a test strip type identification system based on current characteristics according to the present invention is shown.

[0138] like Figure 4 As shown, a second aspect of the present invention discloses a test strip type identification system 4 based on current characteristics, including a memory 41 and a processor 42. The memory includes a test strip type identification method program based on current characteristics. When the test strip type identification method program based on current characteristics is executed by the processor, it performs the following steps:

[0139] The raw current signal is acquired by the test strip electrodes, and the current response curve is obtained based on the electrical signal preprocessing.

[0140] Based on a preset time window, stability indicators are calculated according to the current response curve.

[0141] Determine whether the stability index is lower than a preset stability threshold;

[0142] If not, output an invalid detection signal;

[0143] If so, then extract the dynamic response features from the current response curve;

[0144] The first type of result is obtained by comparing the response time of the dynamic response feature with a preset reference time threshold.

[0145] Based on the first type of result, the peak current threshold and the rise slope threshold are obtained;

[0146] By comparing the peak current and rise slope of the dynamic response characteristics with the peak current threshold and rise slope threshold, a second type of result is obtained;

[0147] The overall confidence level is calculated based on the deviation between the dynamic response characteristics and the corresponding threshold.

[0148] Determine whether the overall confidence level is higher than a preset first confidence threshold;

[0149] If so, output the result of type 2;

[0150] If not, a retest will be triggered or an invalid detection signal will be output.

[0151] It should be noted that in this embodiment, firstly, the raw current signal corresponding to the electrochemical reaction with the test strip is acquired through the test strip electrode, and the signal is preprocessed, including filtering and smoothing, to generate a stable current response curve. Then, data points are extracted from the initial stage of the current response curve based on a preset time window, and their statistical variance is calculated as a stability index. Whether to continue detection is determined by whether this stability index is lower than a preset stability threshold. If the stability index is not lower than the threshold, an invalid detection signal is output; if it is lower than the threshold, the dynamic response feature extraction stage begins. In the feature extraction stage, dynamic response features, including response time, peak current, and current rise slope, are extracted from the current response curve. Then, based on the comparison between the response time and the preset reference time thresholds for LH and HCG test strips, a first type of result is generated, and the first type of result is mapped to the corresponding peak current threshold and rise slope threshold. The peak current and rise slope in the dynamic response features are further compared with the mapped thresholds to generate a second type of result. Finally, the overall confidence level is calculated based on the deviation between the dynamic response feature values ​​and the corresponding thresholds. By determining whether the overall confidence level is higher than the first confidence threshold, the system decides whether to output a second type of result, trigger a retest, or output an invalid detection signal. This embodiment effectively eliminates misjudgments caused by signal instability by organically combining multi-level threshold judgment and confidence assessment, and achieves intelligent handling of uncertainties through the confidence mechanism, thereby improving the accuracy of test strip type identification and the robustness of the system.

[0152] According to an embodiment of the present invention, the step of acquiring the original current signal through the test strip electrode and obtaining the current response curve based on the electrical signal preprocessing specifically includes:

[0153] In response to the test strip insertion signal, a preset excitation voltage signal is applied to the test strip electrodes;

[0154] The raw microcurrent signal flowing through the test strip electrodes is acquired synchronously and converted into a digital current sequence;

[0155] The digital current sequence is filtered.

[0156] The filtered digital current sequence is smoothed to generate a current response curve.

[0157] It should be noted that in this embodiment, when the test strip insertion operation is detected by a mechanical or electronic sensor, an excitation voltage signal with a precisely preset amplitude and waveform is automatically applied to the electrode in contact with the test strip to stimulate the test strip to produce a characteristic electrochemical response. Simultaneously, a high-precision current acquisition module is activated to continuously capture the weak raw current signal flowing through the test strip electrode and converts it into a discrete digital current sequence using an analog-to-digital converter. Subsequently, this digital current sequence undergoes digital filtering processing, primarily low-pass filtering, to effectively remove high-frequency noise and power frequency interference from the signal. The filtered signal is then further smoothed using algorithms based on moving averages or spline interpolation to correct glitches in the data and optimize the curve's consistency. Finally, a smooth and stable current response curve is generated. This embodiment, through standardized signal excitation and refined preprocessing, ensures the quality and consistency of the raw current signal, laying a reliable signal foundation for all subsequent analyses and judgments.

[0158] According to an embodiment of the present invention, the step of calculating the stability index based on the current response curve within a preset time window specifically includes:

[0159] Based on a preset stabilization time window, the window current data points are divided at the beginning stage of the current response curve.

