A syphilis non-specific antibody detection signal pattern recognition method and system

Abstract

The application discloses a syphilis non-specific antibody detection signal mode recognition method and system, relates to the field of syphilis non-specific antibody detection signal mode recognition, and is used for improving the accuracy and reliability of syphilis non-specific antibody detection and comprises the following steps: obtaining interference indication information of a to-be-detected sample, wherein the interference indication information reflects inherent optical characteristics of the to-be-detected sample; adjusting parameters for feature extraction of a signal waveform of the to-be-detected sample according to the interference indication information; extracting features from the signal waveform according to the adjusted parameters; and adjusting interpretation logic according to the interference indication information and the extracted features to obtain an interpretation result of the to-be-detected sample.

Description

Technical Field

[0001] This invention relates to the field of pattern recognition of non-specific antibody detection signals for syphilis, and particularly to a method and system for pattern recognition of non-specific antibody detection signals for syphilis. Background Technology

[0002] In clinical laboratory testing, non-specific antibody testing for syphilis is a crucial step in screening and assisting in the diagnosis of syphilis. Automated testing systems capture the light signals generated by biological reactions and convert them into electrical waveforms for analysis, enabling objective interpretation of the results. However, in practical applications, the complexity of the samples themselves, such as the presence of interfering substances like hyperlipidemia, and the challenges of improving detection sensitivity, often cause weak true reaction signals to be confused with interfering signals. This makes it difficult for traditional methods to accurately identify samples in critical states, thus affecting the accuracy and efficiency of diagnosis.

[0003] In automated detection of nonspecific antibodies for syphilis, methods based on the principle of flocculation reaction are crucial for large-scale screening. This method uses an optical detection system to capture changes in the optical signal caused by the formation of flocculent agglutination particles from antigen-antibody binding, converting these changes into electrical signal waveforms for analysis. However, to improve the detection sensitivity for samples from patients with early infection or those whose antibody titers have decreased after treatment, the detection system often requires a high-sensitivity setting. This high-sensitivity setting, while amplifying weak biological reaction signals, also amplifies background interference from the sample itself. For example, hyperlipidemic samples (chylous blood) contain a large number of lipoprotein particles, which are similar in size to early-formed antigen-antibody agglutination particles. When the reaction plate rotates and shakes within the device, these lipoprotein particles also move, aggregate, and disperse irregularly in the liquid, scattering and blocking the passing light beam, creating interference signals unrelated to the actual agglutination reaction.

[0004] The interference signal generated by the physical characteristics of the sample is not simple random noise; it also exhibits certain time-varying patterns. Its waveform may even closely resemble the waveform of a genuine weak positive reaction, exhibiting a slow rise or plateau. This results in a composite waveform composed of a genuine, weak biochemical reaction signal and a stronger interference signal generated by the sample's physical characteristics. In this situation, traditional waveform feature extraction methods struggle to distinguish which changes are caused by the target reaction and which are caused by lipoprotein particle movement. For example, a signal peak caused by lipoprotein particle aggregation might be incorrectly identified as a feature of a positive reaction, leading to a false positive for a hyperlipidemia sample that was originally negative. Conversely, a genuine weak positive reaction signal may be masked or distorted by interference signals, preventing its features from being correctly extracted and resulting in a false negative. This confusion significantly increases the false positive and false negative rates in the detection of special samples such as hyperlipidemia using traditional feature extraction and pattern recognition methods, severely impacting the reliability of the test results. Summary of the Invention

[0005] This invention provides a method for pattern recognition of syphilis nonspecific antibody detection signals, which can improve the accuracy and reliability of syphilis nonspecific antibody detection.

[0006] In a first aspect, this application discloses a method for recognizing patterns in non-specific antibody detection signals for syphilis, including: Obtain interference indication information of the sample under test, which reflects the inherent optical characteristics of the sample under test; Based on the interference indication information, adjust the parameters for feature extraction of the signal waveform of the sample under test; Based on the adjusted parameters, extract features from the signal waveform; Based on the interference indication information and the extracted features, the interpretation logic is adjusted to obtain the interpretation result of the sample to be tested.

[0007] Through this technical solution, this application can dynamically adjust the signal feature extraction and interpretation logic based on the interference information reflected by the inherent optical properties of the sample itself, thereby effectively distinguishing between real biological reaction signals and interference signals. This solves the problem of low accuracy in identification of complex samples by traditional methods and significantly improves the accuracy and reliability of syphilis non-specific antibody detection.

[0008] Furthermore, based on the above method, according to the interference indication information, the parameters for feature extraction of the signal waveform of the sample under test are adjusted, including: Continuously monitor the local variation trend and frequency components of the signal waveform; When the local change trend of the signal waveform is detected to match the preset interference mode, or when the frequency components of the signal waveform are enhanced in a specific frequency band, dynamic interference is determined to exist. In response to the detection of dynamic interference, the cutoff frequency of the digital filter and the time window length of the waveform analysis are adjusted in real time according to the intensity and frequency characteristics of the dynamic interference to obtain the adjusted parameters.

[0009] Through this technical solution, this application can monitor the dynamic changes of signal waveforms in real time and adaptively adjust the feature extraction parameters according to the identified dynamic interference characteristics, effectively filtering out interference signals and ensuring that the true characteristics of biological reactions can still be accurately captured in complex interference environments, further improving the accuracy of feature extraction.

[0010] Based on this, this application further proposes adjusting the interpretation logic according to the interference indication information and the extracted features to obtain the interpretation result of the sample to be tested, including: Based on the interference indication information and the extracted features, a preliminary interpretation is performed to obtain the preliminary interpretation result; When the preliminary interpretation result is within the preset critical interpretation range, a consistency evaluation is performed on multiple independent features to obtain a consistency evaluation result; Based on the consistency assessment results, the interpretation logic is adjusted to obtain the interpretation result of the sample to be tested.

[0011] Through this technical solution, this application introduces a multi-feature consistency evaluation mechanism for critical samples based on preliminary interpretation, which effectively avoids the limitations of single-feature interpretation, improves the accuracy of interpretation of fuzzy samples, and reduces the risk of misjudgment and missed judgment.

[0012] Furthermore, in some preferred embodiments, when the preliminary interpretation result is within a preset critical interpretation interval, a consistency evaluation is performed on multiple independent features, including: Each of the multiple independent features is weighted, and the weight of each independent feature is adjusted based on the feature’s historical performance in distinguishing real biological response signals from specific interference signals and the signal-to-noise ratio of the current signal waveform; Calculate the weighted aggregate score of these multiple independent features; The aggregated score is compared with multiple preset interpretation regions to determine whether the aggregated score falls within a fuzzy undetermined region. When the aggregated score falls into the fuzzy undetermined region, a cross-validation mechanism based on the nonlinear relationship between features is initiated to identify highly overlapping signal patterns in that feature. Based on the results of cross-validation, the aggregation score is adjusted to obtain the consistency assessment result.

[0013] Through this technical solution, this application refines the processing of samples in the critical interpretation interval by using dynamic weighting and cross-validation mechanisms, which can more accurately identify and distinguish highly overlapping signal patterns, and significantly improve the interpretation accuracy and robustness of critical samples under complex interference.

[0014] As a technical improvement, the weight of each individual feature is determined, including: Continuously monitor the dynamic changes in biological response characteristics and interference characteristics in the signal waveform; When the first or second condition is met, a slight change is identified in the biological reaction kinetics or interference characteristics of the sample to be tested; the first condition is that the upward trend, plateau duration, or response stability to mechanical stirring of the biological reaction signal deviates from the preset reference template; the second condition is that the frequency components and amplitude variation of the interference signal show a new match with the preset interference mode. In response to the identified change, the weight of each individual feature is dynamically adjusted.

[0015] Through this technical solution, this application can monitor the minute dynamic changes of biological reactions and interference characteristics in real time, and dynamically adjust the feature weights accordingly, so that the interpretation system can adapt to individual differences in samples and environmental changes, further improving the adaptability and accuracy of interpretation.

[0016] To enhance functionality, the dynamic changes in biological response characteristics and interference characteristics within the signal waveform are continuously monitored, including: Multiple optical detection channels are set up, and each channel is equipped with a light source and detector of different wavelengths or different polarization states; The signals acquired by each optical detection channel are preliminarily processed to extract their respective waveform features; The differences and correlations of signal characteristics among different optical detection channels were analyzed, and the independent effects of different interfering substances on the signal were identified and separated. The separated interference components are removed from the signal to monitor the dynamic changes of biological response characteristics and interference characteristics in the signal waveform.

[0017] Through this technical solution, this application utilizes multi-channel optical detection technology to effectively separate and remove the independent influence of different interfering substances on the signal, thereby obtaining biological reaction characteristics more purely and significantly improving the extraction accuracy and anti-interference ability of biological reaction signals.

