Nested verification method and device based on multi-modal signals, equipment and storage medium

Abstract

The application provides a nested verification method and device based on multi-modal signals, equipment and a storage medium. The method comprises the following steps: collecting a first multi-modal physiological signal of an acupoint of a current batch of subjects before treatment and a second multi-modal physiological signal of the acupoint after treatment, and generating a treatment information set; selecting a frequency with the smallest signal change in the treatment information set as a core frequency, calculating a treatment effect of the current batch of subjects according to a signal representation value under the core frequency; obtaining a signal change amplitude of a target acupoint under the core frequency, and calculating a correlation coefficient between the signal change amplitude and a clinical symptom improvement degree; and generating an acupoint state judgment result of the current batch of subjects after treatment according to the signal change amplitude, the treatment effect and the correlation coefficient. The application can provide an effective systematic method for verifying the nested relationship among the acupoint, the meridian and the zang-fu organ.

Description

Technical Field

[0001] This invention relates to the field of biomedical engineering technology, and in particular to a nested verification method, apparatus, device, and storage medium based on multimodal signals. Background Technology

[0002] In Traditional Chinese Medicine (TCM) theory, the meridian system is the core network structure that circulates Qi and blood, connects the internal organs, communicates between the interior and exterior, and runs through the upper and lower body. Its function relies on a multi-layered and dynamic nested relationship between acupoints, meridians, and internal organs. Acupoints are special reaction points of Qi and blood on the body surface, reflecting the state of the internal organs and serving as targets for external therapies; meridians are channels for information and energy transmission; and the internal organs are the core of functional activities. These three elements form a nested regulatory mechanism of "low-frequency stimulation (acupoints) – mid-frequency transmission (meridians) – high-frequency response (internal organs)." This mechanism helps to deepen the understanding of the regulatory processes of human physiological functions, provides a theoretical basis for TCM treatment, and promotes the circulation of Qi and blood, the coordination of internal organ functions, and the maintenance of overall health.

[0003] Sensitized acupoints refer to acupoints whose biological characteristics change under disease conditions, making them more sensitive and active compared to their normal state. Sensitized acupoints can serve as an important reference for disease diagnosis. By detecting the sensitization of acupoints and combining this information with the patient's symptoms and signs, doctors can more accurately determine the location, nature, and severity of the disease. For example, in some visceral diseases, corresponding acupoints on the body surface exhibit sensitization; detecting and analyzing these sensitized acupoints can aid in the diagnosis of visceral diseases. However, current TCM digital diagnostic and treatment equipment has significant shortcomings in verifying the nested relationship between acupoints, meridians, and internal organs. Most meridian detectors use a single frequency (e.g., 50kHz) for impedance measurement, with an error rate as high as ±5%, making it difficult to capture the impedance characteristics of tissues at different depths. This limits the spatial exploration of the "three-dimensional structure" characteristics of acupoints and lacks the ability to simultaneously acquire multimodal signals. Furthermore, they often rely on subjective experience to judge the functional state of acupoints, failing to systematically construct verification models. The biological nested modular structure theory proposed by GoekoopJG, KoperJW, and SmeetsTJ in their 2019 paper "Hierarchical network structure as the source of hierarchical dynamics in biological systems" published in Volume 10 of Frontiers in Physiology provides theoretical support for the "low-frequency stimulus triggers high-frequency response" theory. However, empirical research on the meridian system in traditional Chinese medicine lacks multimodal data support and systematic algorithm implementation, making it impossible to correlate acupoint sensitization with the functional state of internal organs, lacking early disease identification and prediction capabilities, and failing to form a standardized and scalable diagnostic and treatment evaluation system. Summary of the Invention

[0004] This invention aims to provide a nested verification method, apparatus, device, and storage medium based on multimodal signals to solve the aforementioned technical problems. It introduces the concept of cross-scale verification, using measurable multimodal physical signals from the body surface, specifically impedance and temperature signals, to construct a quantitative model. This enables objective verification of the cross-scale nested relationship and resonance conduction mechanism between "acupoints (body surface scale) - meridians (body surface channel scale) - viscera (internal deep scale)". It reveals whether a specific correspondence exists between acupoints and viscera, and the changing patterns of this correspondence under different physiological and pathological states. This provides a practical, scientific, and objective systematic method for verifying the nested relationship between "acupoints, meridians, and viscera".

[0005] To address the aforementioned technical problems, this invention provides a nested verification method based on multimodal signals, comprising: The first multimodal physiological signal of acupoints in the current batch of subjects before treatment and the second multimodal physiological signal after treatment were collected; among them, the multimodal physiological signal includes acupoint impedance information, acupoint temperature information and meridian segment impedance information between acupoints at multiple preset frequencies; A treatment information set is generated based on the first and second multimodal physiological signals; wherein, the treatment information set includes the signal representation values ​​of the subject's acupoints at each preset frequency, and the magnitude of signal changes in the subject after the application of treatment; The frequency with the smallest signal change in the treatment information set is selected as the core frequency, and the treatment effect size of the current batch of subjects is calculated based on the signal characterization value at the core frequency; whereby the treatment effect size is used to quantify the magnitude of clinical symptom changes in the current batch of subjects after treatment. The signal variation amplitude of the target acupoint at the core frequency was obtained, and the correlation coefficient between the signal variation amplitude and the degree of improvement of clinical symptoms was calculated. The target acupoint is the acupoint associated with the internal organs, and the correlation coefficient is used to characterize the correlation between the changes in the multimodal physiological signals of the acupoint and the degree of improvement of clinical symptoms. The acupoint status assessment results after treatment are generated based on the signal change amplitude, therapeutic effect size, and correlation coefficient. The acupoint status assessment results are used to verify the correlation between acupoint sensitization status and organ function.

[0006] The above scheme collects multimodal physiological signals, including acupoint impedance, temperature, and meridian segment impedance information at multiple preset frequencies, before and after treatment. This multidimensional data collection method provides a rich and comprehensive information foundation for subsequent analysis. It also closely aligns with the concept of cross-scale validation, building an information bridge from acupoints at the body surface scale and meridians at the body surface channel scale to the internal organs at the deep internal scale. The resulting treatment information set covers signal representation values ​​at each preset frequency and the amplitude of signal changes after treatment, demonstrating the dynamic changes of acupoints during treatment in detail. Next, the frequency with the smallest signal change is selected as the core frequency to calculate the treatment effect. This approach eliminates interference factors during treatment, allowing the calculated effect to more accurately quantify the changes in the subjects' clinical symptoms, reflecting the intrinsic connection and resonance conduction mechanism between acupoints, meridians, and internal organs. Subsequently, by acquiring the signal change amplitude of the target acupoints at the core frequency and calculating its correlation coefficient with the degree of clinical symptom improvement, the intrinsic connection between multimodal physiological signal changes at acupoints and clinical symptom improvement is revealed. This step further deepens the connotation of cross-scale validation. The target acupoints are associated with the internal organs, and the amplitude of their signal changes at the core frequency reflects the cross-scale information transmission between the acupoints on the body surface and the deep internal organs. Finally, the acupoint status judgment result is generated by comprehensively considering the signal change amplitude, therapeutic effect, and correlation coefficient. This process organically combines various key indicators, scientifically and systematically verifying the correlation between acupoint sensitization state and organ function. It can provide objective scientific evidence for the theory of acupoints in traditional Chinese medicine and provide strong support for doctors to accurately judge the acupoint status and formulate personalized treatment plans in clinical practice. Based on the acupoint status judgment result, doctors can understand the specific relationship between acupoints, meridians, and internal organs, thereby accurately adjusting treatment strategies and improving the pertinence and effectiveness of treatment.

[0007] In one implementation, the first multimodal physiological signal of acupoints in the current batch of subjects before treatment and the second multimodal physiological signal after treatment are collected, specifically including: Scanning impedance measurements were performed at each preset frequency for a first preset time to obtain acupoint impedance information. Temperature information of acupoints is collected during scanning measurements to obtain acupoint temperature information. Scanning measurements were performed at each preset frequency to obtain the surface channel impedance between two acupoints on the same meridian, thus obtaining the meridian segment impedance information between acupoints. Obtain the acquisition timestamps of multimodal physiological signals and associate each subject ID with the corresponding multimodal physiological signal and the corresponding acquisition timestamp.

[0008] In the above scheme, regarding data acquisition, by performing scanning impedance measurements at each preset frequency and continuously acquiring acupoint impedance information for a first preset time, the impedance characteristics of acupoints at different frequencies can be captured comprehensively and meticulously. Simultaneously acquiring acupoint temperature information during the measurement process enables multi-dimensional data acquisition, providing richer information for subsequent analysis. Obtaining the surface channel impedance between two acupoints on the same meridian yields meridian segment impedance information between acupoints, further expanding the data dimensions and contributing to a deeper understanding of the physiological state of the meridians. Regarding data management, the acquisition timestamps of multimodal physiological signals are obtained, and each subject ID is associated with the corresponding multimodal physiological signal and acquisition timestamp, facilitating data traceability, organization, and analysis, ensuring data accuracy and traceability, and providing a solid and reliable data foundation for subsequent nested verification based on multimodal signals.

[0009] In one implementation, before generating the treatment information set based on the first and second multimodal physiological signals, the method further includes data preprocessing of the acquired multimodal physiological signals according to a preset preprocessing strategy. Specifically: Missing values ​​were detected in multimodal physiological signals, and outliers were detected using the interquartile range method. For multimodal physiological signal samples with missing values ​​and multimodal physiological signal samples marked as outliers, calculate their distances to complete multimodal physiological signal samples of the same type, and sort them in ascending order. A predetermined number of complete multimodal physiological signal samples with the highest ranking are selected. The selected complete multimodal physiological signal samples are then grouped and weighted according to the distance between the multimodal physiological signal samples and multimodal physiological signal samples with missing values ​​or those marked as outliers. Missing or outlier values ​​are imputed based on the weighted average of complete multimodal physiological signal samples.

