A non-invasive blood pressure measurement device detection method, apparatus, device, storage medium and program product

Abstract

This disclosure provides a method, apparatus, device, storage medium, and program product for testing non-invasive blood pressure measurement devices. The method includes extracting invasive blood pressure reference value features from test records based on a preset blood pressure assignment model, extracting pulse wave morphological features from test records, fusing the two features to obtain a multi-dimensional fused feature vector, calculating the multi-dimensional fused feature vector based on the blood pressure assignment model to obtain an invasive blood pressure prediction value, performing a numerical judgment on the invasive blood pressure test value based on the invasive blood pressure prediction value to obtain a numerical judgment result, generating the device accuracy status of the non-invasive blood pressure measurement device based on the numerical judgment result, and generating a test report of the non-invasive blood pressure measurement device containing evaluation conclusions according to a preset report template. This achieves personalized accuracy evaluation of non-invasive blood pressure measurement devices, improves evaluation reliability, reduces long-term dependence on invasive measurements, and forms a standardized evaluation closed loop.

Description

Technical Field

[0001] This disclosure relates to the field of equipment testing technology, and in particular to a method, apparatus, equipment, storage medium, and program product for testing non-invasive blood pressure measurement devices. Background Technology

[0002] Invasive blood pressure measurement, which directly acquires blood pressure signals through an intra-arterial catheter, is widely recognized as the most direct and reliable method for measuring blood pressure. In various clinical guidelines and standards, its data reliability is comparable to that of mercury sphygmomanometers, and theoretically, it can serve as a reference standard for testing and calibrating non-invasive blood pressure measurement devices. However, this method is invasive, complex to operate, and requires strict aseptic conditions, and is typically used only as an auxiliary tool in interventional procedures. In clinical practice, limitations such as surgical safety, aseptic technique, and vascular patency make it difficult to simultaneously acquire invasive blood pressure reference values ​​and measurements from the non-invasive blood pressure monitoring device under test, making it difficult to directly apply invasive methods to the clinical accuracy testing of non-invasive blood pressure measurement devices. Summary of the Invention

[0003] To solve the above-mentioned technical problems, or at least partially solve them, this disclosure provides a non-invasive blood pressure measurement device detection method, apparatus, equipment, storage medium, and program product.

[0004] This disclosure provides a method for detecting blood pressure using a non-invasive blood pressure measurement device, the method comprising: Acquire test records from historical time periods and test records from the test period; extract invasive blood pressure reference value features from the test records based on a preset blood pressure assignment model; and extract pulse wave morphological features from the test records. Based on the blood pressure assignment model, the invasive blood pressure reference value features and the pulse wave morphological features are fused to obtain a multi-dimensional fused feature vector. Based on the blood pressure assignment model, the multi-dimensional fused feature vector is calculated to output the invasive blood pressure prediction value for the detection period. Based on the invasive blood pressure prediction value, the non-invasive blood pressure detection value obtained during the detection period is numerically determined to obtain the numerical determination result, and the accuracy status of the non-invasive blood pressure detection device is generated based on the numerical determination result. Based on the accuracy of the device, a test report containing the evaluation conclusions of the non-invasive blood pressure monitoring device is generated according to a preset report template.

[0005] The method provided in this disclosure includes acquiring test records from historical time periods and test records from the test period, extracting invasive blood pressure reference value features from the test records based on a preset blood pressure assignment model, and extracting pulse wave morphological features from the test records, including: The test records and the test records are time-series aligned to generate associated data pairs; The invasive blood pressure waveform data in the test record is preprocessed to obtain a first preprocessing result. Based on the preset peak detection algorithm and the blood pressure assignment model, the systolic blood pressure peak point and diastolic blood pressure trough point in the first preprocessing result are identified. The invasive blood pressure reference value features are extracted based on the systolic blood pressure peak point and diastolic blood pressure trough point. The pulse wave signal in the detection record is subjected to a second preprocessing to obtain a second preprocessing result. Based on the blood pressure assignment model, the pulse wave morphological features are extracted from the second preprocessing result.