[0018] As a further improvement, the method also includes: Set up multiple signal analysis modules, each with specific filtering parameters and time window parameters; Parallel processing of signal waveforms yields the outputs of multiple modules; Evaluate the output of each module independently to determine whether the first condition or the second condition is met. When the evaluation results of multiple modules are contradictory or inconsistent, the signal source tracing mechanism is activated; Based on the source tracing results of the signal, determine whether the first condition or the second condition is met.

[0019] This technical solution introduces a multi-module parallel processing and signal source tracing mechanism, which can effectively solve the problem of inconsistent evaluation results of multiple modules. By tracing the source of the signal, it ensures accurate judgment of changes in biological response or interference characteristics, and further improves the reliability of the system and the accuracy of decision-making.

[0020] Building upon the above, this application further proposes that when the evaluation results of multiple modules contradict or are inconsistent, a signal source tracing mechanism be initiated, including: The contradictory or inconsistent patterns in the evaluation results of the multiple modules are classified and identified to obtain the classification and identification results; For the confusion pattern indicated by the classification and recognition results, a dedicated signal decoupling strategy is activated to decouple the signal waveform and obtain the decoupled signal components. During the signal decoupling process, the purity of each signal component after decoupling is monitored to obtain purity feedback; Based on the purity feedback, adjust the decoupling parameters of the signal decoupling strategy; Based on the adjusted decoupling parameters, the signal waveform is decoupled to obtain a pure biological response signal and interference signal; Based on the pure biological response signal and the interference signal, determine whether the first condition or the second condition is met.

[0021] Through this technical solution, this application can effectively separate confused signals by classifying and identifying the contradiction assessment results and activating a dedicated signal decoupling strategy. By dynamically optimizing the decoupling process through purity feedback, pure biological reaction signals and interference signals can be obtained, which greatly improves the ability to analyze complex signals and the accuracy of judgment.

[0022] To optimize the structure, in response to the identified change, the weights of each individual feature are dynamically adjusted, including: If a slow upward trend in a biological response is identified, the weight of later-stage response features is increased to compensate for the weakness of earlier signals. If an increase in the frequency components of a specific interference signal is detected, the weight of features that are more affected by interference in that frequency band is reduced.

[0023] Through this technical solution, this application can adjust the feature weights in a targeted manner according to the specific changing trends of biological reactions and interference signals, so that the interpretation system can adapt to different situations more flexibly, effectively compensate for weak signals or suppress interference effects, and further improve the accuracy and robustness of interpretation.

[0024] Secondly, this application also discloses a syphilis non-specific antibody detection signal pattern recognition system, the system comprising: The information acquisition module is used to acquire interference indication information of the sample under test, which reflects the inherent optical characteristics of the sample under test. The parameter adjustment module is used to adjust the parameters for feature extraction of the signal waveform of the sample under test according to the interference indication information. The feature extraction module is used to extract features from the signal waveform based on the adjusted parameters; The interpretation logic adjustment module is used to adjust the interpretation logic based on the interference indication information and the extracted features in order to obtain the interpretation result of the sample to be tested.

[0025] This application provides a system for implementing the above method through this technical solution. Through modular design, it can efficiently acquire interference information, adjust parameters, extract features, and adjust the interpretation logic, providing an integrated and intelligent solution for the detection of non-specific antibodies against syphilis, and significantly improving the automation level and accuracy of the detection.

[0026] Beneficial effects This application discloses a signal pattern recognition method for syphilis nonspecific antibody detection. By acquiring interference indication information of the sample to be tested, which reflects the sample's inherent optical properties, the method dynamically adjusts the signal waveform feature extraction parameters and interpretation logic accordingly. This method effectively solves the problem in existing technologies where interference from complex samples such as hyperlipidemia leads to confusion between genuine weak biological reaction signals and interference signals, making it difficult for traditional methods to accurately identify samples in critical states. Specifically, this application uses the sample's inherent optical properties as interference indication information, enabling the system to pre-sensitize and quantify potential interference sources, thus allowing for targeted parameter adjustments in the early stages of signal processing. This adaptive adjustment mechanism based on interference indication information effectively distinguishes interference signals generated by the sample's physical properties from genuine biochemical reaction signals, avoiding misjudgments and missed judgments caused by the similarity between interference signals and weak positive reaction waveforms in traditional methods. In this way, this application significantly improves the accuracy and reliability of syphilis nonspecific antibody detection, especially when dealing with special samples such as hyperlipidemia, effectively reducing false positive and false negative rates, thereby improving the efficiency and quality of clinical diagnosis. Attached Figure Description

[0027] Figure 1This is a schematic flowchart of a method for recognizing non-specific antibody detection signals for syphilis provided in an embodiment of the present invention; Figure 2 This is a schematic diagram of another method for recognizing non-specific antibody detection signals for syphilis provided in an embodiment of the present invention; Figure 3 This is a schematic diagram of a syphilis non-specific antibody detection signal pattern recognition system provided in an embodiment of the present invention. Detailed Implementation

[0028] The technical solutions of this application will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are merely some embodiments of this application, and not all embodiments. The components of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely to illustrate selected embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0029] It should be noted that similar reference numerals and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. Furthermore, in the description of this application, terms such as "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0030] In clinical laboratory testing, non-specific antibody testing for syphilis is a crucial step in screening and assisting in the diagnosis of syphilis. Automated testing systems capture the light signals generated by biological reactions and convert them into electrical waveforms for analysis, enabling objective interpretation of the results. However, in practical applications, the complexity of samples, such as those containing interfering substances like hyperlipidemia, and the challenges of improving detection sensitivity, often lead to confusion between weak true reaction signals and interfering signals. This makes it difficult for traditional methods to accurately identify samples in borderline states, thus affecting the accuracy and efficiency of diagnosis. When processing special samples such as those with hyperlipidemia, existing traditional non-specific antibody testing methods for syphilis generate interfering signals similar to true biological reaction signals due to the inherent optical properties of the samples. This makes it difficult to accurately distinguish the signal waveforms, significantly increasing the false positive and false negative rates, severely impacting the reliability of the test and the accuracy of diagnosis.

[0031] In response, this application proposes a method for pattern recognition of non-specific antibody detection signals for syphilis, comprising: Obtain interference indication information of the sample under test, which reflects the inherent optical properties of the sample under test; Based on the interference indication information, adjust the parameters for feature extraction of the signal waveform of the sample under test; Based on the adjusted parameters, features are extracted from the signal waveform; Based on the interference indication information and the extracted features, the interpretation logic is adjusted to obtain the interpretation result of the sample to be tested.

[0032] This application effectively distinguishes between real biological reaction signals and interference signals by acquiring interference indication information of the sample to be tested and dynamically adjusting feature extraction parameters and interpretation logic based on this information, which significantly improves the accuracy and reliability of syphilis non-specific antibody detection, especially when dealing with complex samples.

[0033] To better understand the syphilis non-specific antibody detection signal pattern recognition method proposed in this application, some key terms involved will be explained first.

[0034] "Interference indication information" refers to data that reflects the inherent optical characteristics of the sample under test, such as the turbidity, color, and lipid content of the sample. These characteristics may cause scattering, absorption, or refraction of light signals, thereby generating interference signals.

[0035] "Signal waveform" refers to an electrical signal that changes over time and is acquired through an optical detection system. It includes biological response signals and potential interference signals.

[0036] "Feature extraction parameters" refer to the configuration of the algorithm or model used when analyzing signal waveforms, such as the filter cutoff frequency, time window length, and threshold of the feature extraction algorithm.

[0037] "Analysis logic" refers to the rules or models used to transform extracted features into final analysis results (such as positive, negative, or critical).

[0038] The following specific embodiments will provide a detailed description and explanation of the syphilis non-specific antibody detection signal pattern recognition method provided in this application.

[0039] Reference Figure 1 This invention provides a method for recognizing patterns in non-specific antibody detection signals for syphilis, comprising the following steps: S1, Obtain interference indication information of the sample to be tested.

[0040] Among them, the interference indication information reflects the inherent optical characteristics of the sample under test.

[0041] One possible approach is to integrate an additional optical sensor into the detection system. This sensor, located outside the main detection optical path, is specifically designed to measure the turbidity or absorbance of the sample. Specifically, a separate LED light source and photodiode array can be used to measure the transmitted or scattered light at multiple wavelengths without affecting the main reaction detection, thereby obtaining data on the sample's inherent optical properties. This data, such as absorbance values ​​or scattered light intensity at specific wavelengths, can be directly used as interference indication information.

[0042] Another approach is to perform a rapid scan using the light source and detector of the main detection system after the sample is added to the reaction well, before the biological reaction begins, and record the baseline signal at this point. Since the biological reaction has not yet occurred, fluctuations or shifts in the baseline signal mainly reflect the inherent optical properties of the sample itself, and therefore can be used as interference indicators. For example, for hyperlipidemic samples, their absorbance at a specific wavelength will be significantly higher than that of normal samples; this difference in absorbance can serve as an interference indicator.