[0160] Calculate the statistical variance of the current data points within the window as a stability indicator;

[0161] The statistical variance is compared with a preset stability threshold;

[0162] If the statistical variance is lower than the stability threshold, the stability of the current signal is determined to meet the requirements, and the dynamic feature extraction step is initiated.

[0163] It should be noted that in this embodiment, after obtaining the current response curve, a stabilization time window of a certain length is first defined in the initial stage of the curve. The selection of this window aims to cover the critical stage from the initial transition of the signal to its tendency to stabilize. Subsequently, the statistical variance of all current data points within this window is calculated, and this value is used as the core indicator for quantifying signal stability. Next, the calculated statistical variance is compared with a stability threshold that has been verified by numerous experiments. If the statistical variance is lower than the stability threshold, the quality of the current signal is determined to meet the reliability requirements, and the subsequent dynamic response feature extraction process begins. If it does not meet the requirements, the process is immediately interrupted and an error is reported. This embodiment focuses on the stability of the signal in the initial stage and uses statistical variance as an objective indicator for quantitative evaluation, thereby achieving rapid and effective diagnosis of the quality of detection conditions and avoiding invalid feature extraction and type identification under unstable signal conditions.

[0164] According to an embodiment of the present invention, extracting dynamic response features from the current response curve specifically includes:

[0165] Identify the current rise phase in the current response curve and calculate the current rise slope during the current rise phase through linear fitting.

[0166] Traverse the entire current response curve and extract the global maximum value as the peak current;

[0167] The response time is calculated as the duration required from the start of the current response until the current value reaches a specified percentage of its peak value.

[0168] The current rise slope, peak current, and response time are combined into a dynamic response feature vector.

[0169] It should be noted that in this embodiment, morphological analysis is first performed on the preprocessed current response curve to accurately identify the current rise phase characterizing the electrochemical reaction rate, and the slope value of this phase, i.e., the current rise slope, is calculated using a linear fitting algorithm. Subsequently, the data points of the entire current response curve are traversed to find its global maximum value and determine it as the peak current, which reflects the maximum current intensity of the reaction. Simultaneously, the time interval from the start of the current response to the current value rising to a specific preset proportion (e.g., 80%) of the peak current is calculated and defined as the response time, which characterizes the speed at which the reaction reaches near saturation. Finally, the extracted current rise slope, peak current, and response time—the three key parameters—are combined to form a comprehensive dynamic response feature vector. This embodiment constructs a feature model characterizing the differences between test strip types by extracting multi-dimensional features closely related to the chemical reaction kinetics of the test strip.

[0170] According to an embodiment of the present invention, the step of comparing the response time of the dynamic response feature with a preset reference time threshold to obtain a first type of result specifically includes:

[0171] The response time is compared with the preset LH test strip response time threshold and HCG test strip response time threshold to obtain the LH response time deviation and HCG response time deviation.

[0172] Determine whether the LH response time deviation is lower than the HCG response time deviation;

[0173] If so, then the first type result is set to type LH;

[0174] If not, the first type result is set to HCG type;

[0175] Based on the preset type threshold mapping table, the corresponding peak current threshold and rise slope threshold are obtained according to the first type result.

[0176] It should be noted that in this embodiment, after obtaining the response time characteristic value characterizing the reaction kinetics, it is first compared with the preset standard response time thresholds for LH test strips and HCG test strips to obtain the LH response time deviation and HCG response time deviation, respectively. Then, a judgment logic is established by comparing the magnitudes of these two deviations. If the LH response time deviation is less than the HCG response time deviation, the test strip reaction characteristic is determined to be closer to the typical response mode of the LH test strip, and the first type result is initialized to the LH type. Otherwise, it is initialized to the HCG type. Based on this, the type-threshold mapping database pre-stored in the device is accessed, and the peak current threshold and rise slope threshold parameters matching the determined first type result are automatically indexed and called. This embodiment achieves rapid pre-classification of test strip types through preliminary screening of response time; it not only improves the retrieval efficiency of subsequent feature comparison but also provides a targeted comparison benchmark for subsequent accurate discrimination by establishing a dynamic mapping relationship between type and threshold, enhancing the adaptability to different test strips.

[0177] According to an embodiment of the present invention, by comparing the peak current and rise slope of the dynamic response characteristics with the peak current threshold and rise slope threshold, a second type of result is obtained, specifically including:

[0178] Based on the results of the first type, the peak current deviation range and the rise slope deviation range are obtained;

[0179] Determine whether the peak current is within the peak current deviation range with the peak current threshold as the midpoint, and whether the rising slope is within the rising slope deviation range with the rising slope as the midpoint.