[0006] The method provided in this disclosure, based on the blood pressure assignment model, fuses the invasive blood pressure reference value features and the pulse wave morphological features to obtain a multi-dimensional fused feature vector, calculates the multi-dimensional fused feature vector based on the blood pressure assignment model, and outputs the invasive blood pressure prediction value for the detection period, including: Based on the blood pressure assignment model and the associated data pair, the invasive blood pressure reference value features and the pulse wave morphological features are concatenated to obtain a multi-dimensional fused feature vector. Based on the blood pressure assignment model, the invasive blood pressure reference value features in the multi-dimensional fusion feature vector are used to calibrate and map the pulse wave morphology features to obtain the invasive blood pressure prediction waveform data corresponding to the detection period. Based on the invasive blood pressure prediction waveform data, the systolic blood pressure value and diastolic blood pressure value corresponding to the detection period are calculated using the preset peak detection algorithm, and the systolic blood pressure value and diastolic blood pressure value are determined as the invasive blood pressure prediction value for the detection period.

[0007] The method provided in the embodiments of this disclosure, The blood pressure assignment model includes an input layer, a multi-layer feature extraction layer, a feature fusion layer, a fully connected regression layer, and an output layer; Based on the blood pressure assignment model, the multi-dimensional fused feature vector is calculated, and the predicted invasive blood pressure value for the detection period is output, including: The input layer of the blood pressure assignment model obtains the invasive blood pressure waveform data in the test record and the pulse wave signal in the detection record. Based on the multi-layer feature extraction layer of the blood pressure assignment model, the pulse wave signal and the invasive blood pressure waveform data are subjected to hierarchical feature extraction to obtain the pulse wave morphological features and the invasive blood pressure reference value features. Based on the feature fusion layer of the blood pressure assignment model, the pulse wave morphological features and the invasive blood pressure reference value features are fused to obtain a multi-dimensional fused feature vector. Based on the fully connected regression layer of the blood pressure assignment model, the multi-dimensional fused feature vector is subjected to multi-layer nonlinear transformation and regression mapping to obtain invasive blood pressure prediction waveform data. Based on the output layer of the blood pressure assignment model, the invasive blood pressure prediction waveform data is output.

[0008] The method provided in this disclosure embodiment performs numerical determination on the non-invasive blood pressure detection value obtained during the detection period based on the invasive blood pressure prediction value, obtains a numerical determination result, and generates the device accuracy status of the non-invasive blood pressure detection device based on the numerical determination result, including: Non-invasive blood pressure readings were obtained based on the aforementioned detection period. Based on the invasive blood pressure prediction value, it is compared with the non-invasive blood pressure detection value, and the difference in measured blood pressure is calculated according to the comparison result; Obtain the preset blood pressure error threshold and blood pressure error standard deviation threshold, calculate the mean and standard deviation of the measured blood pressure difference based on the measured blood pressure difference during the detection period, and extract the measured blood pressure difference with the largest absolute difference in the calculation results as the maximum measured blood pressure difference; If the mean of the measured blood pressure difference is less than or equal to the blood pressure error threshold, and the standard deviation of the measured blood pressure difference is less than or equal to the blood pressure error standard deviation threshold, then the numerical determination result is passed. If the mean of the measured blood pressure difference is greater than the blood pressure error threshold, or the standard deviation of the measured blood pressure difference is greater than the blood pressure error standard deviation threshold, then the numerical judgment result is "not passed". If a preset maximum measurement error threshold is obtained, and the numerical judgment result is passed, and the maximum measurement difference is less than or equal to the maximum measurement error threshold, then the non-invasive blood pressure detection device is accurate. If the numerical result is "fail", then the non-invasive blood pressure monitoring device is inaccurate. The accuracy of the equipment includes both accurate and inaccurate equipment.

[0009] The method provided in this disclosure, based on the accuracy of the device, generates a test report for a non-invasive blood pressure monitoring device containing evaluation conclusions according to a preset report template, including: If the device is in good condition, then the first report template in the preset report template library is invoked; If the device is inaccurate, then the second report template in the preset report template library is invoked. The second report template includes an area indicating the reason for non-compliance.