[0010] In the above scheme, outlier detection using missing value detection and interquartile range (ICM) can accurately identify problematic data in multimodal physiological signals. For samples with missing values ​​or marked as outliers, their distance to complete samples of the same type is calculated and ranked, allowing for the selection of complete samples with high relevance. Selecting a predetermined number of top-ranked complete samples and grouping them with weights fully considers their proximity to problematic samples, resulting in a more reasonable weight allocation. Finally, the weighted sample average is used to imput missing or outlier values, effectively filling in missing information, correcting abnormal data, and improving the accuracy, completeness, and reliability of multimodal physiological signals. This provides a high-quality data foundation for subsequent generation of treatment information sets and the entire nested validation method, enhancing the credibility of subsequent analysis and validation results.

[0011] In one implementation, a treatment information set is generated based on a first multimodal physiological signal and a second multimodal physiological signal, specifically including: The average values ​​of the first and second multimodal physiological signals at each preset frequency are calculated respectively, and the average values ​​of the multimodal physiological signals are used as signal characterization values. Among them, the signal characterization values ​​include acupoint impedance information characterization values, acupoint temperature information characterization values, and meridian segment impedance information characterization values ​​between acupoints. The signal change amplitude at each preset frequency is calculated based on the signal characterization values ​​before and after treatment; the expression for the signal change amplitude is: ; In the formula, The amplitude of the signal change. The multimodal physiological signal characterization value at a single preset frequency before treatment is applied; The multimodal physiological signal characterization value at the same preset frequency after treatment is applied; A treatment information set is generated based on each signal characterization value and signal change amplitude of the current batch of subjects.

[0012] In the above scheme, the average values ​​of the first and second multimodal physiological signals at each preset frequency are calculated as signal characterization values, thereby comprehensively and accurately reflecting the characteristics of various aspects such as acupoint impedance, temperature, and meridian segment impedance between acupoints at different frequencies. By using a specific formula to calculate the signal change amplitude at each preset frequency before and after treatment, the degree of influence of treatment on multimodal physiological signals can be clearly quantified. A treatment information set is generated based on each signal characterization value and signal change amplitude, providing a systematic, detailed, and reliable data basis for further analysis of treatment effects, determination of core frequencies, and calculation of treatment effect magnitude, which helps to deeply explore the correlation between treatment and changes in multimodal physiological signals.

[0013] In one implementation, the frequency with the smallest signal change in the treatment information set is selected as the core frequency, and the treatment effect magnitude of the current batch of subjects is calculated based on the signal characterization value at the core frequency, specifically including: Calculate the variance of the signal variation amplitude at each preset frequency, and select the frequency with the smallest variance as the core frequency; The multimodal physiological signal characterization values ​​of each subject in the current batch before and after treatment at the core frequency are obtained one by one, and the difference between the two multimodal physiological signal characterization values ​​is calculated. The average value of the obtained difference is obtained by averaging the difference. The standard deviation of the acquired differences is calculated, and the treatment effect size of the current batch of subjects is calculated based on the mean of the signal representation and the standard deviation of the differences; wherein, the expression for the treatment effect size is: ; In the formula, For effect size, The average value represents the signal. This represents the standard deviation of the difference.

[0014] In the above scheme, by calculating the variance of signal variation amplitude at each preset frequency and selecting the frequency with the smallest variance as the core frequency, frequency interference with large signal fluctuations is effectively eliminated, making the core frequency more stable and representative, and accurately reflecting the true impact of treatment on multimodal physiological signals. The multimodal physiological signal characterization values ​​before and after treatment at the core frequency are obtained, and the average difference is taken to obtain the average signal characterization value, which can intuitively reflect the overall signal changes of the current batch of subjects at the core frequency. The standard deviation of the difference is calculated, and combined with the average signal characterization value to calculate the treatment effect size. This comprehensively considers the central tendency and dispersion of signal changes, making the calculation of the treatment effect size more scientific and accurate, and providing a reliable basis for quantifying the magnitude of clinical symptom changes after treatment in the current batch of subjects.

[0015] In one implementation, calculating the therapeutic effect size of the current batch of subjects based on the signal representation value at the core frequency further includes: performing a paired-samples t-test based on the signal representation value and standard deviation of the current batch of subjects at the core frequency, specifically: A null hypothesis is pre-defined; the null hypothesis is that the treatment received by the current batch of subjects is ineffective. The t-statistic for the current batch of subjects is calculated based on the signal representation value and standard deviation; the expression for the t-statistic is as follows: ; In the formula, For statistical purposes, This represents the number of subjects in the current batch. The corresponding P-value is obtained based on the t-statistic. When the P-value is less than the pre-set significance level, the null hypothesis is deemed invalid, and the changes in multimodal physiological signals before and after treatment are statistically significant.

[0016] In the above scheme, paired sample validation is carried out when calculating the treatment effect size. The null hypothesis of "treatment ineffectiveness" is set in advance. The t-statistic is calculated based on the signal characterization value and standard deviation at the core frequency and the p-value is obtained. When the p-value is less than the significance level, the null hypothesis is determined to be invalid. This method can scientifically, objectively and accurately assess whether the treatment is effective, provide quantitative statistical evidence for the treatment effect, avoid subjective judgment errors, and enhance the credibility and persuasiveness of the research results.

[0017] In one implementation, the signal variation amplitude of the target acupoint at the core frequency is obtained, and the correlation coefficient between the signal variation amplitude and the degree of improvement in clinical symptoms is calculated, specifically including: The symptoms of the current batch of subjects before and after treatment were scored using a clinically recognized assessment scale for visceral diseases, and the changes in clinical symptom scores were obtained. The correlation coefficient is calculated based on the amplitude of signal changes and the changes in clinical symptom scores; the expression for the correlation coefficient is as follows: ; In the formula, The correlation coefficient is denoted as n; n is the number of subjects in the current batch. For the first The amplitude of signal changes in each subject; This represents the average amplitude of signal changes in the current batch of subjects; For the first Changes in clinical symptom scores for each subject; This represents the average change in clinical symptom scores of the current batch of subjects. When the correlation coefficient is less than 0 and the p-value is less than the significance level, it is determined that the sensitization state of acupoints is associated with the functional abnormalities of the internal organs.

[0018] In the above scheme, the potential association between acupoint signal changes and clinical symptom improvement is quantified by calculating the correlation coefficient between the amplitude of signal changes at the core frequency of the target acupoint and the degree of improvement in clinical symptoms, providing specific data indicators for studying the relationship between the two. A clinically recognized visceral disease assessment scale is used to score the symptoms of subjects before and after treatment, ensuring the objectivity and authority of the changes in clinical symptom scores, and allowing the subsequent calculation of the correlation coefficient to be based on reliable symptom data. The association between acupoint sensitization status and visceral dysfunction is determined based on the correlation coefficient and p-value. Combined with statistical methods, the determination results are more scientific and credible, contributing to a deeper understanding of the intrinsic connection between acupoints and viscera, and providing theoretical support and practical guidance for TCM acupoint treatment of visceral diseases.

[0019] Secondly, this application also provides a nested verification device based on multimodal signals, including: a signal acquisition module, a signal integration module, a first processing module, a second processing module, and a result generation module; The signal acquisition module is used to acquire the first multimodal physiological signals of acupoints of the current batch of subjects before treatment and the second multimodal physiological signals after treatment; wherein, the multimodal physiological signals include acupoint impedance information, acupoint temperature information and meridian segment impedance information between acupoints at multiple preset frequencies; The signal integration module is used to generate a treatment information set based on the first multimodal physiological signal and the second multimodal physiological signal; wherein, the treatment information set includes the signal representation values ​​of the acupoints of the subject at each preset frequency, and the signal change amplitude of the subject after the application of treatment; The first processing module is used to select the frequency with the smallest signal change in the treatment information set as the core frequency, and calculate the treatment effect quantity of the current batch of subjects based on the signal characterization value under the core frequency; wherein, the treatment effect quantity is used to quantify the magnitude of clinical symptom change of the current batch of subjects after treatment. The second processing module is used to obtain the signal change amplitude of the target acupoint at the core frequency and calculate the correlation coefficient between the signal change amplitude and the degree of improvement of clinical symptoms. The target acupoint is an acupoint associated with the internal organs, and the correlation coefficient is used to characterize the correlation between the changes in the multimodal physiological signals of the acupoint and the degree of improvement of clinical symptoms. The results generation module is used to generate the acupoint status judgment results after the current batch of subjects have received treatment, based on the signal change amplitude, treatment effect size, and correlation coefficient; among them, the acupoint status judgment results are used to characterize the correlation between acupoint sensitization status and organ function.

[0020] In one implementation, the signal acquisition module is used to acquire first multimodal physiological signals of acupoints of the current batch of subjects before treatment and second multimodal physiological signals after treatment, specifically including: Scanning impedance measurements were performed at each preset frequency for a first preset time to obtain acupoint impedance information. Temperature information of acupoints is collected during scanning measurements to obtain acupoint temperature information. Scanning measurements were performed at each preset frequency to obtain the surface channel impedance between two acupoints on the same meridian, thus obtaining the meridian segment impedance information between acupoints. Obtain the acquisition timestamps of multimodal physiological signals and associate each subject ID with the corresponding multimodal physiological signal and the corresponding acquisition timestamp.