[0180] If both conditions are met, then the second type of result is set according to the first type of result;

[0181] If any condition is not met, the type identification is deemed to have failed.

[0182] It should be noted that in this embodiment, after obtaining the first type of result and its corresponding threshold parameter, the allowable deviation range of peak current and rising slope is first determined based on the statistical characteristics of historical data for this type, forming a reasonable fluctuation range centered on the threshold. Subsequently, a dual condition judgment is performed: on the one hand, it is checked whether the actually measured peak current falls within the preset deviation range centered on the peak current threshold; on the other hand, it is simultaneously checked whether the current rising slope is within the allowable deviation range of the corresponding rising slope threshold. Only when both of these key feature parameters simultaneously meet their respective range requirements is the first type of result directly confirmed as the second type of result. If any feature value exceeds its corresponding reasonable fluctuation range, the identification process is immediately interrupted and the type identification failure status is returned. This embodiment, by establishing a dual feature verification mechanism, fully utilizes the directional guidance provided by the pre-classification results and avoids misjudgment caused by mismatch of a single feature through strict interval verification, significantly improving the reliability of the type discrimination result.

[0183] It is worth mentioning that the comprehensive confidence level calculated based on the deviation between the dynamic response characteristics and the corresponding threshold specifically includes:

[0184] Calculate the deviation between the peak current and the peak current threshold to obtain the first confidence component;

[0185] The deviation of the rising slope from the rising slope threshold is calculated to obtain the second confidence component;

[0186] The first confidence component and the second confidence component are weighted and fused to obtain the comprehensive confidence score;

[0187] If the overall confidence level is higher than the first confidence level threshold, then output the second type of result;

[0188] If the overall confidence level is between the first confidence threshold and the second confidence threshold, a retest mechanism is triggered;

[0189] If the overall confidence level is lower than the second confidence level threshold, the output type is an unknown signal.

[0190] It should be noted that in this embodiment, the normalized deviation between the measured peak current and the corresponding threshold is calculated as the first confidence component, and the normalized deviation between the measured rising slope and the corresponding threshold is calculated as the second confidence component. The two components are linearly weighted and fused using preset weighting coefficients to generate a comprehensive confidence score with a normalized value range. Based on the higher first confidence threshold and the lower second confidence threshold, three judgment intervals are formed: when the comprehensive confidence is higher than the first threshold, the second type of result is directly output as the final judgment; when the comprehensive confidence is between the two thresholds, it is determined to be in a critical state and a retesting mechanism is automatically initiated to obtain new detection data; when the comprehensive confidence is lower than the second threshold, the signal quality or feature matching is determined to be too low, and an unknown type signal is output. This embodiment, by establishing a quantitative confidence evaluation index and a hierarchical decision-making mechanism, intelligently distinguishes between confirmed results, boundary cases, and invalid detections, ensuring the efficiency of direct output of high-confidence results while providing a review opportunity for critical cases, thus improving the rationality of decision-making in practical applications.

[0191] It is worth mentioning that it also includes:

[0192] After the test strip is inserted but before the detection excitation is applied, the background current signal of the test strip electrode is collected;

[0193] The background current signal is matched with a pre-stored background feature library to identify the production batch of the test strip;

[0194] Based on the identified production batch, select the corresponding LH response time threshold, HCG response time threshold, peak current threshold, and rise slope threshold from the threshold group mapping table.

[0195] After completing the test strip type identification, the dynamic response characteristics and the final judgment result are fed back to the background feature database.

[0196] It should be noted that in this embodiment, before applying the detection excitation signal, the background current signal of the test strip electrodes in a resting state is first collected to obtain raw information reflecting the characteristics of the test strip substrate. This background current signal is then matched with a pre-stored background feature database to identify the specific production batch or material specification of the test strip through feature comparison. Based on the identified production batch identifier, the LH response time threshold, HCG response time threshold, peak current threshold, and rise slope threshold suitable for that batch of test strips are precisely selected from a mapping table storing multiple threshold parameter groups. After completing the entire process of test strip type identification, the dynamic response feature data obtained in this detection and the final judgment result are fed back to the background feature database as new samples for iterative optimization of the threshold selection strategy. This embodiment effectively eliminates systematic errors caused by differences in test strip production batches through pre-identification of the background and dynamic threshold matching; simultaneously, the continuous learning mechanism enables the system to have self-optimization capabilities, thereby significantly improving the adaptability of the detection system to test strips from different sources and its long-term accuracy.