[0010] This disclosure also provides a non-invasive blood pressure measurement device, the device comprising: The acquisition module is used to acquire the test records of historical time periods and the test records of the test period, extract invasive blood pressure reference value features from the test records based on a preset blood pressure assignment model, and extract pulse wave morphological features from the test records. The prediction module is used to obtain a multi-dimensional fusion feature vector by fusing the invasive blood pressure reference value features and the pulse wave morphology features based on the blood pressure assignment model, calculate the multi-dimensional fusion feature vector based on the blood pressure assignment model, and output the invasive blood pressure prediction value for the detection period. The determination module is used to make a numerical determination on the non-invasive blood pressure detection value obtained during the detection period based on the invasive blood pressure prediction value, obtain a numerical determination result, and generate the device accuracy status of the non-invasive blood pressure detection device based on the numerical determination result. The generation module is used to generate a test report for the non-invasive blood pressure monitoring device, which includes evaluation conclusions, according to the accuracy of the device and a preset report template.

[0011] The apparatus provided in this disclosure, wherein the acquisition module is specifically used for: The test records and the test records are time-series aligned to generate associated data pairs; The invasive blood pressure waveform data in the test record is preprocessed to obtain a first preprocessing result. Based on the preset peak detection algorithm and the blood pressure assignment model, the systolic blood pressure peak point and diastolic blood pressure trough point in the first preprocessing result are identified. The invasive blood pressure reference value features are extracted based on the systolic blood pressure peak point and diastolic blood pressure trough point. The pulse wave signal in the detection record is subjected to a second preprocessing to obtain a second preprocessing result. Based on the blood pressure assignment model, the pulse wave morphological features are extracted from the second preprocessing result.

[0012] The apparatus provided in this disclosure, wherein the prediction module is specifically used for: Based on the blood pressure assignment model and the associated data pair, the invasive blood pressure reference value features and the pulse wave morphological features are concatenated to obtain a multi-dimensional fused feature vector. Based on the blood pressure assignment model, the invasive blood pressure reference value features in the multi-dimensional fusion feature vector are used to calibrate and map the pulse wave morphology features to obtain the invasive blood pressure prediction waveform data corresponding to the detection period. Based on the invasive blood pressure prediction waveform data, the systolic blood pressure value and diastolic blood pressure value corresponding to the detection period are calculated using the preset peak detection algorithm, and the systolic blood pressure value and diastolic blood pressure value are determined as the invasive blood pressure prediction value for the detection period.

[0013] The apparatus provided in this disclosure, wherein the acquisition module is specifically used for: The blood pressure assignment model includes an input layer, a multi-layer feature extraction layer, a feature fusion layer, a fully connected regression layer, and an output layer; The input layer of the blood pressure assignment model obtains the invasive blood pressure waveform data in the test record and the pulse wave signal in the detection record. Based on the multi-layer feature extraction layer of the blood pressure assignment model, the pulse wave signal and the invasive blood pressure waveform data are subjected to hierarchical feature extraction to obtain the pulse wave morphological features and the invasive blood pressure reference value features. Based on the feature fusion layer of the blood pressure assignment model, the pulse wave morphological features and the invasive blood pressure reference value features are fused to obtain a multi-dimensional fused feature vector. Based on the fully connected regression layer of the blood pressure assignment model, the multi-dimensional fused feature vector is subjected to multi-layer nonlinear transformation and regression mapping to obtain invasive blood pressure prediction waveform data. Based on the output layer of the blood pressure assignment model, the invasive blood pressure prediction waveform data is output.