[0043] S2. Based on the interference indication information, adjust the parameters for feature extraction of the signal waveform of the sample under test.

[0044] As one possible approach, if the interference indication information shows high sample turbidity, indicating strong scattering interference, the cutoff frequency of the digital filter can be adjusted by a preset step size to lower it, thereby filtering out high-frequency scattering noise. Simultaneously, the time window length for waveform analysis can be shortened by a preset step size to reduce random interference accumulated over a long period.

[0045] In some embodiments, if the interference indication information shows that the sample color is dark, there may be light absorption interference. In this case, the baseline correction parameters of the feature extraction algorithm can be adjusted to make it more actively compensate for baseline drift, or the threshold of feature recognition can be adjusted to avoid misjudging baseline changes caused by color as biological response signals.

[0046] S3. Based on the adjusted parameters, extract features from the signal waveform.

[0047] As one possible approach, by adjusting the cutoff frequency and time window length of the digital filter, a Fourier transform can be performed on the signal waveform to extract its energy distribution within a specific frequency band as a feature.

[0048] Alternatively, wavelet transform can be applied to extract local features of the signal waveform at different scales, such as peaks, valleys, slopes, and areas. These features can be single numerical values ​​or multi-dimensional vectors. For example, the rising slope, plateau duration, and maximum signal strength of the signal waveform can be extracted as features.

[0049] S4. Based on the interference indication information and the extracted features, adjust the interpretation logic to obtain the interpretation result of the sample to be tested.

[0050] As one possible approach, if the interference indication information shows that the sample has slight interference and the extracted features are in the critical positive range, the interpretation threshold can be manually increased so that only stronger signals will be interpreted as positive, thereby reducing the false positive rate.

[0051] As another possible approach, if the interference indication information shows that the sample has serious interference, even if the extracted features show a weak positive result, the interpretation result can be marked as "to be retested" or "uncertain" instead of directly giving a positive result.

[0052] The syphilis non-specific antibody detection signal pattern recognition method proposed in this application effectively overcomes the limitations of traditional detection methods by incorporating the inherent optical characteristics of the sample (i.e., interference indication information) into the entire process of signal analysis and interpretation. Traditional methods often use uniform feature extraction parameters and interpretation logic for all samples. In special samples such as hyperlipidemia, the confusion between interference signals and real biological reaction signals leads to high false positive and false negative rates.

[0053] Specifically, this application first acquires interference indication information of the sample to be tested, which directly reflects the inherent optical properties of the sample, such as its turbidity or color. This step is one of the core innovations that distinguishes this application from existing technologies, because it enables the system to "sense" the potential level of interference in the sample. For example, when a high level of turbidity is detected in the sample, the system can predict the possible scattering interference caused by lipoprotein particles.

[0054] Subsequently, based on the acquired interference indication information, the system can dynamically adjust the parameters for feature extraction from the signal waveform. For example, if the interference indication information indicates high turbidity in the sample, the system can automatically adjust the cutoff frequency of the digital filter to more effectively filter out high-frequency scattering noise and shorten the waveform analysis time window, thereby reducing the impact of interference on feature extraction. This adaptive parameter adjustment mechanism ensures that purer and more accurate biological reaction features can be extracted from samples with different levels of interference, avoiding the problem of extracting erroneous features from interference samples using traditional fixed-parameter methods.

[0055] Next, features are extracted from the signal waveform based on the adjusted parameters. Since the parameters have been optimized for interference, the extracted features more accurately reflect the true biological response signal while reducing the intrusion of interference signals. For example, after optimized filtering and time window processing, the extracted signal peaks, slopes, and other features will more realistically represent the kinetics of the antigen-antibody agglutination reaction.

[0056] Finally, based on the interference indication information and the extracted features, the interpretation logic is adjusted to obtain the interpretation result of the sample to be tested. This step further enhances the ability to cope with interference. For example, if the interference indication information shows that the sample has severe interference, even if the extracted features slightly exceed the standard positive threshold, the system can still interpret it as "borderline" or "awaiting retesting" according to the adjusted interpretation logic, instead of directly interpreting it as "positive". This interpretation logic adjustment based on the degree of interference effectively avoids misjudgment caused by interference and significantly improves the accuracy and reliability of the interpretation results.

[0057] Compared to the closest existing technology, the advantage of this application lies in its closed-loop control of interference through "sensing-adjustment-interpretation." Existing technologies typically perform feature extraction and interpretation directly after signal acquisition, lacking proactive assessment and adaptive adjustment of the inherent interference characteristics of the sample. For example, traditional methods may use the same filter parameters for all samples, which can lead to ineffective filtering of interference signals in hyperlipidemic samples, thus affecting the accuracy of subsequent feature extraction. This application, however, introduces interference indication information, enabling the system to dynamically optimize signal processing and interpretation strategies based on the specific characteristics of the sample, thereby achieving high accuracy even in complex samples. This adaptability and intelligence are the core innovations of this application, significantly reducing false positive and false negative rates in syphilis non-specific antibody detection when facing special samples such as hyperlipidemic samples, thus improving the reliability of clinical diagnosis.

[0058] In some embodiments described above, a method for adjusting signal waveform feature extraction parameters based on interference indication information of the sample under test is proposed. However, in actual detection processes, in addition to static interference that may be caused by the inherent optical properties of the sample, there may also be dynamic interference caused by environmental factors or the detection process itself, such as bubbles, particle movement, or instrument vibration. These dynamic interferences are instantaneous and variable. If the parameters are adjusted only based on the pre-acquired static interference indication information, it may not be able to effectively cope with these real-time changing interferences, thereby affecting the accuracy of feature extraction and the reliability of the interpretation results.

[0059] In this regard, such as Figure 2 As shown, in order to adjust the parameters for feature extraction of the signal waveform of the sample under test based on the interference indication information, this application may further include the following steps: S101. Continuously monitor the local variation trend and frequency components of the signal waveform.

[0060] Specifically, continuous monitoring of the local variation trend and frequency components of a signal waveform refers to the uninterrupted analysis of the signal waveform during signal acquisition. The local variation trend can be understood as the slope, curvature, or instantaneous amplitude fluctuation of the signal within a short time window, with the aim of capturing sudden changes or abnormal behavior in the signal. Frequency component monitoring aims to analyze the energy distribution of the signal in different frequency bands in real time using methods such as Fourier transform and wavelet analysis to identify noise or interference signals at specific frequencies.

[0061] S102. When the local change trend of the signal waveform is detected to match the preset interference mode, or when the frequency components of the signal waveform are enhanced in a specific frequency band, it is determined that dynamic interference exists.

[0062] The preset interference patterns can be trained and defined in advance using experiments or historical data. For example, a bubble passing through the detection area may generate specific spikes or step signals, while instrument vibration may cause periodic fluctuations at specific frequencies. When the signal characteristics monitored in real time closely match these preset patterns, or when a significant enhancement is detected in a frequency band where high energy should not normally occur, dynamic interference can be identified.

[0063] S103. In response to the determination of the existence of dynamic interference, the cutoff frequency of the digital filter and the time window length of the waveform analysis are adjusted in real time according to the intensity and frequency characteristics of the dynamic interference to obtain the adjusted parameters.

[0064] Specifically, the intensity of dynamic interference can be quantified based on its amplitude, duration, or energy, while frequency characteristics refer to the main frequency range of the interference signal. Adjusting the cutoff frequency of digital filters aims to precisely filter out interference frequency bands while preserving biological response signals to the maximum extent possible. For example, if high-frequency noise is detected, the cutoff frequency of the low-pass filter can be lowered; if specific narrowband interference is detected, a band-stop filter can be enabled and its center frequency and bandwidth adjusted. Adjusting the time window length for waveform analysis can be optimized based on the duration or periodicity of the interference. For example, for transient interference, the analysis window can be shortened for a faster response; for periodic interference, the window length can be adjusted to avoid interference peaks or for averaging.

[0065] This application's solution introduces continuous monitoring of local variation trends and frequency components of the signal waveform, enabling the system to perceive and identify dynamic interference that is difficult to capture using traditional static interference indication information. It is precisely because of this ability to dynamically determine the existence and characteristics of interference that the system can adaptively adjust the cutoff frequency of the digital filter and the time window length of the waveform analysis in real time, based on the intensity and frequency characteristics of the interference. This dynamic adjustment mechanism ensures that the feature extraction parameters can quickly adapt to the new signal environment when interference occurs, thereby effectively suppressing the masking or distortion of biological response signals by interference and guaranteeing the accuracy of subsequent feature extraction.