[0021] In one implementation, before generating the treatment information set based on the first and second multimodal physiological signals, the method further includes data preprocessing of the acquired multimodal physiological signals according to a preset preprocessing strategy. Specifically: Missing values ​​were detected in multimodal physiological signals, and outliers were detected using the interquartile range method. For multimodal physiological signal samples with missing values ​​and multimodal physiological signal samples marked as outliers, calculate their distances to complete multimodal physiological signal samples of the same type, and sort them in ascending order. A predetermined number of complete multimodal physiological signal samples with the highest ranking are selected. The selected complete multimodal physiological signal samples are then grouped and weighted according to the distance between the multimodal physiological signal samples and multimodal physiological signal samples with missing values ​​or those marked as outliers. Missing or outlier values ​​are imputed based on the weighted average of complete multimodal physiological signal samples.

[0022] In one implementation, the signal integration module is used to generate a treatment information set based on a first multimodal physiological signal and a second multimodal physiological signal, specifically including: The average values ​​of the first and second multimodal physiological signals at each preset frequency are calculated respectively, and the average values ​​of the multimodal physiological signals are used as signal characterization values. Among them, the signal characterization values ​​include acupoint impedance information characterization values, acupoint temperature information characterization values, and meridian segment impedance information characterization values ​​between acupoints. The signal change amplitude at each preset frequency is calculated based on the signal characterization values ​​before and after treatment; the expression for the signal change amplitude is: ; In the formula, The amplitude of the signal change. The multimodal physiological signal characterization value at a single preset frequency before treatment is applied; The multimodal physiological signal characterization value at the same preset frequency after treatment is applied; A treatment information set is generated based on each signal characterization value and signal change amplitude of the current batch of subjects.

[0023] In one implementation, the first processing module selects the frequency with the smallest signal change in the treatment information set as the core frequency, and calculates the treatment effect of the current batch of subjects based on the signal representation value at the core frequency, specifically including: Calculate the variance of the signal variation amplitude at each preset frequency, and select the frequency with the smallest variance as the core frequency; The multimodal physiological signal characterization values ​​of each subject in the current batch before and after treatment at the core frequency are obtained one by one, and the difference between the two multimodal physiological signal characterization values ​​is calculated. The average value of the obtained difference is obtained by averaging the difference. The standard deviation of the acquired differences is calculated, and the treatment effect size of the current batch of subjects is calculated based on the mean of the signal representation and the standard deviation of the differences; wherein, the expression for the treatment effect size is: ; In the formula, For effect size, The average value represents the signal. This represents the standard deviation of the difference.

[0024] In one implementation, calculating the therapeutic effect size of the current batch of subjects based on the signal representation value at the core frequency further includes: performing a paired-samples t-test based on the signal representation value and standard deviation of the current batch of subjects at the core frequency, specifically: A null hypothesis is pre-defined; the null hypothesis is that the treatment received by the current batch of subjects is ineffective. The t-statistic for the current batch of subjects is calculated based on the signal representation value and standard deviation; the expression for the t-statistic is as follows: ; In the formula, For statistical purposes, This represents the number of subjects in the current batch. The corresponding P-value is obtained based on the t-statistic. When the P-value is less than the pre-set significance level, the null hypothesis is deemed invalid, and the changes in multimodal physiological signals before and after treatment are statistically significant.

[0025] In one implementation, the second processing module is used to acquire the signal change amplitude of the target acupoint at the core frequency and calculate the correlation coefficient between the signal change amplitude and the degree of improvement in clinical symptoms, specifically including: The symptoms of the current batch of subjects before and after treatment were scored using a clinically recognized assessment scale for visceral diseases, and the changes in clinical symptom scores were obtained. The correlation coefficient is calculated based on the amplitude of signal changes and the changes in clinical symptom scores; the expression for the correlation coefficient is as follows: ; In the formula, The correlation coefficient is denoted as n; n is the number of subjects in the current batch. For the first The amplitude of signal changes in each subject; This represents the average amplitude of signal changes in the current batch of subjects; For the first Changes in clinical symptom scores for each subject; This represents the average change in clinical symptom scores of the current batch of subjects. When the correlation coefficient is less than 0 and the p-value is less than the significance level, it is determined that the sensitization state of acupoints is associated with the functional abnormalities of the internal organs.

[0026] Thirdly, this application also provides a terminal device, including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein the processor executes the computer program to implement the above-described nested verification method based on multimodal signals.

[0027] Fourthly, this application also provides a computer-readable storage medium, which includes a stored computer program, wherein the computer program, when running, controls the device where the computer-readable storage medium is located to execute the above-described nested verification method based on multimodal signals. Attached Figure Description

[0028] Figure 1 This is a flowchart illustrating a nested verification method based on multimodal signals provided in one embodiment of the present invention; Figure 2 This is an example diagram of outlier detection based on the interquartile range method provided in one embodiment of the present invention; Figure 3 This is a schematic diagram of a nested verification device based on multimodal signals provided in one embodiment of the present invention. Detailed Implementation

[0029] The specific embodiments of the present invention will be described in further detail below with reference to the accompanying drawings and examples. The following examples are for illustrative purposes only and are not intended to limit the scope of the invention.

[0030] The terms "first" and "second," etc., in the specification, claims, and drawings of this application are used to distinguish different objects, not to describe a specific order. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or apparatus that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to such processes, methods, products, or apparatus.

[0031] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0032] Example 1 See Figure 1 , Figure 1 This is a flowchart illustrating a nested verification method based on multimodal signals according to an embodiment of the present invention. The embodiment of the present invention provides a nested verification method based on multimodal signals, including steps 101 to 105, each step being as follows: Step 101: Collect the first multimodal physiological signals of the acupoints of the current batch of subjects before the application of treatment and the second multimodal physiological signals after the application of treatment; wherein, the multimodal physiological signals include acupoint impedance information, acupoint temperature information and meridian segment impedance information between acupoints at multiple preset frequencies.

[0033] In this embodiment of the invention, multimodal physiological signals of acupoints of the current batch of subjects before and after treatment are collected. The multimodal physiological signals include information such as acupoint impedance, acupoint temperature and meridian segment impedance between acupoints, which can reflect the changes in the physiological state of acupoints and meridians before and after treatment from different perspectives, and help to more comprehensively and accurately assess the specific impact of treatment on human physiological functions.

[0034] In one embodiment, collecting the first multimodal physiological signals of acupoints of the current batch of subjects before treatment and the second multimodal physiological signals after treatment specifically includes: performing scanning impedance measurements at each preset frequency for a first preset time to obtain acupoint impedance information; collecting acupoint temperature information during scanning measurements to obtain acupoint temperature information; performing scanning measurements at each preset frequency to obtain the surface channel impedance between two acupoints on the same meridian to obtain meridian segment impedance information between acupoints; obtaining the collection timestamp of the multimodal physiological signals, and associating each subject ID with the corresponding multimodal physiological signal and the corresponding collection timestamp.

[0035] In this embodiment of the invention: at each preset frequency (e.g., 10 specific frequencies: 60Hz, 70Hz, ..., 150Hz), the AD5933 chip is used to perform scanning impedance measurements on acupoints for a first preset time (e.g., 60 seconds). At each frequency, a time series containing 60 raw impedance values ​​is obtained, representing the acupoint impedance information, with a sampling rate of 1Hz. Then, simultaneously with the acquisition of acupoint impedance information, a high-precision temperature sensor (e.g., LMT70) is used to record the temperature value of the acupoint. Synchronous acquisition for 60 seconds at a sampling rate of 1Hz yields a time series of 60 temperature values, representing the acupoint temperature information. Furthermore, for two acupoints on the same meridian (e.g., ST36 and ST35), scanning measurements are performed at 10 frequencies, acquiring a representative value of the channel impedance at each frequency, thus obtaining the meridian segment impedance information between acupoints. The acquisition timestamps of the multimodal physiological signals are obtained, and each subject ID is associated with the corresponding multimodal physiological signal and the corresponding acquisition timestamp. The data were clearly divided into two groups: pre-treatment and post-treatment. They were linked by “subject ID”, “acupoint / channel name”, and “collection timestamp” to facilitate subsequent paired statistical comparisons (such as paired t-tests).

[0036] For example, suppose we want to study the effect of a certain traditional Chinese medicine acupuncture treatment on the physiological signals of acupoints. Ten subjects were selected, and multimodal physiological signals of their Zusanli acupoint (ST36) were collected before and after treatment. Using an AD5933 chip, scanning impedance measurements were performed on the ST36 acupoint at 10 specific frequencies (60Hz-150Hz, 10Hz steps), each frequency lasting 60 seconds, with a sampling rate of 1Hz. For example, at 70Hz, the raw impedance data for the left side of the ST36 acupoint of subject 001 is as follows: Acupoint: ST36_left, Frequency: 70Hz, Time points: t1~t5 { "acupoint": "ST36_left", "frequency": "70Hz", "impedance_time_series": [54.2, 54.5, 53.9, 54.1, 53.8, ...] # An array of length 60 } Then, in sync with the impedance data, the temperature values ​​of acupoints ST36 and ST37 were acquired using an LMT70 temperature sensor for 60 seconds at a sampling rate of 1Hz. For example, the raw temperature data of the left side of acupoint ST36 for subject 001 is as follows: # Acupoint: ST36_left { "acupoint": "ST36_left", "temperature_time_series": [36.52, 36.53, 36.55, 36.57, 36.58, ...] # An array of length 60 } Furthermore, the surface channel impedance between ST36 and ST37 was measured, scanned at 10 frequencies, and a representative value of the channel impedance was obtained at each frequency. For example, at 70 Hz, the impedance data of the left channel of the ST36-ST35 acupoints for subject 001 are as follows: # Channels: ST36_left-ST35_left, Frequency: 70Hz { "channel": "ST36_left-ST35_left", "frequency": "70Hz", "impedance_value": 48.7 } Step 102: Generate a treatment information set based on the first multimodal physiological signal and the second multimodal physiological signal; wherein, the treatment information set includes the signal representation values ​​of the subject's acupoints at each preset frequency, and the signal change amplitude of the subject after the application of treatment.