[0014] The apparatus provided in this disclosure, wherein the determination module is specifically used for: Non-invasive blood pressure readings were obtained based on the aforementioned detection period. Based on the invasive blood pressure prediction value, it is compared with the non-invasive blood pressure detection value, and the difference in measured blood pressure is calculated according to the comparison result; Obtain the preset blood pressure error threshold and blood pressure error standard deviation threshold, calculate the mean and standard deviation of the measured blood pressure difference based on the measured blood pressure difference during the detection period, and extract the measured blood pressure difference with the largest absolute difference in the calculation results as the maximum measured blood pressure difference; If the mean of the measured blood pressure difference is less than or equal to the blood pressure error threshold, and the standard deviation of the measured blood pressure difference is less than or equal to the blood pressure error standard deviation threshold, then the numerical determination result is passed. If the mean of the measured blood pressure difference is greater than the blood pressure error threshold, or the standard deviation of the measured blood pressure difference is greater than the blood pressure error standard deviation threshold, then the numerical judgment result is "not passed". If a preset maximum measurement error threshold is obtained, and the numerical judgment result is passed, and the maximum measurement difference is less than or equal to the maximum measurement error threshold, then the non-invasive blood pressure detection device is accurate. If the numerical result is "fail", then the non-invasive blood pressure monitoring device is inaccurate. The accuracy of the equipment includes both accurate and inaccurate equipment.

[0015] The apparatus provided in this disclosure, wherein the generation module is specifically used for: If the device is in good condition, then the first report template in the preset report template library is invoked; If the device is inaccurate, then the second report template in the preset report template library is invoked. The second report template includes an area indicating the reason for non-compliance.

[0016] This disclosure also provides an electronic device, the electronic device comprising: a processor; a memory for storing executable instructions of the processor; the processor being configured to read the executable instructions from the memory and execute the instructions to implement the non-invasive blood pressure measurement device detection method provided in this disclosure.

[0017] This disclosure also provides a computer-readable storage medium storing a computer program for executing the non-invasive blood pressure measurement device detection method provided in this disclosure.

[0018] The technical solution provided in this disclosure has the following advantages compared with the prior art: The non-invasive blood pressure measurement device detection method provided in this disclosure extracts invasive blood pressure reference value features from test records based on a blood pressure assignment model, extracts pulse wave morphological features from test records, fuses the invasive blood pressure reference value features and pulse wave morphological features to obtain a multi-dimensional fused feature vector, and calculates and outputs invasive blood pressure predicted values. The accuracy of the non-invasive blood pressure detection value is evaluated using the invasive blood pressure predicted values, the accuracy of the non-invasive blood pressure measurement device is determined, and a test report is output. This method enables personalized accuracy evaluation of non-invasive blood pressure measurement devices, improves evaluation reliability, reduces long-term reliance on invasive measurements, and forms a standardized evaluation closed loop. Attached Figure Description

[0019] The above and other features, advantages, and aspects of the embodiments of this disclosure will become more apparent from the accompanying drawings and the following detailed description. Throughout the drawings, the same or similar reference numerals denote the same or similar elements. It should be understood that the drawings are schematic, and the originals and elements are not necessarily drawn to scale.

[0020] Figure 1 A schematic flowchart illustrating the non-invasive blood pressure measurement device testing method provided in this embodiment of the disclosure; Figure 2 This is a schematic diagram of the structure of the non-invasive blood pressure measurement device provided in the embodiments of this disclosure; Figure 3 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this disclosure. Detailed Implementation

[0021] Embodiments of this disclosure will now be described in more detail with reference to the accompanying drawings. While some embodiments of this disclosure are shown in the drawings, it should be understood that this disclosure can be implemented in various forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided to provide a more thorough and complete understanding of this disclosure. It should be understood that the accompanying drawings and embodiments of this disclosure are for illustrative purposes only and are not intended to limit the scope of protection of this disclosure.

[0022] It should be understood that the steps described in the method embodiments of this disclosure may be performed in different orders and / or in parallel. Furthermore, the method embodiments may include additional steps and / or omit the steps shown. The scope of this disclosure is not limited in this respect.

[0023] The term "comprising" and its variations as used herein are open-ended inclusions, meaning "including but not limited to". The term "based on" means "at least partially based on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments". Definitions of other terms will be given in the description below.

[0024] It should be noted that the concepts of "first" and "second" mentioned in this disclosure are used only to distinguish different devices, modules or units, and are not used to limit the order of functions performed by these devices, modules or units or their interdependencies.