[0066] Through the above technical solution, this application overcomes the limitations of relying solely on static interference indication information for parameter adjustment. By real-time monitoring and dynamic adjustment, the system can effectively cope with various instantaneous and variable dynamic interferences occurring during the detection process, significantly improving the robustness and accuracy of signal waveform feature extraction. Therefore, even in complex detection environments, it can more accurately extract real biological response characteristics from signal waveforms, thus providing a solid foundation for subsequent interpretation logic adjustments and the reliability of the final interpretation results.

[0067] In some preferred embodiments, a specific example is given below. Suppose that during the detection of nonspecific antibodies against syphilis, due to instability in sample processing or the fluid transport system, tiny bubbles occasionally pass through the optical detection area. These bubbles manifest as transient, high-amplitude spikes or rapid localized changes in the signal waveform, and may also cause energy enhancement at higher frequencies.

[0068] The system first continuously monitors the local variation trend and frequency components of the signal waveform. When a bubble passes through, the monitoring module immediately detects a steep rising and falling edge in the signal waveform, and its local variation trend closely matches the preset "bubble interference mode." Simultaneously, the frequency analysis module may detect a transient increase in signal energy within a specific frequency band, such as 500Hz to 1kHz.

[0069] At this point, the system detects the presence of dynamic interference. In response, the system makes real-time adjustments based on the intensity of the bubble interference (e.g., the amplitude of the spike) and its frequency characteristics (e.g., primarily concentrated in the 500Hz-1kHz range). Specifically, the cutoff frequency of the digital filter may be temporarily lowered to more effectively filter out high-frequency noise components caused by the bubble; simultaneously, the time window length for waveform analysis may be shortened to quickly skip or isolate the brief period affected by the bubble, preventing it from contaminating feature extraction within the entire analysis window. Through this real-time, adaptive parameter adjustment, even when bubble interference occurs, the system can ensure the extraction of purer and more accurate biological response features from the interfered signal waveform, thereby avoiding misjudgments or missed detections.

[0070] In some embodiments described above, while adjusting the interpretation logic based on interference indication information and extracted features to obtain the interpretation results of the sample to be tested is proposed, in practical applications, when the preliminary interpretation results are in a critical state or ambiguous range, simple logic adjustment alone may not be able to effectively distinguish between real biological reaction signals and complex interference signals, thus leading to a decrease in the accuracy and reliability of the interpretation results. If the above problems are not addressed, misjudgments or missed judgments may occur, affecting the clinical diagnostic accuracy of syphilis non-specific antibody detection. Therefore, this application further proposes a more refined method for adjusting the interpretation logic, introducing preliminary interpretation, critical interpretation range judgment, and consistency assessment mechanisms to improve the robustness and accuracy of the interpretation results.

[0071] In response, this application further proposes adjusting the interpretation logic based on the interference indication information and extracted features to obtain the interpretation result of the sample to be tested, including: S201. Based on the interference indication information and the extracted features, a preliminary interpretation is performed to obtain the preliminary interpretation result.

[0072] Specifically, preliminary interpretation refers to the system's initial and rapid assessment of the current sample under test based on the acquired interference indication information and features extracted from the signal waveform, using a pre-set interpretation model or algorithm to generate a preliminary interpretation result. This preliminary interpretation result can be a continuous numerical value, such as a response intensity score, or a preliminary classification, such as "positive tendency," "negative tendency," or "pending." Its purpose is to quickly screen samples with a clear interpretation tendency and identify those critical samples that require further refined analysis.

[0073] S202. When the preliminary interpretation result is within the preset critical interpretation range, perform a consistency evaluation on multiple independent features to obtain a consistency evaluation result.

[0074] The pre-defined critical interpretation interval can be understood as one or more intervals defined within the numerical range of the initial interpretation results. Results within these intervals indicate that the interpretation status of the sample is unclear and may be influenced by a combination of factors, requiring further analysis to reach a final conclusion. For example, in the detection of non-specific antibodies for syphilis, this interval may correspond to the signal intensity range of weakly positive, weakly negative, or indeterminate results. Its purpose is to accurately identify samples that are difficult to interpret and easily affected by interference, avoiding simplistic and arbitrary interpretations.

[0075] In practical applications, consistency evaluation of multiple independent features refers to the process where, when the initial interpretation result falls within the critical interpretation range, the system no longer relies solely on a single comprehensive interpretation result. Instead, it cross-validates and comprehensively considers multiple independent features extracted from the signal waveform, each with different biological or physical meanings. For example, these independent features may include the signal's peak height, rise slope, plateau duration, background noise level, and the intensity of specific frequency components. Consistency evaluation aims to determine the reliability of the initial interpretation result by analyzing the interrelationships, synergies, or contradictions among these independent features. Its purpose is to improve the accuracy and anti-interference capability of the interpretation through multi-dimensional and multi-angle verification.

[0076] S203. Based on the consistency assessment results, adjust the interpretation logic to obtain the interpretation results of the sample to be tested.

[0077] For example, if a consistency assessment indicates that multiple independent features point to a positive result, the interpretation logic will be adjusted to be more inclined towards a positive result, even if the initial interpretation is in the critical range. Conversely, if the assessment results show contradictions or interference between features, the interpretation logic may be adjusted to be more inclined towards a negative result or require retesting. The purpose is to make the final interpretation result more accurate and reliable, especially when dealing with complex or ambiguous signals.

[0078] This application's solution effectively addresses the problem of insufficient accuracy in processing critical or ambiguous signals using traditional methods by introducing a phased interpretation logic adjustment mechanism. First, through preliminary interpretation, the system can quickly filter out samples with clear interpretation tendencies and identify those in a critical state. Separating these critical samples from the clear samples makes subsequent refined processing possible. Second, when the preliminary interpretation result falls within a preset critical interpretation range, the system no longer simply performs an interpretation but initiates a consistency evaluation of multiple independent features. This multi-feature cross-validation mechanism can examine the authenticity of the signal from different dimensions. For example, by comparing the biological response characteristics and interference characteristics of the signal, the source and nature of the signal can be determined more accurately. It is this multi-dimensional evaluation that enables the system to identify complex patterns that a single feature might not reveal, thereby effectively distinguishing genuine biological response signals from various interference signals. Finally, based on the consistency evaluation results, the interpretation logic is dynamically adjusted to ensure the robustness of the final interpretation result. For example, when the consistency assessment results show that the signal has a high degree of biological consistency, the system can more confidently give a positive interpretation even if the initial interpretation result is low; conversely, when the assessment results show significant interference or inconsistency in characteristics, the system will tend to give a negative or indeterminate interpretation to avoid misdiagnosis.

[0079] Through the above technical solutions, this application significantly improves the accuracy and reliability of syphilis non-specific antibody detection when handling critical or ambiguous signals. Compared to basic solutions that rely solely on interference indications and extracted features for direct interpretation, the preliminary interpretation and critical interpretation interval judgment mechanism introduced in this application effectively identifies samples requiring refined processing, avoiding misjudgments that may result from simple interpretation. Furthermore, by conducting consistency assessments on multiple independent features, the system can verify the authenticity of signals from multiple dimensions and angles, effectively distinguishing between genuine biological reaction signals and complex interference signals, thereby significantly reducing the risk of false positives or false negatives caused by inherent optical properties or dynamic interference of samples. This refined interpretation logic adjustment makes the final interpretation results more robust and reliable when facing complex and variable clinical samples, providing a more reliable basis for clinical diagnosis.

[0080] In some preferred embodiments, a specific example is given below. Suppose that in a syphilis nonspecific antibody test, after feature extraction, the initial interpretation of a test sample's signal waveform shows a reaction intensity score of 0.45. At this point, the preset critical interpretation interval is defined as 0.35 to 0.55. Since 0.45 falls within this interval, the system will initiate a consistency assessment of multiple independent features. Specifically, the system will evaluate independent features such as the signal's peak height, rise slope, plateau duration, background noise level, and the intensity of specific interference frequency components. For example, if the peak height and rise slope both show a trend similar to a typical positive reaction, but the plateau duration is slightly shorter and the background noise level is slightly higher, this indicates a possible weak positive reaction accompanied by some degree of interference. In the consistency assessment, the system will synthesize these features, for example, by calculating an aggregate score using a weighted average or machine learning model. If the aggregate score is still in the ambiguous undetermined region, the system will further initiate a cross-validation mechanism based on the nonlinear relationship between features to identify highly overlapping signal patterns. For example, the true source of the signal can be determined by analyzing the nonlinear relationship between peak height and rise slope, and their interaction with background noise levels. Based on the cross-validation results, if the final consistency assessment leans towards a genuine biological response, the interpretation logic will be adjusted to "weakly positive, retest recommended" even if the initial interpretation is borderline; conversely, if the assessment strongly points to interference, the interpretation logic will be adjusted to "negative, interference may exist." In this way, the proposed solution avoids making simple but potentially erroneous judgments in borderline situations, thereby improving the accuracy and reliability of the detection.