[0037] In this embodiment of the invention, a treatment information set is generated based on the first multimodal physiological signals of the subject's acupoints before treatment and the second multimodal physiological signals after treatment. This treatment information set includes two parts: first, the signal representation values ​​of the subject's acupoints at each preset frequency; and second, the amplitude of changes in each physiological signal after treatment. The treatment information set integrates signal representation values ​​at multiple preset frequencies and the amplitude of signal changes after treatment, reflecting the impact of treatment on the physiological signals of the subject's acupoints from multiple dimensions and frequencies. This avoids the limitations of evaluation based on a single frequency or single indicator, making the evaluation of treatment effects more comprehensive and accurate.

[0038] In one embodiment, before generating the treatment information set based on the first and second multimodal physiological signals, the method further includes data preprocessing of the acquired multimodal physiological signals according to a preset preprocessing strategy. Specifically, this involves: detecting missing values ​​and outliers in the multimodal physiological signals using the interquartile range method; calculating the distance between multimodal physiological signal samples with missing values ​​and those marked as outliers and complete multimodal physiological signal samples of the same type, and sorting them in ascending order; selecting a preset number of complete multimodal physiological signal samples from the top of the sorted list; grouping and weighting the selected complete multimodal physiological signal samples according to the distance between the multimodal physiological signal samples and the multimodal physiological signal samples with missing values ​​or those marked as outliers; and imputing missing or outliers based on the average value of the weighted complete multimodal physiological signal samples.

[0039] Before generating the treatment information set, data preprocessing of the acquired multimodal physiological signals is essential because the raw data may contain missing and outlier values, which can affect the accuracy and reliability of subsequent analysis. In this embodiment of the invention, the presence of missing values ​​in the multimodal physiological signal data is checked. This could be due to reasons such as measurement equipment malfunction or data transmission problems. The interquartile range (IQR) method is used to identify outliers. Figure 2This is an example diagram of outlier detection based on the interquartile range (IQR) method provided in one embodiment of the present invention. The whiskers above and below the box plot represent the range of data, excluding points considered outliers. Typically, the upper bound of the whiskers is Q3 plus 1.5 times the IQR, and the lower bound is Q1 minus 1.5 times the IQR. Points outside the whisker range are generally considered outliers. The small dots in the diagram represent outliers, i.e., data points falling outside the whiskers. These points may be due to measurement errors, data entry errors, or are part of the true data distribution.

[0040] First, calculate the first quartile (Q1) and third quartile (Q3) of the data, then obtain the interquartile range (IQR) = Q3 - Q1. Define upper and lower bounds: lower bound = Q1 - 1.5 × IQR, upper bound = Q3 + 1.5 × IQR. For samples with missing values ​​and samples marked as outliers, calculate their distances to complete multimodal physiological signal samples of the same type, and then sort them in ascending order according to their distances. Common distance metrics such as Euclidean distance and Manhattan distance can be used here. Select a predetermined number of complete multimodal physiological signal samples from the top of the sorted list, and group them into weighted groups based on their distances to samples with missing values ​​or outliers. Samples with closer distances have higher weights, for example, weights of 0.5, 0.3, 0.2, etc. Based on the weighted average of the selected complete multimodal physiological signal samples, imputation is performed on the missing or outliers, i.e., the missing or outliers are replaced with the weighted average. Alternatively, besides Euclidean and Manhattan distances, other suitable distance metrics can be chosen based on the characteristics of the data, such as Chebyshev distance and Mahalanobis distance. Different distance metrics may affect the final interpolation results; therefore, the optimal distance metric can be selected by comparing the effects of different methods experimentally.

[0041] For example, impedance data of the Zusanli (left) acupoint were collected from 10 subjects, and some data are shown below: Subject ID | Impedance Value | | 001| 52.1, 52.3, 52.0, 52.2, … | | 002| 53.0, 53.2, NaN, 53.4, … | / / Missing values ​​exist. | 003| 51.8, 51.9, 52.1, 52.0, … | | 004| 54.0, 54.2, 54.1, 54.3, … | | 005| 52.5, 52.6, 52.7, 52.8, … | | 006| 53.5, 53.6, 53.7, 53.8, … | | 007| 51.5, 51.6, 51.7, 51.8, … | | 008| 55.0, 55.2, 55.1, 55.3, … | | 009| 52.2, 52.3, 52.4, 52.5, … | | 010| 56.0, 56.2, 56.1, 56.3, … | / / Outliers may exist. Missing values ​​(NaN) were found in the impedance data of subject 002. Outlier detection was performed on all data using the interquartile range (IQR). The calculated IQR values ​​were Q1 = 52.0, Q3 = 53.8, and IQR = 53.8 - 52.0 = 1.8. The lower bound was 52.0 - 1.5 × 1.8 = 49.3, and the upper bound was 53.8 + 1.5 × 1.8 = 56.5. Some impedance values ​​of subject 010 exceeded the upper bound and were marked as outliers. Taking the missing value sample of subject 002 as an example, the Euclidean distance between it and other complete samples (subjects 001, 003, 004, 005, 006, 007, 008, 009) was calculated and sorted in ascending order. The sorting results are assumed to be as follows: |Subject ID| Distance| | 003| 0.5 | | 009| 0.6 | | 001| 0.7 | | 005| 0.8 | | 006| 0.9 | | 007 | 1.0 | The top 6 complete samples (subjects 003, 009, 001, 005, 006, and 007) were selected and weighted according to weights of 0.5, 0.3, and 0.2. Assuming the impedance values ​​corresponding to these samples are 51.8, 52.2, 52.1, 52.5, 53.5, and 51.5 respectively, the weighted impedance values ​​are as follows: Subject 003: 51.8 × 0.5 = 25.9; Subject 009: 52.2 × 0.3 = 15.66; Subject 001: 52.1 × 0.2 = 10.42; Subject 005: 52.5 × 0 = 0; Subject 006: 53.5 × 0 = 0; Subject 007: 51.5 × 0 = 0; Calculate the weighted sample mean: (25.9+15.66+10.42) / (0.5+0.3+0.2)=52.0 (rounded). Use this value to impute the missing value of subject 002.

[0042] For outliers in the subject's 010 range, imputation was performed following the same steps described above, which will not be elaborated here. This completes the data preprocessing. These data preprocessing steps improve data quality and provide a reliable data foundation for subsequent generation of treatment information sets and analysis of treatment effects.

[0043] In one embodiment, generating a treatment information set based on a first multimodal physiological signal and a second multimodal physiological signal specifically includes: calculating the average value of the first multimodal physiological signal and the second multimodal physiological signal at each preset frequency, and using the average value of the multimodal physiological signal as a signal characterization value; wherein, the signal characterization value includes acupoint impedance information characterization value, acupoint temperature information characterization value, and meridian segment impedance information characterization value between acupoints; The signal change amplitude at each preset frequency is calculated based on the signal characterization values ​​before and after treatment; the expression for the signal change amplitude is: ; In the formula, The amplitude of the signal change. The multimodal physiological signal characterization value at a single preset frequency before treatment is applied; The multimodal physiological signal characterization value at the same preset frequency after treatment is applied; A treatment information set is generated based on each signal characterization value and signal change amplitude of the current batch of subjects.

[0044] In this embodiment of the invention, for the first multimodal physiological signal (before treatment) and the second multimodal physiological signal (after treatment), the average value of the multimodal physiological signal is calculated at each preset frequency. That is, for each subject, at each frequency dimension (e.g., 70Hz), the average value of impedance or temperature data of all sampling points (n=60) is calculated. This is to eliminate the influence of time series fluctuations and obtain a stable value that can represent the physiological signal characteristics at that frequency, as the signal characterization value. The signal characterization value includes acupoint impedance information characterization value, acupoint temperature information characterization value, and meridian segment impedance information characterization value between acupoints. These different types of characterization values ​​reflect the physiological state of acupoints and meridians from multiple aspects. To make the signal changes between different acupoints or individuals comparable, a percentage change method is used to calculate the signal change amplitude. Because the initial baseline values ​​of different acupoints or individuals vary greatly, using absolute change amounts lacks comparability, while percentage change can standardize the change amplitude. A treatment information set is generated based on each signal characterization value and signal change amplitude of the current batch of subjects. This dataset contains the signal characteristics and changes of each subject, each acupoint / channel, before and after treatment at each frequency, providing a comprehensive data foundation for subsequent treatment effect evaluation and analysis.

[0045] For example, suppose a traditional Chinese medicine clinic performed acupuncture treatment on 5 subjects and collected impedance and temperature information of their Zusanli acupoints. The preset frequencies were 50Hz, 70Hz, and 90Hz, with 60 sampling points collected at each frequency. Taking subject 001's acupoint impedance data at 70Hz as an example, assuming the collected data from the 60 sampling points are... , ... The impedance information of acupoints at this frequency is characterized by: = The same method was used to calculate the acupoint impedance information, acupoint temperature information, and meridian segment impedance information of other subjects at each preset frequency. The signal characteristics and signal variation amplitudes of each subject at each preset frequency were compiled into a dataset, as shown in the table below:

[0046] This treatment information set can be used for subsequent treatment effect evaluation and analysis, such as comparing the treatment effects at different frequencies and the differences in treatment response among different subjects.