[0025] It should be noted that the terms "a" and "a plurality of" used in this disclosure are illustrative rather than restrictive, and those skilled in the art should understand that, unless otherwise expressly indicated in the context, they should be understood as "one or more".

[0026] The names of messages or information exchanged between multiple devices in the embodiments of this disclosure are for illustrative purposes only and are not intended to limit the scope of such messages or information.

[0027] To address the aforementioned issues, this disclosure provides a method for detecting non-invasive blood pressure measurement devices. The method will be described below with reference to specific embodiments.

[0028] Figure 1 This is a flowchart illustrating a non-invasive blood pressure measurement device detection method provided in an embodiment of the present disclosure. The method can be executed by a non-invasive blood pressure measurement device detection device, which can be implemented using software and / or hardware and is generally integrated into an electronic device.

[0029] Example 1: A non-invasive blood pressure measurement device detection method provided in this embodiment includes: S101: Obtain the test records of historical time periods and the test records of the test period, extract invasive blood pressure reference value features from the test records based on a preset blood pressure assignment model, and extract pulse wave morphological features from the test records. S102: Based on the blood pressure assignment model, a multi-dimensional fusion feature vector is obtained by fusing the invasive blood pressure reference value features and the pulse wave morphological features. The multi-dimensional fusion feature vector is calculated based on the blood pressure assignment model, and the invasive blood pressure prediction value for the detection period is output. S103: Based on the invasive blood pressure prediction value, the non-invasive blood pressure detection value obtained during the detection period is numerically determined to obtain the numerical determination result, and the accuracy status of the non-invasive blood pressure detection device is generated according to the numerical determination result. S104: Based on the accuracy of the device, generate a test report for the non-invasive blood pressure monitoring device containing evaluation conclusions according to a preset report template.

[0030] In this embodiment, the historical time period refers to the time period before the detection time period, and the detection time period is the time period in which the pulse wave signal is currently collected. The latest state of the non-invasive blood pressure measurement device is determined based on the detection time period. The test record refers to the data set containing invasive blood pressure waveforms and corresponding timestamps acquired within the historical time period. The detection record refers to the data set containing pulse wave signals and timestamps acquired in real time within the detection time period. The blood pressure assignment model has two input and output stages. The first stage inputs the test record and the detection record, and outputs invasive blood pressure reference value features and pulse wave morphological features. The second stage inputs a multi-dimensional fusion feature vector of invasive blood pressure reference value features and pulse wave morphological features, and outputs the invasive blood pressure prediction value corresponding to the detection stage. The invasive blood pressure reference value features are the systolic blood pressure value, diastolic blood pressure value, and pressure waveform envelope area extracted from the test record. Historical real blood pressure status is provided for calibrating the pulse wave morphological features. The pulse wave morphological features are the pulse wave conduction time, peak height, trough height, rising limb slope, and waveform area extracted from the detection record. It represents the pulse wave morphology during the current detection period, providing data for the model to fuse features with reference values ​​in calculations.

[0031] In this embodiment, the multi-dimensional fused feature vector is a combined feature formed by concatenating invasive blood pressure reference value features and pulse wave morphological features along the channel dimension. The fusion process involves concatenating the two types of features end-to-end along the channel dimension in the feature fusion layer of the blood pressure replication model. The calculation process involves the fully connected regression layer of the blood pressure assignment model performing multi-layer nonlinear transformations and regression mappings on the fused feature vector, transforming the multi-dimensional fused feature vector into continuous blood pressure waveform data, thus completing the mapping from features to blood pressure values. The invasive blood pressure prediction value is the systolic and diastolic blood pressure values ​​output by the blood pressure assignment model during the detection period. It serves as a high-precision reference benchmark in the evaluation of non-invasive devices, replacing the actual invasive measurement values. The invasive blood pressure prediction value serves as a reference standard for comparing non-invasive blood pressure detection values, quantifying and determining the measurement difference and accuracy level of non-invasive blood pressure detection devices, and generating the device accuracy status.