[0081] In some embodiments described above, when the preliminary interpretation result falls within a preset critical interpretation interval, a consistency assessment of multiple independent features is required. However, in practical applications, signals within the critical interpretation interval are often complex and susceptible to various factors. Simple feature consistency assessment may not effectively distinguish between genuine biological reaction signals and background interference signals, leading to the risk of misjudgment or missed judgment. Failure to address these issues may reduce the accuracy and reliability of the detection results, especially in the interpretation of low-concentration or high-interference samples. To address this, this application further proposes a more refined and robust consistency assessment mechanism. By introducing dynamic weighting, aggregated score calculation, fuzzy undetermined region identification, and cross-validation, the accuracy of interpretation results within the critical interpretation interval is improved.

[0082] When the preliminary interpretation result is within the preset critical interpretation interval, the consistency evaluation of multiple independent features is performed, including: S301. Perform weighted processing on each of the multiple independent features.

[0083] The weight of each independent feature is adjusted based on its historical performance in distinguishing real biological response signals from specific interference signals and the signal-to-noise ratio of the current signal waveform.

[0084] Specifically, weighting each independent feature among multiple independent features refers to dynamically assigning different weights based on each feature's ability to distinguish genuine biological response signals from specific interference signals in historical data, as well as the signal-to-noise ratio (SNR) of the current signal waveform. For example, features exhibiting higher sensitivity and specificity in distinguishing biological responses from specific optical interference can be assigned higher weights; while features with low SNR in the current signal waveform can have their weights appropriately reduced to minimize the impact of noise on the interpretation results. This weight adjustment aims to optimize the contribution of each feature to the final interpretation result, making it more accurately reflect the true state of the biological response.

[0085] S302. Calculate the aggregate score of multiple independent features after weighting.

[0086] Specifically, the weighted individual feature values ​​can be combined to obtain a single score that represents the overall feature information. This aggregated score can be obtained using linear weighted summation, nonlinear combination, or other multivariate analysis methods. The purpose is to integrate scattered feature information into a unified interpretation index, facilitating subsequent comparison of interpretation regions.

[0087] S303. Compare the aggregated score with multiple preset interpretation regions to determine whether the aggregated score falls within the fuzzy undetermined region.

[0088] These interpretation regions typically include clearly defined positive regions, clearly defined negative regions, and one or more ambiguous undetermined regions. Ambiguous undetermined regions refer to intervals where a clear interpretation result cannot be directly given when the aggregated score falls within them; these usually correspond to situations where the signal strength is close to the interpretation threshold or where complex interference exists.

[0089] S304. When the aggregated score falls into the fuzzy undetermined region, a cross-validation mechanism based on the nonlinear relationship between features is initiated to identify highly overlapping signal patterns among the features.

[0090] Specifically, this cross-validation mechanism aims to deeply analyze signals in a state of ambiguity by examining whether there are nonlinear correlations or interactions between different features, thereby revealing complex signal patterns that may be overlooked by simple linear models. For example, machine learning algorithms (such as support vector machines and neural networks) can be used to learn and identify nonlinear relationships between features, thus more accurately determining the true attributes of the signal. Highly overlapping signal patterns refer to situations where biological response signals and interference signals exhibit similarity in certain feature dimensions, making them difficult to distinguish.

[0091] S305. Based on the results of cross-validation, adjust the aggregation score to obtain a consistency evaluation result.

[0092] For example, if the cross-validation mechanism identifies a signal within a vague region of uncertainty as actually exhibiting a specific biological response pattern, the aggregate score will be adjusted upwards to favor a positive interpretation; conversely, if it identifies a interference pattern, the aggregate score will be adjusted downwards. This adjustment improves the accuracy and reliability of the final consensus assessment results.

[0093] This application's solution introduces dynamic weighting processing, allowing the contribution of each independent feature to the final interpretation result to be optimized based on its historical performance and current signal quality. This effectively reduces the negative impact of low-quality or easily interfered features on the overall interpretation. By calculating the weighted aggregation score, the information from multiple independent features is integrated into a more representative comprehensive index, avoiding the limitations of single-feature interpretation. When the aggregation score falls within the ambiguous region, traditional simple interpretation methods may struggle. This application further initiates a cross-validation mechanism based on the nonlinear relationship between features. This mechanism can deeply explore the potential complex correlations between features, identify highly overlapping signal patterns, and thus provide more refined discrimination capabilities even when the signal is unclear. The aggregation score is corrected through the cross-validation results, ensuring that the final consistency evaluation result fully utilizes multi-dimensional and deep-level signal information, effectively overcoming the limitations of traditional methods that are susceptible to interference and inaccurate in interpretation within the critical interpretation interval.

[0094] Through the above technical solutions, this application can significantly improve the accuracy and reliability of syphilis non-specific antibody detection within the critical interpretation range. Dynamic weighting processing enables the system to adaptively adjust the importance of different features, effectively suppressing the influence of noise and interference. The introduction of aggregated scores simplifies the complexity of multi-feature interpretation and provides a comprehensive interpretation basis. Especially when the initial interpretation result is within the critical range, the activation of a cross-validation mechanism based on the nonlinear relationship between features can identify complex signal patterns that are difficult to distinguish using traditional methods, thereby avoiding misjudgments and missed judgments. Therefore, the solution of this application can provide more accurate and robust interpretation results when facing low-concentration samples, high background interference, or complex biological reaction dynamics, greatly enhancing the clinical application value and diagnostic efficacy of the detection method.

[0095] In some preferred embodiments, assuming that the preliminary interpretation result of a test sample shows that its aggregate score falls within the critical interpretation interval, for example, the score is between the positive and negative thresholds, and a clear conclusion cannot be given directly, the system will activate the consistency evaluation mechanism proposed in this application. First, multiple independent features are weighted. For example, if the "signal rise slope" feature shows high discriminative power in distinguishing true positive reactions and non-specific adsorption in historical data, and the current signal-to-noise ratio of the signal waveform is high, its weight will be increased; while if the "background fluorescence intensity" feature has a low signal-to-noise ratio, its weight will be decreased. Next, the weighted aggregate score is calculated. If the aggregate score still falls within the ambiguous undetermined region, the system will activate a cross-validation mechanism based on the nonlinear relationship between features. Specifically, a pre-trained neural network model can be used, inputting multiple independent features (such as signal rise slope, plateau duration, maximum signal intensity, background fluctuation frequency, etc.), and the model can learn the nonlinear combination pattern between these features. For example, if the neural network identifies a high match between the current feature combination pattern and a known high-interference pattern, even if the aggregate score is slightly biased towards positive, the cross-validation result will indicate negative or require further confirmation. The aggregate score is then adjusted based on the cross-validation result. For instance, if the cross-validation result strongly indicates negative, the aggregate score will be adjusted downwards, and the final interpretation may be changed to negative. Conversely, if the cross-validation result indicates positive, the aggregate score will be adjusted upwards, and the final interpretation may be changed to positive. In this way, even in critical situations where the signal is unclear, more accurate and reliable interpretation results can be obtained.

[0096] In one design, to determine the weight of each independent feature, this application further includes: S401. Continuously monitor the dynamic changes of biological response characteristics and interference characteristics in the signal waveform.

[0097] Specifically, continuous monitoring of the dynamic changes in biological response characteristics and interference characteristics in signal waveforms refers to the system's real-time analysis of signal waveforms to capture the changes over time in signal characteristics related to biological responses (such as signal strength, rate of change, peak position, etc.) and signal characteristics related to interference (such as noise level, specific frequency components, baseline drift, etc.). This monitoring can employ various signal processing techniques, such as Fourier transform, wavelet analysis, and moving average, to extract dynamic information across different time scales and frequency ranges.

[0098] In some embodiments, multiple optical detection channels can be set up, each channel is configured with a light source and detector of different wavelengths or different polarization states; the signals collected by each optical detection channel are preliminarily processed to extract their respective waveform features; the differences and correlations of signal features between different optical detection channels are analyzed to identify and separate the independent effects of different interfering substances on the signal; the separated interfering components are removed from the signal to monitor the dynamic changes of biological response features and interference features in the signal waveform.

[0099] The purpose of setting up multiple optical detection channels is to acquire optical response information of the sample from different dimensions. Each optical detection channel can be configured with a light source and detector of different wavelengths or polarization states. For example, light sources of different wavelengths such as visible light and near-infrared light, or light sources of different polarization states such as linearly polarized light and circularly polarized light, can be used to detect the sample's response under different optical conditions. Through this multi-channel, multi-dimensional data acquisition method, the biological response signals and potential interference signals of the sample can be captured more comprehensively.

[0100] Furthermore, the signals acquired by each optical detection channel undergo preliminary processing to remove noise, calibrate the baseline, and extract their respective waveform characteristics. These waveform characteristics may include, but are not limited to, signal amplitude, rise time, fall time, peak value, integral area, and frequency components. These characteristics can reflect the dynamics of biological reactions and the characteristics of interference signals.