[0047] Step 103: Select the frequency with the smallest signal change in the treatment information set as the core frequency, and calculate the treatment effect size of the current batch of subjects based on the signal characterization value under the core frequency; wherein, the treatment effect size is used to quantify the magnitude of clinical symptom changes in the current batch of subjects after treatment. In this embodiment of the invention, the frequency with the smallest signal change is selected from the treatment information set and determined as the core frequency. Then, based on the signal characterization value at this core frequency, the treatment effect size for the current batch of subjects is calculated. This effect size can be used to measure the magnitude of change in clinical symptoms after treatment in the current batch of subjects. Selecting the frequency with the smallest signal change as the core frequency helps to accurately locate frequencies that play a key role in the treatment effect. In multimodal physiological signals, different frequencies may represent different physiological mechanisms and treatment responses. The frequency with the smallest signal change may be related to stable and continuous physiological regulation induced by treatment. Using this as the core frequency for subsequent analysis can more accurately reflect the core mechanism of action of the treatment. By calculating the treatment effect size at the core frequency, the magnitude of change in clinical symptoms after treatment in the subjects is quantified. This quantification method makes the treatment effect no longer a vague subjective judgment, but an objective indicator with specific numerical values, making it easier for doctors and researchers to more intuitively and accurately assess the effectiveness of the treatment.

[0048] In one embodiment, the frequency with the smallest signal change in the treatment information set is selected as the core frequency, and the treatment effect magnitude of the current batch of subjects is calculated based on the signal characterization value at the core frequency, specifically including: Calculate the variance of the signal variation amplitude at each preset frequency, and select the frequency with the smallest variance as the core frequency; The multimodal physiological signal characterization values ​​of the current batch of subjects before and after treatment at the core frequency are obtained. The average value of the signal characterization is obtained by averaging the difference between the two multimodal physiological signal characterization values. Calculate the standard deviation of the difference, and then calculate the treatment effect size for the current batch of subjects based on the mean and standard deviation of the signal representation; the expression for the treatment effect size is: ; In the formula, For effect size, The average value represents the signal. The standard deviation is denoted as .

[0049] In this embodiment of the invention, a core frequency is determined from the treatment information set, and the therapeutic effect of the current batch of subjects is calculated based on the signal characterization values ​​at this frequency. This quantifies the change amplitude before and after treatment, providing a standardized indicator for evaluating the treatment effect. The variance of the signal change amplitude at each preset frequency is calculated. Variance reflects the dispersion of the data. The smaller the variance, the closer the signal change amplitude of different subjects is at that frequency, meaning the change in the group data is more consistent and stable, with minimal noise interference. Such a frequency is most suitable as the core frequency for subsequent analysis. Furthermore, in practical applications, there may be situations where the variances of multiple frequencies are relatively close. In this case, other indicators, such as the average value of the signal change amplitude at that frequency, can be combined to comprehensively judge and select the most suitable core frequency. The multimodal physiological signal characterization values ​​include acupoint impedance information characterization values, acupoint temperature information characterization values, and meridian segment impedance information characterization values ​​between acupoints, among other information. When calculating the average value of the signal characterization, it should be ensured that each signal is reasonably processed and comprehensively analyzed. For example, corresponding weights can be assigned according to the importance of different signals, and then a weighted average calculation can be performed. As an alternative, only one type of information characterization value can be selected as the object of subsequent processing. The therapeutic effect size (Cohen's d) standardizes the mean change (mean value of the signal representation) using the degree of variability (standard deviation) of individual changes. This makes effect sizes comparable across different studies and different indicators, allowing for a more objective assessment of the magnitude of treatment effects. Generally, a larger absolute value of Cohen's d indicates a stronger therapeutic effect. Typically, d=0.2 is considered a small effect, d=0.5 a moderate effect, and d=0.8 a large effect. However, in practical applications, the magnitude of the effect size needs to be comprehensively judged in conjunction with the specific research field and clinical circumstances.

[0050] For example, a rehabilitation center conducted a month-long rehabilitation treatment on 10 patients, collecting acupoint impedance information before and after treatment at preset frequencies of 30Hz, 50Hz, 70Hz, and 90Hz. The acupoint impedance data for each patient before and after treatment at each preset frequency was collected, and the signal change amplitude (percentage change) at each frequency was calculated using the previous method, as shown in the table below:

[0051] The variance data above shows that the variance is smallest at 70Hz, therefore 70Hz was chosen as the core frequency. The acupoint impedance information of 10 patients before and after treatment at 70Hz was obtained, and the difference calculation results are shown in the table below:

[0052] Calculate the average signal representation based on the table above. = Then calculate the sum of squares of their differences: (5 4.6) 2 +(4 4.6) 2 +(4 4.6) 2 +(5 4.6) 2 +(4 4.6) 2 +(5 4.6) 2 +(4 4.6) 2 +(5 4.6) 2 +(4 4.6) 2 +(5 4.6) 2 =2.4, standard deviation = The therapeutic effect size can be calculated based on the expression for it. Since the effect size of 8.85 is much greater than 0.8, it indicates that the rehabilitation therapy has a very significant effect on the acupoint impedance of patients at a frequency of 70Hz, and the therapeutic effect is very strong.

[0053] In one embodiment, calculating the therapeutic effect size of the current batch of subjects based on the signal representation value at the core frequency further includes: performing a paired-samples t-test based on the signal representation value and standard deviation of the current batch of subjects at the core frequency, specifically: A null hypothesis is pre-defined; the null hypothesis is that the treatment received by the current batch of subjects is ineffective. The t-statistic for the current batch of subjects is calculated based on the signal representation value and standard deviation; the expression for the t-statistic is as follows: ; In the formula, For statistical purposes, This represents the number of subjects in the current batch. The corresponding P-value is obtained based on the t-statistic. When the P-value is less than the pre-set significance level, the null hypothesis is deemed invalid, and the changes in multimodal physiological signals before and after treatment are statistically significant.

[0054] When calculating the treatment effect size for the current batch of subjects based on the signal characterization values ​​at the core frequency, in addition to calculating Cohen's d effect size, a paired-samples t-test is also introduced. These two steps complement each other, assessing the treatment effect from different dimensions. The paired-samples t-test focuses on whether the difference before and after treatment is statistically significant, while Cohen's d effect size measures the actual magnitude and clinical significance of this difference.

[0055] In this embodiment of the invention, the null hypothesis is set as the treatment received by the current batch of subjects being ineffective, meaning there is no difference in the mean values ​​of multimodal physiological signals before and after treatment. This is the basis of the statistical test, and subsequent analyses revolve around whether to reject this null hypothesis. In the formula for calculating the statistic, It is the mean of the differences in signal characteristics before and after treatment for all subjects, reflecting the average magnitude of change. The standard deviation of the difference reflects the consistency of the variation range among different subjects. The t-statistic takes into account both the mean variation and individual differences, and is used to measure the degree of deviation of the sample data from the null hypothesis. Paired-samples t-tests and Cohen's d effect size calculations share the same data source, describing the treatment effect from the two dimensions of "statistical significance" and "effect strength," respectively. Only when both are analyzed together can the effectiveness and practical significance of the treatment be more comprehensively assessed. For example, even if the t-test shows a significant difference before and after treatment, if the Cohen's d effect size is small, it indicates that this difference may not be important in practical applications; conversely, if the Cohen's d effect size is large, but the p-value is greater than the significance level, it may be due to insufficient sample size or other reasons leading to statistical insignificance.

[0056] For example, the null hypothesis The rehabilitation treatment received by the current batch of patients was ineffective, meaning there was no difference in the mean values ​​of acupoint impedance information before and after treatment. (Known) , If n=10, then according to the expression of the t-statistic, we can obtain t. The p-value corresponding to the t-statistic was calculated using statistical analysis software (such as the SciPy library in Python). Assuming the calculated p-value is much smaller than the pre-set significance level α=0.01, the null hypothesis is invalid, indicating that the change in the acupoint impedance information representation value before and after treatment is statistically significant. Since Cohen's effect size d≈8.85≥0.5 and the p-value is much smaller than 0.01, this indicates that the rehabilitation treatment not only produced a statistically significant change, but also a large magnitude of change, possessing clear clinical or physiological significance.

[0057] Step 104: Obtain the signal change amplitude of the target acupoint at the core frequency, and calculate the correlation coefficient between the signal change amplitude and the degree of improvement of clinical symptoms; where the target acupoint is an acupoint associated with the internal organs, and the correlation coefficient is used to characterize the correlation between the changes in the multimodal physiological signals of the acupoint and the degree of improvement of clinical symptoms.

[0058] In this embodiment of the invention, the signal change amplitude of target acupoints associated with viscera at the core frequency is obtained, and then the correlation coefficient between this signal change amplitude and the degree of improvement of clinical symptoms is calculated, so as to characterize the correlation between the multimodal physiological signal changes of acupoints and the degree of improvement of clinical symptoms.

[0059] In one embodiment, the signal change amplitude of the target acupoint at the core frequency is obtained, and the correlation coefficient between the signal change amplitude and the degree of improvement of clinical symptoms is calculated. Specifically, this includes: using a clinically recognized visceral disease assessment scale to score the symptoms of the current batch of subjects before and after treatment, and obtaining the change value of clinical symptom score. The correlation coefficient is calculated based on the amplitude of signal changes and the changes in clinical symptom scores; the expression for the correlation coefficient is as follows: ; In the formula, The correlation coefficient is denoted as n; n is the number of subjects in the current batch. For the first The amplitude of signal changes in each subject; This represents the average amplitude of signal changes in the current batch of subjects; For the first Changes in clinical symptom scores for each subject; This represents the average change in clinical symptom scores of the current batch of subjects. When the correlation coefficient is less than 0 and the p-value is less than the significance level, it is determined that the sensitization state of acupoints is associated with the functional abnormalities of the internal organs.