[0032] In this embodiment, numerical judgment involves comparing non-invasive blood pressure readings with invasive blood pressure predictions, calculating the difference between the measured blood pressure values, and then comparing the mean and standard deviation of the deviation with a preset threshold to quantitatively assess whether the measurement error of the non-invasive device is within an acceptable range. The numerical judgment result includes two states: pass or fail. Pass indicates that neither the mean nor the standard deviation exceeds the corresponding threshold, while fail indicates that at least one exceeds the threshold, providing a prerequisite for subsequent determination of device accuracy. Device accuracy is the final conclusion generated by combining the numerical judgment result with the comparison result of the maximum measurement difference, including two states: accurate and inaccurate. This directly characterizes whether the non-invasive blood pressure monitoring device meets the accuracy requirements. First, the numerical judgment is performed to determine pass or fail. Then, the maximum measurement difference is obtained and compared with the maximum measurement error threshold. If the numerical judgment passes and the maximum deviation does not exceed the threshold, the device is considered accurate; otherwise, the device is considered inaccurate.

[0033] In this embodiment, the preset report template is a pre-built report format framework that includes a basic device information area, a detailed indicator data area, and an evaluation conclusion, providing a unified format for generating standardized test reports. The test report is a complete report file generated after filling in the relevant data and evaluation conclusions. It records all process data and final conclusions of the non-invasive blood pressure monitoring device accuracy evaluation, providing traceable evaluation evidence. The evaluation conclusion is a textual conclusion generated based on the device's accuracy status, including whether the device accuracy meets the standards or not, and can indicate the reasons for non-compliance. It can intuitively present the device performance judgment results, providing a basis for clinical decision-making or device calibration.

[0034] The working principle and beneficial effects of this embodiment are as follows: Invasive blood pressure reference features are extracted from test records based on a blood pressure assignment model; pulse wave morphological features are extracted from test records; the invasive blood pressure reference features and pulse wave morphological features are fused to obtain a multi-dimensional fused feature vector; and the invasive blood pressure predicted value is calculated and output. The accuracy of the non-invasive blood pressure detection value is evaluated using the invasive blood pressure predicted value, the accuracy of the non-invasive blood pressure measurement device is determined, and a test report is output. This achieves personalized accuracy evaluation of non-invasive blood pressure detection devices, improves evaluation reliability, reduces long-term dependence on invasive measurements, and forms a standardized evaluation closed loop.

[0035] Example 2: The method provided in this embodiment of the present disclosure acquires test records from historical time periods and test records from the test period, extracts invasive blood pressure reference value features from the test records based on a preset blood pressure assignment model, and extracts pulse wave morphological features from the test records, including: The test records and the test records are time-series aligned to generate associated data pairs; The invasive blood pressure waveform data in the test record is preprocessed to obtain a first preprocessing result. Based on the preset peak detection algorithm and the blood pressure assignment model, the systolic blood pressure peak point and diastolic blood pressure trough point in the first preprocessing result are identified. The invasive blood pressure reference value features are extracted based on the systolic blood pressure peak point and diastolic blood pressure trough point. The pulse wave signal in the detection record is subjected to a second preprocessing to obtain a second preprocessing result. Based on the blood pressure assignment model, the pulse wave morphological features are extracted from the second preprocessing result.