[0197] It is worth mentioning that it also includes:

[0198] A composite excitation signal comprising multiple frequency components is applied to the test strip electrodes;

[0199] Current response signals under different excitation frequencies were collected to obtain multiple sets of current time-series curves;

[0200] Dynamic response feature extraction is performed on each set of current time-series curves to obtain multiple sets of feature vectors;

[0201] Multiple feature vectors are fused at the feature level to generate enhanced feature vectors;

[0202] The dynamic response features are modified based on the enhanced feature vector.

[0203] It should be noted that in this embodiment, firstly, a composite periodic electrical signal containing multiple components of different frequencies is applied to the test strip electrodes as a detection excitation. Subsequently, the current response signals generated under each frequency component excitation are simultaneously acquired and separated, obtaining multiple sets of current time-series curves that are parallel in the time domain. Secondly, a complete dynamic response feature extraction process is independently performed on the current time-series curves corresponding to each frequency, resulting in multiple feature vectors containing response time, peak current, and rise slope. Simultaneously, a feature-level fusion algorithm is used to integrate the multiple feature vectors from different frequency responses into a high-dimensional enhanced feature vector, which comprehensively reflects the response characteristics of the test strip under different electrical excitation conditions. Finally, this enhanced feature vector is used to correct and supplement the dynamic response features obtained under the original single-excitation condition. This embodiment obtains richer electrochemical response information through multi-frequency excitation, effectively overcoming the limitation of features being easily interfered with under single detection conditions, and improving anti-interference capability and recognition accuracy in complex usage environments.

[0204] A third aspect of the present invention provides a computer-readable storage medium including a test strip type identification method program based on current characteristics, wherein when the test strip type identification method program based on current characteristics is executed by a processor, it implements the steps of the test strip type identification method based on current characteristics as described in any of the preceding claims.

[0205] In summary, this invention provides a method, system, and storage medium for identifying test strip types based on current characteristics. First, raw current signals are acquired through test strip electrodes and preprocessed to generate a current response curve. Second, a stability index for the initial time window of the curve is calculated. When the stability index is lower than a preset stability threshold, dynamic response features are extracted. Then, a first type of result is determined based on the deviation between the response time and a preset reference threshold. A second type of result is obtained by comparing the corresponding peak current with a rise slope threshold. Finally, a comprehensive confidence level is generated by calculating the deviation between the dynamic features and the threshold, which determines the final output type, triggers a retest, or reports an error. This invention overcomes the limitations of single-feature discrimination through multi-level collaboration of signal stability screening, chain threshold discrimination, and confidence level assessment, improving the accuracy and reliability of HCG and LH test strip identification in complex application environments.

[0206] If the aforementioned functions are implemented as software functional modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0207] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for identifying test strip types based on current characteristics, characterized in that, The method includes: The raw current signal is acquired by the test strip electrodes, and the current response curve is obtained based on the electrical signal preprocessing. Based on a preset time window, stability indicators are calculated according to the current response curve. Determine whether the stability index is lower than a preset stability threshold; If not, output an invalid detection signal; If so, dynamic response features are extracted from the current response curve, specifically including: identifying the current rise phase in the current response curve, calculating the current rise slope of the current rise phase through linear fitting; traversing the entire current response curve and extracting the global maximum value as the peak current; calculating the duration required from the start of the current response to the current value reaching a specified proportion of the peak value as the response time; and combining the current rise slope, peak current, and response time into a dynamic response feature vector. The first type of result is obtained by comparing the response time of the dynamic response feature with a preset reference time threshold. Specifically, this includes: comparing the response time with preset LH test strip response time thresholds and HCG test strip response time thresholds to obtain LH response time deviation and HCG response time deviation; determining whether the LH response time deviation is lower than the HCG response time deviation; if so, the first type of result is set to LH type; if not, the first type of result is set to HCG type; and based on a preset type threshold mapping table, the corresponding peak current threshold and rising slope threshold are obtained according to the first type of result. Based on the first type of result, the peak current threshold and the rise slope threshold are obtained; By comparing the peak current and rise slope of the dynamic response characteristics with the peak current threshold and rise slope threshold, a second type of result is obtained; The overall confidence level is calculated based on the deviation between the dynamic response characteristics and the corresponding threshold. Determine whether the overall confidence level is higher than a preset first confidence threshold; If so, output the result of type 2; If not, a retest will be triggered or an invalid detection signal will be output.