[0060] In this embodiment of the invention, the acquisition of clinical symptom scores should emphasize the importance of clinically recognized visceral disease assessment scales. Different visceral diseases should be assessed using scales that are highly targeted and have high reliability and validity. For example, specific assessment scales related to spleen and stomach function can be used for spleen and stomach diseases, while specialized cardiovascular disease assessment scales should be used for cardiovascular diseases. When using assessment scales for scoring, it should be ensured that the assessors are professionally trained and familiar with the usage and scoring criteria of the scales to guarantee the accuracy and consistency of the scores. Simultaneously, the scoring process should be conducted in a relatively stable environment to avoid interference from external factors.

[0061] For example, a rehabilitation center conducted a one-month rehabilitation treatment on 10 patients with abnormal spleen and stomach function, collecting acupoint impedance information before and after treatment, with preset frequencies of 30Hz, 50Hz, 70Hz, and 90Hz. After calculation, 70Hz was selected as the core frequency, and the target acupoint was Zusanli (ST36). The relevant treatment information is shown in the table below:

[0062] Based on the above table, the average signal change amplitude and the average clinical symptom score change value of the current batch of subjects can be calculated to obtain the following results. , The correlation coefficient is obtained from the above. The p-value corresponding to the correlation coefficient was calculated using statistical analysis software (such as Python's SciPy library). Assuming the calculated p-value is 0.03, since r < 0 and p < 0.05, this indicates a significant negative correlation between the decrease in acupoint impedance and the improvement of clinical symptoms, meaning that the sensitization of this acupoint (manifested as a decrease in impedance) is associated with abnormal spleen and stomach function.

[0063] Step 105: Generate the acupoint status judgment results after treatment for the current batch of subjects based on the signal change amplitude, therapeutic effect size and correlation coefficient; wherein, the acupoint status judgment results are used to characterize the correlation between acupoint sensitization state and organ function.

[0064] First, the signal change amplitude of the target acupoint at the core frequency was obtained. Simultaneously, a clinically recognized visceral disease assessment scale was used to score the symptoms of the current batch of subjects before and after treatment, obtaining the change values ​​of clinical symptom scores. Next, the correlation coefficient between the signal change amplitude and the change values ​​of clinical symptom scores was calculated. Finally, combining the signal change amplitude, therapeutic effect size, and correlation coefficient, the acupoint status assessment results after treatment for the current batch of subjects were generated, thereby verifying the correlation between acupoint sensitization status and visceral function.

[0065] In this embodiment of the invention, the acupoint status judgment result is generated by comprehensively considering the signal change amplitude (% Change), therapeutic effect size (Cohen's d), and correlation coefficient (r and P value). The judgment process requires the establishment of clear judgment criteria. For example, when the acupoint impedance decreases significantly, the effect size reaches a certain threshold, and the correlation coefficient is significant (P<0.05), the acupoint can be judged to be in a sensitized state and associated with abnormal organ function. The final output result is generated as a structured report in JSON or CSV format, facilitating data storage, transmission, and further analysis.

[0066] For example, the acupoint status assessment results include the following: (1) Percentage change in impedance and temperature of each acupoint / channel at the optimal frequency (% Change): The signal changes of each acupoint or channel at the optimal frequency are listed in detail, and the differences before and after treatment are shown intuitively; (2) Corresponding statistical significance (P value) and effect size (Cohen's d): Provides the P value and effect size for each acupoint or channel to help readers judge the significance and magnitude of the change; (3) Results of the correlation analysis of viscera and organs (including Pearson correlation coefficient r and its P value): show the strength and direction of the linear correlation between acupoint sensitization and improvement of visceral function, as well as the significance of this correlation.

[0067] (4) Sensitization status judgment of acupoints: Based on the comprehensive analysis results, give conclusive diagnostic suggestions, clearly indicating which acupoints are in a sensitized state and their correlation with abnormal organ function; (5) Suggestions for further examination or treatment: Based on the results of the acupoint condition assessment, provide targeted suggestions to clinicians, such as suggesting further stimulation treatment of certain acupoints or more detailed examination of related organs.

[0068] In this embodiment of the invention, transcutaneous electrical stimulation applied to specific acupoints is a low-frequency intervention at the body surface scale. Acupoints are sites on the body surface where Qi and blood from the meridians are infused. When transcutaneous electrical stimulation is applied to acupoints, it triggers changes in the physical properties of the acupoints, such as alterations in impedance and temperature. These changes are considered low-frequency resonant signals. This signal is transmitted along the meridian channels, causing synergistic changes in other acupoints and meridian segments on the same meridian. This process demonstrates the nesting and resonance of "acupoint-meridian" at the same body surface scale. As channels for the flow of Qi and blood, the meridians connect various acupoints into an organic whole. When an acupoint is stimulated and generates a resonant signal, the meridian, like a transmission line, transmits the signal to other related acupoints, allowing acupoints on the same meridian to influence and synergize with each other. This fully demonstrates the close nesting relationship between acupoints and meridians at the body surface scale and the resonant transmission mechanism. From a cross-scale perspective, this "acupoint-meridian" resonance phenomenon at the body surface scale does not exist in isolation but is closely related to the internal organs at a deeper level. The physiological and pathological states of the internal organs are reflected on acupoints on the body surface through the meridians, and the stimulation of acupoints on the body surface can also affect the function of the internal organs through the meridians.

[0069] The study collected first-mode multimodal physiological signals from acupoints of the current batch of subjects before treatment (transcutaneous electrical stimulation) and second-mode multimodal physiological signals after treatment. These multimodal physiological signals included acupoint impedance information, acupoint temperature information, and meridian segment impedance information between acupoints at multiple preset frequencies. These signals not only contained acupoint and meridian information at the body surface scale but also contained relevant information about deep internal organs. The treatment information set generated based on the first-mode and second-mode multimodal physiological signals covered the signal characterization values ​​of the subject's acupoints at each preset frequency and the amplitude of signal changes after treatment. By selecting the frequency with the smallest signal change in the treatment information set as the core frequency and calculating the treatment effect of the current batch of subjects based on the signal characterization values ​​at the core frequency, interference factors could be eliminated, and the amplitude of changes in the subjects' clinical symptoms could be accurately quantified. Changes in clinical symptoms are often closely related to changes in organ function.

[0070] Furthermore, by acquiring the signal variation amplitude of target acupoints (acupoints associated with internal organs) at the core frequency and calculating the correlation coefficient between this amplitude and the degree of improvement in clinical symptoms, this process reveals the intrinsic connection between changes in multimodal physiological signals of acupoints and improvements in organ function (manifested as improvements in clinical symptoms), verifying the cross-scale nested relationship and resonance conduction mechanism of "acupoint-meridian-internal organs." That is, stimulation of acupoints on the body surface is conducted through the meridians, triggering changes in the function of internal organs, and these changes in organ function are then fed back to acupoints on the body surface through the meridians, forming a complete cross-scale nested and resonant system.

[0071] Finally, based on the signal change amplitude, therapeutic effect size, and correlation coefficient, the acupoint status assessment results of the current batch of subjects after treatment were generated. These results, combined with various key indicators, scientifically and systematically verified the correlation between acupoint sensitization and organ function, further confirming the cross-scale nesting relationship and resonance conduction mechanism between "acupoints (surface scale) - meridians (surface channel scale) - organs (deep internal scale)". This theoretical framework provides an objective scientific basis for TCM acupoint theory and solid theoretical support for accurately assessing acupoint status and developing personalized treatment plans in clinical practice.

[0072] In this embodiment of the invention, a nested verification device based on multimodal signals is also provided, including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, it implements the above-described nested verification method based on multimodal signals.

[0073] In this embodiment of the invention, a computer-readable storage medium is also provided, which includes a stored computer program, wherein the computer program controls the device where the computer-readable storage medium is located to execute the above-described nested verification method based on multimodal signals when it is running.

[0074] For example, a computer program can be divided into one or more modules, one or more of which are stored in memory and executed by a processor to perform the present invention. The one or more modules can be a series of computer program instruction segments capable of performing a specific function, which describe the execution process of the computer program in a nested verification device based on multimodal signals.

[0075] The nested verification device based on multimodal signals can be a computing device such as a desktop computer, laptop, handheld computer, or cloud server. The nested verification device based on multimodal signals may include, but is not limited to, a processor, memory, and a display. Those skilled in the art will understand that the above components are merely examples of a nested verification device based on multimodal signals and do not constitute a limitation on the device. It may include more or fewer components, combinations of certain components, or different components. For example, a nested verification device based on multimodal signals may also include input / output devices, network access devices, buses, etc.

[0076] The processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor can be a microprocessor or any conventional processor. The processor is the control center of the nested verification device based on multimodal signals, connecting all parts of the device through various interfaces and lines.

[0077] The memory can be used to store computer programs and / or modules. The processor implements various functions of the nested verification device based on multimodal signals by running or executing the computer programs and / or modules stored in the memory, and by calling data stored in the memory. The memory can mainly include a program storage area and a data storage area. The program storage area can store the operating system, at least one application program required for a function (such as sound playback function, text conversion function, etc.), etc.; the data storage area can store data created according to the use of the mobile phone (such as audio data, text message data, etc.). In addition, the memory can include high-speed random access memory, and can also include non-volatile memory, such as hard disk, RAM, plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, at least one disk storage device, flash memory device, or other volatile solid-state storage device.

[0078] In this invention, the module based on nested verification using multimodal signals, if implemented as a software functional unit and sold or used as an independent product, can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the above embodiments can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include: any entity or device capable of carrying computer program code, recording media, USB flash drives, portable hard drives, magnetic disks, optical disks, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc. Those skilled in the art can understand and implement this invention without any inventive effort.