[0036] In this embodiment, time alignment processing is an operation that confirms the continuous adjacency of the end time of the test record and the start time of the detection record through timestamp matching. Based on time alignment processing, a temporal causal relationship is established between the two data segments, providing a foundation for generating associated data pairs. Associated data pairs are paired data combinations formed after time alignment of the test record and the detection record, binding historical invasive waveforms and current pulse wave signals into a continuous sequence from the same source for subsequent feature extraction and fusion. The first preprocessing involves bandpass filtering and baseline calibration of the invasive blood pressure waveform in the test record. This first preprocessing eliminates high-frequency noise and baseline drift interference, improving the accuracy of peak point identification. The result of the first preprocessing is clean invasive blood pressure waveform data output after the first preprocessing. The preset peak detection algorithm is an algorithm that automatically locates peaks and troughs based on waveform morphology features, accurately identifying key pressure points from the preprocessed invasive waveform, providing a location basis for extracting reference value features. The systolic blood pressure peak and diastolic blood pressure trough are the highest and lowest pressure points of the invasive blood pressure waveform within a cardiac cycle (the detection period provides the pulse wave morphology of several consecutive cardiac cycles). These extreme positions of the pressure waveform are marked and used to extract the systolic and diastolic blood pressure values. The second preprocessing involves bandpass filtering and motion artifact removal of the pulse wave signal in the detection record. This preprocessing eliminates high-frequency interference and body motion noise. The result of the second preprocessing is clean pulse wave signal data, which serves as the input for pulse wave morphological feature extraction, ensuring that the extracted morphological parameters reflect the true physiological state.

[0037] The working principle and beneficial effects of this embodiment are as follows: Time-series alignment of the test records and detection records generates associated data pairs; after preprocessing the waveforms in the test records, peak and trough points are identified using a peak detection algorithm to extract invasive blood pressure reference value features; after preprocessing the pulse wave signals in the detection records, pulse wave morphological features are extracted for model fusion and calibration; data association is established through time-series alignment to ensure the temporal consistency of feature pairing; targeted preprocessing strategies are adopted to accurately extract invasive reference value features and pulse wave morphological features, providing the model with high-quality personalized prior knowledge and current detection features, thereby improving the reliability of subsequent prediction and evaluation.

[0038] Example 3: The method provided in this embodiment of the present disclosure, based on the blood pressure assignment model, fuses the invasive blood pressure reference value features and the pulse wave morphological features to obtain a multi-dimensional fused feature vector, calculates the multi-dimensional fused feature vector based on the blood pressure assignment model, and outputs the invasive blood pressure prediction value for the detection period, including: Based on the blood pressure assignment model and the associated data pair, the invasive blood pressure reference value features and the pulse wave morphological features are concatenated to obtain a multi-dimensional fused feature vector. Based on the blood pressure assignment model, the invasive blood pressure reference value features in the multi-dimensional fusion feature vector are used to calibrate and map the pulse wave morphology features to obtain the invasive blood pressure prediction waveform data corresponding to the detection period. Based on the invasive blood pressure prediction waveform data, the systolic blood pressure value and diastolic blood pressure value corresponding to the detection period are calculated using the preset peak detection algorithm, and the systolic blood pressure value and diastolic blood pressure value are determined as the invasive blood pressure prediction value for the detection period.

[0039] In this embodiment, feature splicing is an operation that connects the invasive blood pressure reference value features and pulse wave morphology features end-to-end along the feature channel dimension. By combining historical invasive reference information and current pulse wave morphology information through feature splicing, a multi-dimensional fused feature vector is formed, providing a complete input representation for the model. Calibration mapping is a process of dynamically correcting and mapping the pulse wave morphology features based on the invasive blood pressure reference value features in the fused feature vector through nonlinear transformations within the model. This transforms the pulse wave morphology features into a blood pressure waveform corresponding to invasive pressure, achieving personalized calibration. The invasive blood pressure prediction waveform data is a continuous blood pressure waveform sequence generated after calibration mapping, containing complete systolic and diastolic pressure change curves. It serves as input to the peak detection algorithm, from which specific values ​​of systolic and diastolic pressure are extracted, providing a quantifiable reference benchmark for subsequent evaluation.

[0040] The working principle and beneficial effects of this embodiment are as follows: Based on the associated data pairs, invasive blood pressure reference value features and pulse wave morphological features are spliced ​​together to form a multi-dimensional fused feature vector. The pulse wave morphological features are calibrated and mapped based on the invasive blood pressure reference value features to generate invasive blood pressure prediction waveform data. The systolic blood pressure value and diastolic blood pressure value are extracted from the waveform through a preset peak detection algorithm and determined as the invasive blood pressure prediction value for the detection period. The fused feature vector contains both historical invasive reference values ​​and current pulse wave information to achieve personalized calibration mapping. The invasive blood pressure prediction value is determined through the two-step output of waveform generation and peak detection, which improves the prediction accuracy of systolic and diastolic blood pressure and provides a highly reliable quantitative benchmark for the accuracy evaluation of non-invasive devices.