2. The test strip type identification method based on current characteristics according to claim 1, characterized in that, The process of acquiring the raw current signal through test strip electrodes and obtaining the current response curve based on electrical signal preprocessing specifically includes: In response to the test strip insertion signal, a preset excitation voltage signal is applied to the test strip electrodes; The raw microcurrent signal flowing through the test strip electrodes is acquired synchronously and converted into a digital current sequence; The digital current sequence is filtered. The filtered digital current sequence is smoothed to generate a current response curve.

3. The test strip type identification method based on current characteristics according to claim 1, characterized in that, The calculation of stability indicators based on the current response curve within a preset time window specifically includes: Based on a preset stabilization time window, the window current data points are divided at the beginning stage of the current response curve. Calculate the statistical variance of the current data points within the window as a stability indicator; The statistical variance is compared with a preset stability threshold; If the statistical variance is lower than the stability threshold, the stability of the current signal is determined to meet the requirements, and the dynamic feature extraction step is initiated.

4. The test strip type identification method based on current characteristics according to claim 1, characterized in that, By comparing the peak current and rise slope of the dynamic response characteristics with the peak current threshold and rise slope threshold, a second type of result is obtained, which specifically includes: Based on the results of the first type, the peak current deviation range and the rise slope deviation range are obtained; Determine whether the peak current is within the peak current deviation range with the peak current threshold as the midpoint, and whether the rising slope is within the rising slope deviation range with the rising slope as the midpoint. If both conditions are met, then the second type of result is set according to the first type of result; If any condition is not met, the type identification is deemed to have failed.

5. A test strip type identification system based on current characteristics, characterized in that, The system includes a memory and a processor. The memory includes a test strip type identification method program based on current characteristics. When the test strip type identification method program based on current characteristics is executed by the processor, it performs the following steps: The raw current signal is acquired by the test strip electrodes, and the current response curve is obtained based on the electrical signal preprocessing. Based on a preset time window, stability indicators are calculated according to the current response curve. Determine whether the stability index is lower than a preset stability threshold; If not, output an invalid detection signal; If so, dynamic response features are extracted from the current response curve, specifically including: identifying the current rise phase in the current response curve, calculating the current rise slope of the current rise phase through linear fitting; traversing the entire current response curve and extracting the global maximum value as the peak current; calculating the duration required from the start of the current response to the current value reaching a specified proportion of the peak value as the response time; and combining the current rise slope, peak current, and response time into a dynamic response feature vector. The first type of result is obtained by comparing the response time of the dynamic response feature with a preset reference time threshold. Specifically, this includes: comparing the response time with preset LH test strip response time thresholds and HCG test strip response time thresholds to obtain LH response time deviation and HCG response time deviation; determining whether the LH response time deviation is lower than the HCG response time deviation; if so, the first type of result is set to LH type; if not, the first type of result is set to HCG type; and based on a preset type threshold mapping table, the corresponding peak current threshold and rising slope threshold are obtained according to the first type of result. Based on the first type of result, the peak current threshold and the rise slope threshold are obtained; By comparing the peak current and rise slope of the dynamic response characteristics with the peak current threshold and rise slope threshold, a second type of result is obtained; The overall confidence level is calculated based on the deviation between the dynamic response characteristics and the corresponding threshold. Determine whether the overall confidence level is higher than a preset first confidence threshold; If so, output the result of type 2; If not, a retest will be triggered or an invalid detection signal will be output.

6. The test strip type identification system based on current characteristics according to claim 5, characterized in that, The process of acquiring the raw current signal through test strip electrodes and obtaining the current response curve based on electrical signal preprocessing specifically includes: In response to the test strip insertion signal, a preset excitation voltage signal is applied to the test strip electrodes; The raw microcurrent signal flowing through the test strip electrodes is acquired synchronously and converted into a digital current sequence; The digital current sequence is filtered. The filtered digital current sequence is smoothed to generate a current response curve.

7. The test strip type identification system based on current characteristics according to claim 5, characterized in that, The calculation of stability indicators based on the current response curve within a preset time window specifically includes: Based on a preset stabilization time window, the window current data points are divided at the beginning stage of the current response curve. Calculate the statistical variance of the current data points within the window as a stability indicator; The statistical variance is compared with a preset stability threshold; If the statistical variance is lower than the stability threshold, the stability of the current signal is determined to meet the requirements, and the dynamic feature extraction step is initiated.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that, The computer-readable storage medium includes a test strip type identification method program based on current characteristics. When the test strip type identification method program based on current characteristics is executed by a processor, it implements the steps of the test strip type identification method based on current characteristics as described in any one of claims 1 to 4.

Images