[0079] This invention provides a nested verification method based on multimodal signals. The method collects multimodal physiological signals, including acupoint impedance, temperature, and meridian segment impedance information at multiple preset frequencies, before and after treatment. This multi-dimensional data acquisition provides a rich and comprehensive information foundation for subsequent analysis. It closely aligns with the concept of cross-scale verification, building an information bridge from acupoints at the body surface scale and meridians at the body surface channel scale to the internal organs at the deep internal scale. The resulting treatment information set covers signal characterization values ​​at each preset frequency and the amplitude of signal changes after treatment, demonstrating the dynamic changes of acupoints during treatment. Next, the frequency with the smallest signal change is selected as the core frequency to calculate the treatment effect. This approach eliminates interference factors during treatment, allowing the calculated effect to more accurately quantify the changes in the subject's clinical symptoms, reflecting the intrinsic connection and resonant conduction mechanism between acupoints, meridians, and internal organs. Subsequently, by acquiring the signal variation amplitude of the target acupoint at the core frequency and calculating its correlation coefficient with the degree of improvement in clinical symptoms, the intrinsic connection between the multimodal physiological signal changes of acupoints and the improvement of clinical symptoms was revealed. This step further deepened the connotation of cross-scale validation. The target acupoint is associated with the internal organs, and its signal variation amplitude at the core frequency reflects the cross-scale information transmission between the surface acupoint and the deep internal organs. Finally, the acupoint status judgment result was generated by comprehensively considering the signal variation amplitude, therapeutic effect size, and correlation coefficient. This process organically combines various key indicators, scientifically and systematically verifying the correlation between acupoint sensitization state and organ function. It can provide objective scientific evidence for the theory of acupoints in traditional Chinese medicine and provide strong support for doctors to accurately judge the acupoint status and formulate personalized treatment plans in clinical practice. Doctors can understand the specific relationship between acupoints, meridians, and internal organs based on the acupoint status judgment results, thereby accurately adjusting treatment strategies and improving the pertinence and effectiveness of treatment.

[0080] Example 2 See Figure 3 , Figure 3 This is a schematic diagram of a nested verification device based on multimodal signals according to an embodiment of the present invention. The embodiment of the present invention provides a nested verification device based on multimodal signals, including: a signal acquisition module 201, a signal integration module 202, a first processing module 203, a second processing module 204, and a result generation module 205; The signal acquisition module 201 is used to acquire the first multimodal physiological signals of the acupoints of the current batch of subjects before the application of treatment and the second multimodal physiological signals after the application of treatment; wherein, the multimodal physiological signals include acupoint impedance information, acupoint temperature information and meridian segment impedance information between acupoints at multiple preset frequencies; The signal integration module 202 is used to generate a treatment information set based on the first multimodal physiological signal and the second multimodal physiological signal; wherein, the treatment information set includes the signal representation values ​​of the acupoints of the subject at each preset frequency, and the signal change amplitude of the subject after the treatment is applied; The first processing module 203 is used to select the frequency with the smallest signal change in the treatment information set as the core frequency, and calculate the treatment effect quantity of the current batch of subjects based on the signal characterization value under the core frequency; wherein, the treatment effect quantity is used to quantify the magnitude of clinical symptom change of the current batch of subjects after treatment. The second processing module 204 is used to acquire the signal change amplitude of the target acupoint at the core frequency and calculate the correlation coefficient between the signal change amplitude and the degree of improvement of clinical symptoms; wherein, the target acupoint is an acupoint related to the viscera, and the correlation coefficient is used to characterize the correlation between the change of multimodal physiological signals of the acupoint and the degree of improvement of clinical symptoms. The result generation module 205 is used to generate the acupoint status judgment results after the current batch of subjects have been treated, based on the signal change amplitude, treatment effect size and correlation coefficient; wherein, the acupoint status judgment results are used to characterize the correlation verification results between acupoint sensitization state and organ function.

[0081] In one embodiment, the signal acquisition module 201 is used to acquire the first multimodal physiological signals of acupoints of the current batch of subjects before the application of treatment and the second multimodal physiological signals after the application of treatment, specifically including: Scanning impedance measurements were performed at each preset frequency for a first preset time to obtain acupoint impedance information. Temperature information of acupoints is collected during scanning measurements to obtain acupoint temperature information. Scanning measurements were performed at each preset frequency to obtain the surface channel impedance between two acupoints on the same meridian, thus obtaining the meridian segment impedance information between acupoints. Obtain the acquisition timestamps of multimodal physiological signals and associate each subject ID with the corresponding multimodal physiological signal and the corresponding acquisition timestamp.

[0082] In one embodiment, before generating the treatment information set based on the first and second multimodal physiological signals, the method further includes data preprocessing of the acquired multimodal physiological signals according to a preset preprocessing strategy. Specifically: Missing values ​​were detected in multimodal physiological signals, and outliers were detected using the interquartile range method. For multimodal physiological signal samples with missing values ​​and multimodal physiological signal samples marked as outliers, calculate their distances to complete multimodal physiological signal samples of the same type, and sort them in ascending order. A predetermined number of complete multimodal physiological signal samples with the highest ranking are selected. The selected complete multimodal physiological signal samples are then grouped and weighted according to the distance between the multimodal physiological signal samples and multimodal physiological signal samples with missing values ​​or those marked as outliers. Missing or outlier values ​​are imputed based on the weighted average of complete multimodal physiological signal samples.

[0083] In one embodiment, the signal integration module 202 is used to generate a treatment information set based on the first multimodal physiological signal and the second multimodal physiological signal, specifically including: The average values ​​of the first and second multimodal physiological signals at each preset frequency are calculated respectively, and the average values ​​of the multimodal physiological signals are used as signal characterization values. Among them, the signal characterization values ​​include acupoint impedance information characterization values, acupoint temperature information characterization values, and meridian segment impedance information characterization values ​​between acupoints. The signal change amplitude at each preset frequency is calculated based on the signal characterization values ​​before and after treatment; the expression for the signal change amplitude is: ; In the formula, The amplitude of the signal change. The multimodal physiological signal characterization value at a single preset frequency before treatment is applied; The multimodal physiological signal characterization value at the same preset frequency after treatment is applied; A treatment information set is generated based on each signal characterization value and signal change amplitude of the current batch of subjects.

[0084] In one embodiment, the first processing module 203 is used to select the frequency with the smallest signal change in the treatment information set as the core frequency, and calculate the treatment effect of the current batch of subjects based on the signal characterization value at the core frequency, specifically including: Calculate the variance of the signal variation amplitude at each preset frequency, and select the frequency with the smallest variance as the core frequency; The multimodal physiological signal characterization values ​​of each subject in the current batch before and after treatment at the core frequency are obtained one by one, and the difference between the two multimodal physiological signal characterization values ​​is calculated. The average value of the obtained difference is obtained by averaging the difference. The standard deviation of the acquired differences is calculated, and the treatment effect size of the current batch of subjects is calculated based on the mean of the signal representation and the standard deviation of the differences; wherein, the expression for the treatment effect size is: ; In the formula, For effect size, The average value represents the signal. This represents the standard deviation of the difference.

[0085] In one embodiment, calculating the therapeutic effect size of the current batch of subjects based on the signal representation value at the core frequency further includes: performing a paired-samples t-test based on the signal representation value and standard deviation of the current batch of subjects at the core frequency, specifically: A null hypothesis is pre-defined; the null hypothesis is that the treatment received by the current batch of subjects is ineffective. The t-statistic for the current batch of subjects is calculated based on the signal representation value and standard deviation; the expression for the t-statistic is as follows: ; In the formula, For statistical purposes, This represents the number of subjects in the current batch. The corresponding P-value is obtained based on the t-statistic. When the P-value is less than the pre-set significance level, the null hypothesis is deemed invalid, and the changes in multimodal physiological signals before and after treatment are statistically significant.

[0086] In one embodiment, the second processing module 204 is used to acquire the signal change amplitude of the target acupoint at the core frequency and calculate the correlation coefficient between the signal change amplitude and the degree of improvement of clinical symptoms, specifically including: The symptoms of the current batch of subjects before and after treatment were scored using a clinically recognized assessment scale for visceral diseases, and the changes in clinical symptom scores were obtained. The correlation coefficient is calculated based on the amplitude of signal changes and the changes in clinical symptom scores; the expression for the correlation coefficient is as follows: ; In the formula, The correlation coefficient is denoted as n; n is the number of subjects in the current batch. For the first The amplitude of signal changes in each subject; This represents the average amplitude of signal changes in the current batch of subjects; For the first Changes in clinical symptom scores for each subject; This represents the average change in clinical symptom scores of the current batch of subjects. When the correlation coefficient is less than 0 and the p-value is less than the significance level, it is determined that the sensitization state of acupoints is associated with the functional abnormalities of the internal organs.

[0087] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working process of the device described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0088] This invention provides a nested verification device based on multimodal signals. The aforementioned scheme collects multimodal physiological signals, including acupoint impedance, temperature, and meridian segment impedance information at multiple preset frequencies, before and after treatment. This multi-dimensional data acquisition method provides a rich and comprehensive information foundation for subsequent analysis. It closely aligns with the concept of cross-scale verification, building an information bridge from acupoints at the body surface scale and meridians at the body surface channel scale to the internal organs at the deep internal scale. The resulting treatment information set covers signal characterization values ​​at each preset frequency and the amplitude of signal changes after treatment, demonstrating the dynamic changes of acupoints during treatment. Next, the frequency with the smallest signal change is selected as the core frequency to calculate the treatment effect. This approach eliminates interference factors during treatment, allowing the calculated effect to more accurately quantify the amplitude of changes in the subject's clinical symptoms, reflecting the intrinsic connection and resonant conduction mechanism between acupoints, meridians, and internal organs. Subsequently, by acquiring the signal variation amplitude of the target acupoint at the core frequency and calculating its correlation coefficient with the degree of improvement in clinical symptoms, the intrinsic connection between the multimodal physiological signal changes of acupoints and the improvement of clinical symptoms was revealed. This step further deepened the connotation of cross-scale validation. The target acupoint is associated with the internal organs, and its signal variation amplitude at the core frequency reflects the cross-scale information transmission between the surface acupoint and the deep internal organs. Finally, the acupoint status judgment result was generated by comprehensively considering the signal variation amplitude, therapeutic effect size, and correlation coefficient. This process organically combines various key indicators, scientifically and systematically verifying the correlation between acupoint sensitization state and organ function. It can provide objective scientific evidence for the theory of acupoints in traditional Chinese medicine and provide strong support for doctors to accurately judge the acupoint status and formulate personalized treatment plans in clinical practice. Doctors can understand the specific relationship between acupoints, meridians, and internal organs based on the acupoint status judgment results, thereby accurately adjusting treatment strategies and improving the pertinence and effectiveness of treatment.