[0101] Because different interfering substances (such as turbidity, color, bubbles, and non-specific adsorption in a sample) may exhibit varying optical response characteristics at different wavelengths or polarization states, while real biological reaction signals may show stronger consistency or specific correlation patterns across different channels, comparing and analyzing these differences and correlations can effectively identify and separate the independent effects of different interfering substances on the signal. For example, some interferences may only be significant at specific wavelengths, while biological reaction signals are present at multiple wavelengths.

[0102] For example, by comparing signals from different wavelength channels, it is possible to distinguish between scattering interference caused by sample turbidity and absorbance changes caused by specific binding. Once the interfering components are identified and separated, they can be effectively removed from the original signal, resulting in purer biological response characteristics. This method enhances the ability to capture biological response signals at both the physical and signal processing levels, effectively suppresses various potential interferences, and ensures the accuracy of subsequent feature extraction.

[0103] S402. When the first or second condition is met, a slight change is identified in the biological reaction kinetics or interference characteristics of the sample to be tested.

[0104] The first condition is that the upward trend of the biological reaction signal, the duration of the plateau period, or the stability of the response to mechanical stirring deviates from the preset reference template. For example, the rate of increase of the biological reaction signal is significantly slower or faster than the standard curve, or the time required to reach the plateau period is significantly prolonged or shortened, or the fluctuation of the signal or the recovery speed is different from the normal situation when mechanical stirring is performed.

[0105] The second condition is that the frequency components and amplitude variation patterns of the interference signal find a new match with the preset interference mode.

[0106] For example, within a specific frequency band, the energy of the interfering signal suddenly increases, or its amplitude variation exhibits a new periodicity or randomness similar to known interference patterns. These deviations or new matches indicate that there may be subtle, unconsidered changes within the sample or the detection environment.

[0107] S403. In response to identified changes, dynamically adjust the weight of each individual feature.

[0108] Specifically, if a slow upward trend in a biological response is identified, the weight of later-stage response features should be increased to compensate for the weakness of earlier signals. If an increase in the frequency components of a specific interference signal is detected, the weight of features that are more affected by interference in that frequency band is reduced.

[0109] It should be noted that when the system detects a significantly slower upward trend in the biological response signal of the sample compared to the preset reference template, this may indicate low biological activity or abnormal reaction kinetics. In this case, to ensure accurate capture of weak or delayed biological response signals, the weighting of later-stage features can be increased. For example, the contribution of later-stage features such as the duration of the signal plateau, the time of reaching maximum signal intensity, and the signal decay rate can be increased. In this way, even if the early signals are not obvious, the effective information accumulated later can be fully utilized, thereby avoiding missed detections.

[0110] It should be noted that when the system identifies an enhancement in the frequency components of a specific interference signal through frequency analysis or pattern matching—such as noise in a specific frequency band caused by mechanical vibration, power supply noise, or ambient light fluctuations—this interference may significantly affect certain features that depend on signals in that frequency band. To reduce the negative impact of this interference on the interpretation results, the weight of features more affected by interference in that frequency band can be reduced. For example, if a feature is mainly extracted based on high-frequency signals, and high-frequency interference is enhanced, the weight of that feature can be appropriately reduced, and the interpretation can instead rely on other features less affected by interference in that frequency band.

[0111] In one example, assuming a non-specific antibody test for syphilis is performed, the system first identifies a sample whose biological response signal rise is significantly slower than the normal reference curve by continuously monitoring the dynamic changes in biological response and interference features within the signal waveform. For instance, the signal intensity should typically reach 50% of its maximum value within 5 minutes of the reaction starting, but this sample only reaches that value after 10 minutes. In response to this slower rise in biological response, the system dynamically adjusts the feature weights. Specifically, the system increases the weights on later-reaction features such as the duration of the signal plateau, the maximum signal intensity, and the signal decay rate, while appropriately decreasing the weights on earlier features such as the initial rise slope of the signal.

[0112] Furthermore, suppose that in another detection, the system detects a significant enhancement of frequency components in a specific high-frequency band (e.g., 50-60Hz) of the signal waveform, which is identified as a specific interference signal caused by external power supply noise. In response to this change in the "enhancement of frequency components of the specific interference signal," the system dynamically adjusts the feature weights. Specifically, if a particular feature (e.g., a fluctuation feature based on a high-frequency signal) is particularly sensitive to interference in that frequency band, the system will decrease the weight of that feature while increasing the weights of other features less affected by interference in that frequency band or features based on low-frequency signals (e.g., features based on signal integral area or low-frequency trends). In this way, even in the presence of specific interference, the system can ensure the accuracy of the interpretation results by optimizing the feature combination.

[0113] This application's solution addresses the limitations of traditional weighting methods by introducing continuous dynamic monitoring of biological response characteristics and interference characteristics, and dynamically adjusting the weights of independent features based on identified minute changes. The ability to perceive and quantify subtle changes in biological response kinetics or interference characteristics in real time enables the system to more flexibly and accurately assess the reliability and discriminative power of each independent feature under specific detection conditions. This dynamic adjustment mechanism ensures that the interpretation logic adaptively optimizes in the face of fluctuations in inherent sample characteristics, changes in environmental interference, or abnormal biological responses, thereby avoiding misjudgments or missed judgments caused by fixed weights or insufficient weighting adjustments based on historical data.

[0114] Through the above technical solution, this application can significantly improve the robustness and accuracy of the signal pattern recognition method for syphilis non-specific antibody detection. Especially when facing test samples with complex background interference or atypical biological reaction dynamics, this method can effectively reduce the influence of interference signals and more accurately capture the true biological reaction signals by dynamically adjusting feature weights, thereby obtaining more reliable interpretation results. This adaptive weight adjustment mechanism enables the system to maintain high performance in various practical application scenarios, reduces reliance on manual intervention, and improves the level of automation in detection.

[0115] In some preferred embodiments, a specific example is given below. Suppose that during a non-specific antibody test for syphilis, the system continuously monitors the signal waveform of the sample to be tested. At a certain point in time, the system detects that the upward trend of the biological reaction signal is significantly slower than that of a preset reference template. This is judged to satisfy a first condition, indicating a slight change in the biological reaction kinetics. Simultaneously, the system also detects a sudden increase in the noise amplitude in a specific frequency band of the signal waveform. This is judged to satisfy a second condition, indicating the presence of a new interference pattern.

[0116] In response to these identified changes, the system dynamically adjusts the weights of individual features. Specifically, as the upward trend of the biological response slows, the system reduces the weight of early response rate features and increases the weight of signal strength features during the later plateau phase of the response. This ensures that even with weaker early signals, the biological response can be accurately assessed through more stable later signals. Simultaneously, due to increased interference in specific frequency bands, the system reduces the weights of features susceptible to interference in those bands, such as certain features based on high-frequency components, thereby minimizing the impact of interference on the final aggregate score. Through this dynamic adjustment, even in cases of abnormal biological responses or the presence of novel interferences, the system can still generate a more accurate and representative aggregate score, thus improving the reliability of the interpretation results.

[0117] In some embodiments of this application, the dynamic changes of biological response characteristics and interference characteristics in the signal waveform are continuously monitored, and the biological response dynamics or interference characteristics of the sample under test are identified as having undergone minor changes based on a first condition or a second condition. However, in practical applications, due to the complexity of the signal, the randomness of noise, and the superposition of multiple interference factors, a single monitoring and judgment mechanism may lead to misjudgment or uncertainty regarding the satisfaction of the first or second condition, especially when the signal characteristics are in a critical state or when there is multi-source interference. This may result in contradictory or inconsistent evaluation results, thus affecting the accurate adjustment of subsequent feature weights. If the above problems are not addressed, the system may not be robust enough in identifying sample changes, thereby affecting the reliability of the final interpretation result. To address this, this application further proposes a more refined and robust judgment mechanism, which uses multi-module parallel analysis and signal source tracing to more accurately determine whether the first or second condition is met.

[0118] S501: Set up multiple signal analysis modules, each with specific filtering parameters and time window parameters.

[0119] Here, "specific filtering parameters and time window parameters" refers to the fact that each module is designed to analyze signals from different angles or with a focus on different signal characteristics. For example, one module might be configured with a low-pass filter and a longer time window to capture the slow trends of biological responses, while another module might be configured with a band-pass filter and a shorter time window to detect interference signals at specific frequencies. This diverse parameter configuration aims to comprehensively capture various characteristics of the signal and improve sensitivity to different types of variations.

[0120] S502 processes signal waveforms in parallel to obtain the output of multiple modules.

[0121] Specifically, the original signal waveform is simultaneously input into the aforementioned multiple signal analysis modules, and each module independently performs its preset analysis task.

[0122] The outputs of multiple modules refer to the evaluation results generated by each module after completing its analysis, which describe the characteristics or state of the signal. These outputs can be quantitative indicators or judgments regarding the upward trend of biological response signals, the duration of plateau phases, the stability of responses to mechanical stirring, and the frequency components or amplitude variations of interference signals.

[0123] S503. Evaluate the output of each module independently to determine whether the first or second condition is met.

[0124] S504. When the evaluation results of multiple modules are contradictory or inconsistent, the signal source tracing mechanism shall be activated.

[0125] This mechanism is an advanced diagnostic and analytical process designed to delve into the source, composition, and reasons for discrepancies in signal performance across different modules when discrepancies arise in assessments. The mechanism aims to identify the root cause of the discrepancy, such as specific types of interference, signal superposition, sensor drift, or other unknown factors.

[0126] Specifically, the signal source tracing mechanism can be activated by following these steps: 1. Classify and identify contradictory or inconsistent patterns in the evaluation results of multiple modules to obtain classification and identification results; Specifically, classifying and identifying contradictory or inconsistent patterns in the evaluation results of multiple modules involves using a pre-trained pattern recognition algorithm to analyze the specific manifestations of conflicts between the outputs of different modules. These conflicts may manifest as inconsistencies in amplitude, frequency, phase, or time series, as well as the intensity and duration of these inconsistencies. This allows for the categorization of these contradictory patterns into known types of obfuscation, such as "biological signals being masked by interference at specific frequencies," "multiple interference superposition," or "signal drift being confused with biological responses." The aim is to provide targeted guidance for subsequent signal decoupling.

[0127] 2. For the confusion pattern indicated by the classification and recognition results, activate a dedicated signal decoupling strategy to decouple the signal waveform and obtain the decoupled signal components. Specifically, based on the identified obfuscation pattern, the most suitable signal separation algorithm can be selected from a pre-defined strategy library or dynamically generated. For example, if the obfuscation is between high-frequency noise and biological signals, a high-pass filtering or wavelet denoising strategy is activated; if the obfuscation is between baseline drift and slow biological responses, a baseline correction combined with an adaptive filtering strategy is activated. The aim is to ensure the targeted and efficient decoupling process.

[0128] 3. During the signal decoupling process, monitor the purity of each signal component after decoupling to obtain purity feedback; Specifically, the decoupling effect can be quantified by real-time calculation of metrics such as the signal-to-noise ratio, spectral separation, or similarity to a known pure signal template between the decoupled biological response signal and the interference signal. For example, blind source separation (BSS) technology can be used, and the independence of each separated component can be continuously evaluated during the separation process. The purpose is to provide real-time, quantitative data for adjusting the parameters of the decoupling strategy.

[0129] 4. Adjust the decoupling parameters of the signal decoupling strategy based on purity feedback.

[0130] Specifically, when the detected purity index fails to reach a preset threshold, the system automatically optimizes the parameters of the current decoupling strategy, such as adjusting the filter cutoff frequency, wavelet basis function, number of iterations, or coefficients of the separation matrix, in order to achieve a better decoupling effect. The goal is to achieve adaptive optimization of the decoupling process, ensuring that the final separated signal components are as pure as possible.

[0131] 5. Based on the adjusted decoupling parameters, the signal waveform is decoupled to obtain a pure biological response signal and interference signal.

[0132] Based on the pure biological response signal and the interference signal, determine whether the first or second condition is met.

[0133] Understandably, during the decoupling process, by monitoring the purity of each signal component after decoupling in real time and providing feedback, the parameters of the decoupling strategy can be adaptively adjusted and optimized, thus ensuring the high purity of the finally separated biological reaction signal and interference signal. It is precisely because of the obtained pure biological reaction signal and interference signal that subsequent judgments on the first or second condition can be based on more reliable and accurate data, thereby significantly improving the accuracy of identifying minute changes in biological reaction kinetics or interference characteristics.

[0134] In some preferred embodiments, a specific example is given below. Suppose that during the detection of nonspecific antibodies against syphilis, the evaluation results of two signal analysis modules are contradictory: module A indicates that the upward trend of the biological reaction signal is normal, but module B shows abnormal high-frequency noise. This makes it difficult for the system to determine whether the abnormality is due to the biological reaction itself or external interference.

[0135] At this point, the source tracing mechanism of this application is activated. First, the system classifies and identifies contradictory patterns in the evaluation results of module A and module B. Through analysis, the system identifies this as a typical obfuscation pattern where "biological signals are masked by high-frequency noise."

[0136] To address this obfuscation pattern, the system activates a dedicated signal decoupling strategy, such as an adaptive wavelet denoising algorithm. During signal decoupling, the system continuously monitors the purity of the decoupled biological response signal and noise signal. If the initial denoising parameters lead to distortion of the biological signal or incomplete noise removal, the purity feedback mechanism instructs the system to adjust the wavelet basis function, the number of decomposition layers, or the threshold parameters.

[0137] After several rounds of iteration and parameter adjustments, the system finally obtained a pure biological response signal and a pure interference signal. Based on this pure biological response signal, the system re-evaluated its upward trend, plateau duration, and other characteristics, confirming that it did not deviate from the preset reference template, i.e., it did not meet the first condition. At the same time, based on the pure interference signal, the system confirmed the existence of enhanced high-frequency noise in a specific frequency band, and that its frequency components and amplitude variation patterns matched the preset interference mode, i.e., it met the second condition.

[0138] Therefore, the system can accurately determine that the biological reaction kinetics of the sample under test are normal, and that the contradiction in the evaluation results is due to the presence of specific high-frequency interference signals, thus avoiding misjudgment of the biological reaction.

[0139] S505. Based on the signal tracing results, determine whether the first condition or the second condition is met.

[0140] Specifically, detailed information about signal composition, interference type, noise characteristics, and the reasons for differences in assessments of each module can be obtained through the signal source tracing mechanism. Based on the deeper and more comprehensive information provided by the tracing mechanism, a final and more accurate ruling can be made on previously contradictory judgments. This ensures higher reliability and accuracy in identifying changes in biological response dynamics or interference characteristics in complex or ambiguous situations.

[0141] This application's solution effectively addresses the potential for misjudgment and uncertainty in complex signal environments through the introduction of multi-module parallel analysis and a signal source tracing mechanism. Specifically, multiple signal analysis modules are set up with different filtering and time window parameters, enabling the system to capture biological response and interference characteristics in signal waveforms from multiple dimensions and with varying sensitivities. This parallel processing approach ensures comprehensive signal coverage, avoiding the omission of key information or insensitivity to specific interference due to improper single parameter settings. When the evaluation results of different modules contradict or are inconsistent, it indicates the possible presence of complex superposition effects, novel interferences, or critical states in the signal, at which point the signal source tracing mechanism is activated. This mechanism can deeply analyze the signal components causing the contradictions, such as by decoupling signals from different sources, identifying noise patterns, or analyzing the energy distribution of signals in different frequency bands, thereby revealing the true composition and causes of signal changes. By tracing the signal source, the system can obtain purer and more accurate biological response and interference signal information than a single module, and based on this verified and corrected information, more reliably determine whether the first or second condition is met.

[0142] Through the above technical solutions, this application can significantly improve the accuracy and robustness of identifying minute changes in biological reaction dynamics or interference characteristics in complex signal environments. The multi-module parallel analysis mechanism ensures comprehensive capture of signal features, reducing the risk of misjudgment that may arise from a single analysis method. When contradictory evaluation results occur, the introduction of a signal source tracing mechanism allows the system to deeply analyze the intrinsic composition of the signal and the source of interference, effectively solving the problem of traditional methods' difficulty in accurately judging under critical or complex conditions. Therefore, this application can more accurately identify the true changes in samples, avoiding improper adjustment of feature weights due to misjudgment, thereby improving the overall reliability of syphilis non-specific antibody detection signal pattern recognition and the accuracy of interpretation results.

[0143] like Figure 3 As shown in the figure, this invention also provides a syphilis non-specific antibody detection signal pattern recognition system. The system includes: The information acquisition module is used to acquire interference indication information of the sample under test, which reflects the inherent optical characteristics of the sample under test. The parameter adjustment module is used to adjust the parameters for feature extraction of the signal waveform of the sample under test according to the interference indication information. The feature extraction module is used to extract features from the signal waveform based on the adjusted parameters; The interpretation logic adjustment module is used to adjust the interpretation logic based on the interference indication information and the extracted features in order to obtain the interpretation result of the sample to be tested.

[0144] This application also provides a computer-readable storage medium. All or part of the processes in the above method embodiments can be executed by a computer program instructing related hardware. This program can be stored in the computer-readable storage medium, and when executed, it can include the processes of the above method embodiments. The computer-readable storage medium can be an internal storage unit of the task execution device (including a data sending end and / or a data receiving end) of any of the foregoing embodiments, such as the hard disk or memory of the task execution device. The computer-readable storage medium can also be an external storage device of the terminal device, such as a plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, etc., equipped on the terminal device. Further, the computer-readable storage medium can include both the internal storage unit of the task execution device and an external storage device. The computer-readable storage medium is used to store the computer program and other programs and data required by the task execution device. The computer-readable storage medium can also be used to temporarily store data that has been output or will be output.