[0089] The above are merely preferred embodiments of the present invention. It should be noted that those skilled in the art can make several improvements and substitutions without departing from the technical principles of the present invention, and these improvements and substitutions should also be considered within the scope of protection of the present invention.

Claims

1. A nested verification method based on multimodal signals, characterized in that, include: The first multimodal physiological signal of acupoints of the current batch of subjects before treatment and the second multimodal physiological signal after treatment are collected; wherein, the multimodal physiological signal includes acupoint impedance information, acupoint temperature information and meridian segment impedance information between acupoints at multiple preset frequencies; A treatment information set is generated based on the first multimodal physiological signal and the second multimodal physiological signal; wherein, the treatment information set includes the signal representation values ​​of the subject's acupoints at each preset frequency, and the amplitude of signal changes of the subject after the application of treatment; The frequency with the smallest signal change in the treatment information set is selected as the core frequency, and the treatment effect quantity of the current batch of subjects is calculated based on the signal characterization value at the core frequency; wherein, the treatment effect quantity is used to quantify the magnitude of clinical symptom change in the current batch of subjects after treatment. The signal change amplitude of the target acupoint at the core frequency is obtained, and the correlation coefficient between the signal change amplitude and the degree of improvement of clinical symptoms is calculated; wherein, the target acupoint is an acupoint associated with the internal organs, and the correlation coefficient is used to characterize the correlation between the changes in the multimodal physiological signals of the acupoint and the degree of improvement of clinical symptoms. The acupoint status judgment result after treatment is generated based on the signal change amplitude, the therapeutic effect size, and the correlation coefficient; wherein, the acupoint status judgment result is used to characterize the verification result of the correlation between acupoint sensitization state and organ function.

2. The nested verification method based on multimodal signals as described in claim 1, characterized in that, The collection of the first multimodal physiological signals of acupoints of the current batch of subjects before and after treatment, specifically includes: Scanning impedance measurements are performed at each preset frequency for a first preset time to obtain the acupoint impedance information. Temperature information of acupoints is collected during the scanning measurement to obtain the acupoint temperature information. Scanning measurements were performed at each preset frequency to obtain the surface channel impedance between two acupoints on the same meridian, thus obtaining the meridian segment impedance information between the acupoints. The acquisition timestamps of the multimodal physiological signals are obtained, and each subject ID is associated with the corresponding multimodal physiological signal and the corresponding acquisition timestamp.

3. The nested verification method based on multimodal signals as described in claim 1, characterized in that, Before generating the treatment information set based on the first and second multimodal physiological signals, the method further includes data preprocessing of the acquired multimodal physiological signals according to a preset preprocessing strategy. Specifically: Missing values ​​were detected in the multimodal physiological signals, and outlier detection was performed using the interquartile range method. For multimodal physiological signal samples with missing values ​​and multimodal physiological signal samples marked as outliers, calculate their distances to complete multimodal physiological signal samples of the same type, and sort them in ascending order. A predetermined number of complete multimodal physiological signal samples with the highest ranking are selected. Based on the distance between the selected complete multimodal physiological signal samples and the multimodal physiological signal samples with missing values ​​or those marked as outliers, the selected complete multimodal physiological signal samples are grouped and weighted. Missing or outlier values ​​are imputed based on the weighted average of the complete multimodal physiological signal samples.

4. The nested verification method based on multimodal signals as described in claim 1, characterized in that, The step of generating a treatment information set based on the first multimodal physiological signal and the second multimodal physiological signal specifically includes: The average values ​​of the first and second multimodal physiological signals at each preset frequency are calculated respectively, and the average values ​​of the multimodal physiological signals are used as signal characterization values; wherein, the signal characterization values ​​include acupoint impedance information characterization values, acupoint temperature information characterization values, and meridian segment impedance information characterization values ​​between acupoints; The signal change amplitude at each preset frequency is calculated based on the signal characterization values ​​before and after treatment; wherein, the expression for the signal change amplitude is: ； In the formula, The amplitude of the signal change. The multimodal physiological signal characterization value at a single preset frequency before treatment is applied; The multimodal physiological signal characterization value at the same preset frequency after treatment is applied; The treatment information set is generated based on each signal characterization value and signal change amplitude of the current batch of subjects.

5. The nested verification method based on multimodal signals as described in claim 1, characterized in that, The step of selecting the frequency with the smallest signal change in the treatment information set as the core frequency, and calculating the treatment effect of the current batch of subjects based on the signal characterization value at the core frequency, specifically includes: Calculate the variance of the signal variation amplitude at each preset frequency, and select the frequency with the smallest variance as the core frequency; The multimodal physiological signal characterization values ​​of each subject in the current batch before and after treatment at the core frequency are obtained one by one, and the difference between the two multimodal physiological signal characterization values ​​is calculated. The average value of the obtained difference is obtained by averaging the difference. The standard deviation of the acquired differences is calculated, and the treatment effect size of the current batch of subjects is calculated based on the mean of the signal representation and the standard deviation of the differences; wherein, the expression for the treatment effect size is: ； In the formula, For effect size, The average value represents the signal. This represents the standard deviation of the difference.

6. The nested verification method based on multimodal signals as described in claim 5, characterized in that, The calculation of the therapeutic effect size of the current batch of subjects based on the signal characterization value at the core frequency further includes: performing a paired-samples t-test based on the signal characterization value and standard deviation of the current batch of subjects at the core frequency, specifically: A null hypothesis is pre-defined; wherein the null hypothesis is that the treatment received by the current batch of subjects is ineffective; The t-statistic of the current batch of subjects is calculated based on the signal representation value and the standard deviation; wherein, the expression for the t-statistic is: ； In the formula, For statistical purposes, This represents the number of subjects in the current batch. The corresponding P-value is obtained based on the t-statistic. When the P-value is less than the preset significance level, the null hypothesis is determined to be invalid, and the changes in multimodal physiological signals before and after treatment are statistically significant.

7. The nested verification method based on multimodal signals as described in claim 6, characterized in that, The process of acquiring the signal change amplitude of the target acupoint at the core frequency and calculating the correlation coefficient between the signal change amplitude and the degree of improvement in clinical symptoms specifically includes: The symptoms of the current batch of subjects before and after treatment were scored using a clinically recognized assessment scale for visceral diseases, and the changes in clinical symptom scores were obtained. The correlation coefficient is calculated based on the amplitude of the signal change and the change in the clinical symptom score; wherein, the expression for the correlation coefficient is: ； In the formula, The correlation coefficient is denoted as n; n is the number of subjects in the current batch. For the first The amplitude of signal changes in each subject; This represents the average amplitude of signal changes in the current batch of subjects; For the first Changes in clinical symptom scores for each subject; This represents the average change in clinical symptom scores of the current batch of subjects. When the correlation coefficient is less than 0 and the p-value is less than the significance level, it is determined that the sensitization state of the acupoint is associated with the functional abnormality of the internal organs.

8. A nested verification device based on multimodal signals, characterized in that, include: The system comprises a signal acquisition module, a signal integration module, a first processing module, a second processing module, and a result generation module. The signal acquisition module is used to acquire the first multimodal physiological signals of acupoints of the current batch of subjects before the application of treatment and the second multimodal physiological signals after the application of treatment; wherein, the multimodal physiological signals include acupoint impedance information, acupoint temperature information and meridian segment impedance information between acupoints at multiple preset frequencies; The signal integration module is used to generate a treatment information set based on the first multimodal physiological signal and the second multimodal physiological signal; wherein, the treatment information set includes the signal representation values ​​of the subject's acupoints at each preset frequency, and the signal change amplitude of the subject after the application of treatment; The first processing module is used to select the frequency with the smallest signal change in the treatment information set as the core frequency, and calculate the treatment effect quantity of the current batch of subjects based on the signal characterization value at the core frequency; wherein, the treatment effect quantity is used to quantify the magnitude of clinical symptom change in the current batch of subjects after treatment. The second processing module is used to obtain the signal change amplitude of the target acupoint at the core frequency and calculate the correlation coefficient between the signal change amplitude and the degree of improvement of clinical symptoms; wherein, the target acupoint is an acupoint associated with the internal organs, and the correlation coefficient is used to characterize the correlation between the change of multimodal physiological signals of the acupoint and the degree of improvement of clinical symptoms. The result generation module is used to generate the acupoint status judgment result after the current batch of subjects has been treated, based on the signal change amplitude, the therapeutic effect size and the correlation coefficient; wherein, the acupoint status judgment result is used to characterize the correlation verification result between acupoint sensitization state and organ function.

9. A terminal device, characterized in that, It includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein the processor executes the computer program to implement the nested verification method based on multimodal signals as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a stored computer program, wherein, when the computer program is executed, it controls the device on which the computer-readable storage medium is located to perform the nested verification method based on multimodal signals as described in any one of claims 1 to 7.

Images