[0145] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0146] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a readable storage medium. Based on this understanding, the technical solutions of the embodiments of this application, essentially, or the parts that contribute to the prior art, or all or part of the technical solutions, can be embodied in the form of a software product. This software product is stored in a storage medium and includes several instructions to cause a device (which may be a microcontroller, chip, etc.) or processor to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, ROM, RAM, magnetic disks, or optical disks.

[0147] The above are merely specific embodiments of this application, but the scope of protection of this application is not limited thereto. Any changes or substitutions within the technical scope disclosed in this application should be covered within the scope of protection of this application.

Claims

1. A method for recognizing patterns in non-specific antibody detection signals for syphilis, characterized in that, include: Obtain interference indication information of the sample under test, wherein the interference indication information reflects the inherent optical characteristics of the sample under test; Based on the interference indication information, adjust the parameters for feature extraction of the signal waveform of the sample under test; Based on the adjusted parameters, features are extracted from the signal waveform; Based on the interference indication information and the extracted features, the interpretation logic is adjusted to obtain the interpretation result of the sample to be tested; The step of adjusting the parameters for feature extraction of the signal waveform of the sample under test based on the interference indication information includes: Continuously monitor the local variation trend and frequency components of the signal waveform; When the local change trend of the signal waveform is detected to match the preset interference mode, or when the frequency components of the signal waveform are enhanced in a specific frequency band, dynamic interference is determined to exist. In response to the determination of the presence of dynamic interference, the cutoff frequency of the digital filter and the time window length of the waveform analysis are adjusted in real time according to the intensity and frequency characteristics of the dynamic interference to obtain the adjusted parameters. The step of adjusting the interpretation logic based on the interference indication information and extracted features to obtain the interpretation result of the sample to be tested includes: Based on the interference indication information and the extracted features, a preliminary judgment is made to obtain a preliminary judgment result; When the preliminary judgment result is within the preset critical judgment interval, a consistency evaluation is performed on multiple independent features to obtain a consistency evaluation result; Based on the consistency assessment results, the interpretation logic is adjusted to obtain the interpretation results of the sample to be tested; When the preliminary judgment result is within a preset critical judgment interval, a consistency evaluation is performed on multiple independent features, including: Each of the plurality of independent features is weighted, and the weight of each independent feature is adjusted based on the feature’s historical performance in distinguishing real biological response signals from specific interference signals and the signal-to-noise ratio of the current signal waveform. Calculate the weighted aggregate score of the multiple independent features; The aggregated score is compared with multiple preset interpretation regions to determine whether the aggregated score falls within a fuzzy undetermined region. When the aggregated score falls into the fuzzy undetermined region, a cross-validation mechanism based on the nonlinear relationship between features is initiated to identify highly overlapping signal patterns among the features; Based on the results of cross-validation, the aggregation score is adjusted to obtain the consistency evaluation result; Determining the weight of each independent feature includes: Continuously monitor the dynamic changes of biological response characteristics and interference characteristics in the signal waveform; When either the first or the second condition is met, a slight change is identified in the bio-reaction kinetics or interference characteristics of the sample to be tested; the first condition is that the upward trend, plateau duration, or response stability to mechanical stirring of the bio-reaction signal deviates from a preset reference template; the second condition is that the frequency components and amplitude variation patterns of the interference signal show a new match with a preset interference pattern. In response to the identified changes, the weight of each individual feature is dynamically adjusted.

2. The method for recognizing syphilis non-specific antibody detection signal patterns according to claim 1, characterized in that, The continuous monitoring of the dynamic changes in biological response characteristics and interference characteristics in the signal waveform includes: Multiple optical detection channels are set up, and each channel is equipped with a light source and detector of different wavelengths or different polarization states; The signals acquired by each optical detection channel are preliminarily processed to extract their respective waveform features; The differences and correlations of signal characteristics between different optical detection channels are analyzed to identify and separate the independent effects of different interfering substances on the signal. The separated interference components are removed from the signal to monitor the dynamic changes of biological response characteristics and interference characteristics in the signal waveform.

3. The method for recognizing syphilis non-specific antibody detection signal patterns according to claim 1, characterized in that, The method further includes: Set up multiple signal analysis modules, each with specific filtering parameters and time window parameters; Parallel processing of signal waveforms yields the outputs of multiple modules; The output of each module is evaluated independently to determine whether the first condition or the second condition is met; When the evaluation results of multiple modules are contradictory or inconsistent, the signal source tracing mechanism is activated. Based on the source tracing results of the signal, determine whether the first condition or the second condition is met.

4. The method for recognizing syphilis nonspecific antibody detection signal patterns according to claim 3, characterized in that, When the evaluation results of multiple modules are contradictory or inconsistent, the signal source tracing mechanism is activated, including: The contradictory or inconsistent patterns of the evaluation results of the multiple modules are classified and identified to obtain the classification and identification results; For the confusion pattern indicated by the classification and recognition results, a dedicated signal decoupling strategy is activated to decouple the signal waveform and obtain the decoupled signal components. During the signal decoupling process, the purity of each signal component after decoupling is monitored to obtain purity feedback; Based on the purity feedback, adjust the decoupling parameters of the signal decoupling strategy; Based on the adjusted decoupling parameters, the signal waveform is decoupled to obtain a pure biological reaction signal and interference signal; Based on the pure biological response signal and the interference signal, determine whether the first condition or the second condition is met.

5. The method for recognizing syphilis non-specific antibody detection signal patterns according to claim 1, characterized in that, The dynamic adjustment of the weight of each individual feature in response to the identified change includes: If a slow upward trend in a biological response is identified, the weight of later-stage response features is increased to compensate for the weakness of earlier signals. If an increase in the frequency components of a specific interference signal is detected, the weight of features that are more affected by interference in that frequency band is reduced.

6. A syphilis non-specific antibody detection signal pattern recognition system, characterized in that, The system includes: An information acquisition module is used to acquire interference indication information of the sample under test, wherein the interference indication information reflects the inherent optical characteristics of the sample under test; The parameter adjustment module is used to adjust the parameters for feature extraction of the signal waveform of the sample under test according to the interference indication information. The feature extraction module is used to extract features from the signal waveform based on the adjusted parameters; The interpretation logic adjustment module is used to adjust the interpretation logic based on the interference indication information and the extracted features to obtain the interpretation result of the sample to be tested. The step of adjusting the parameters for feature extraction of the signal waveform of the sample under test based on the interference indication information includes: Continuously monitor the local variation trend and frequency components of the signal waveform; When the local change trend of the signal waveform is detected to match the preset interference mode, or when the frequency components of the signal waveform are enhanced in a specific frequency band, dynamic interference is determined to exist. In response to the determination of the presence of dynamic interference, the cutoff frequency of the digital filter and the time window length of the waveform analysis are adjusted in real time according to the intensity and frequency characteristics of the dynamic interference to obtain the adjusted parameters. The step of adjusting the interpretation logic based on the interference indication information and extracted features to obtain the interpretation result of the sample to be tested includes: Based on the interference indication information and the extracted features, a preliminary judgment is made to obtain a preliminary judgment result; When the preliminary judgment result is within the preset critical judgment interval, a consistency evaluation is performed on multiple independent features to obtain a consistency evaluation result; Based on the consistency assessment results, the interpretation logic is adjusted to obtain the interpretation results of the sample to be tested; When the preliminary judgment result is within a preset critical judgment interval, a consistency evaluation is performed on multiple independent features, including: Each of the plurality of independent features is weighted, and the weight of each independent feature is adjusted based on the feature’s historical performance in distinguishing real biological response signals from specific interference signals and the signal-to-noise ratio of the current signal waveform. Calculate the weighted aggregate score of the multiple independent features; The aggregated score is compared with multiple preset interpretation regions to determine whether the aggregated score falls within a fuzzy undetermined region. When the aggregated score falls into the fuzzy undetermined region, a cross-validation mechanism based on the nonlinear relationship between features is initiated to identify highly overlapping signal patterns among the features; Based on the results of cross-validation, the aggregation score is adjusted to obtain the consistency evaluation result; Determining the weight of each independent feature includes: Continuously monitor the dynamic changes of biological response characteristics and interference characteristics in the signal waveform; When either the first or the second condition is met, a slight change is identified in the bio-reaction kinetics or interference characteristics of the sample to be tested; the first condition is that the upward trend, plateau duration, or response stability to mechanical stirring of the bio-reaction signal deviates from a preset reference template; the second condition is that the frequency components and amplitude variation patterns of the interference signal show a new match with a preset interference pattern. In response to the identified changes, the weight of each individual feature is dynamically adjusted.

Images