[0041] Example 4: The method provided in this embodiment of the present disclosure, The blood pressure assignment model includes an input layer, a multi-layer feature extraction layer, a feature fusion layer, a fully connected regression layer, and an output layer; Based on the blood pressure assignment model, the multi-dimensional fused feature vector is calculated, and the predicted invasive blood pressure value for the detection period is output, including: The input layer of the blood pressure assignment model obtains the invasive blood pressure waveform data in the test record and the pulse wave signal in the detection record. Based on the multi-layer feature extraction layer of the blood pressure assignment model, the pulse wave signal and the invasive blood pressure waveform data are subjected to hierarchical feature extraction to obtain the pulse wave morphological features and the invasive blood pressure reference value features. Based on the feature fusion layer of the blood pressure assignment model, the pulse wave morphological features and the invasive blood pressure reference value features are fused to obtain a multi-dimensional fused feature vector. Based on the fully connected regression layer of the blood pressure assignment model, the multi-dimensional fused feature vector is subjected to multi-layer nonlinear transformation and regression mapping to obtain invasive blood pressure prediction waveform data. Based on the output layer of the blood pressure assignment model, the invasive blood pressure prediction waveform data is output.

[0042] In this embodiment, the data flow within the blood pressure assignment model is as follows: input layer → multi-layer feature extraction layer → feature fusion layer → fully connected regression layer → output layer.

[0043] The working principle and beneficial effects of this embodiment are as follows: by using the progressive structure of hierarchical feature extraction, feature fusion and regression mapping, and utilizing the complementary information of invasive blood pressure waveform data and pulse wave signal, end-to-end learning from the original waveform to the predicted waveform is achieved, thereby improving the hierarchy of feature extraction and mapping accuracy, and providing model support for high-precision blood pressure assignment.

[0044] Example 5: The method provided in this embodiment of the present disclosure performs numerical determination on the non-invasive blood pressure detection value obtained during the detection period based on the invasive blood pressure prediction value, obtains the numerical determination result, and generates the device accuracy status of the non-invasive blood pressure detection device based on the numerical determination result, including: Non-invasive blood pressure readings were obtained based on the aforementioned detection period. Based on the invasive blood pressure prediction value, it is compared with the non-invasive blood pressure detection value, and the difference in measured blood pressure is calculated according to the comparison result; Obtain the preset blood pressure error threshold and blood pressure error standard deviation threshold, calculate the mean and standard deviation of the measured blood pressure difference based on the measured blood pressure difference during the detection period, and extract the measured blood pressure difference with the largest absolute difference in the calculation results as the maximum measured blood pressure difference; If the mean of the measured blood pressure difference is less than or equal to the blood pressure error threshold, and the standard deviation of the measured blood pressure difference is less than or equal to the blood pressure error standard deviation threshold, then the numerical determination result is passed. If the mean of the measured blood pressure difference is greater than the blood pressure error threshold, or the standard deviation of the measured blood pressure difference is greater than the blood pressure error standard deviation threshold, then the numerical judgment result is "not passed". If a preset maximum measurement error threshold is obtained, and the numerical judgment result is passed, and the maximum measurement difference is less than or equal to the maximum measurement error threshold, then the non-invasive blood pressure detection device is accurate. If the numerical result is "fail", then the non-invasive blood pressure monitoring device is inaccurate. The accuracy of the equipment includes both accurate and inaccurate equipment.

[0045] In this embodiment, the measured blood pressure difference is the difference between the predicted invasive blood pressure and the measured non-invasive blood pressure at each moment during the detection period. The blood pressure error threshold is a preset upper limit for judging the mean of the measured blood pressure difference. The blood pressure error standard deviation threshold is a preset upper limit for judging the standard deviation of the measured blood pressure difference. The blood pressure error standard deviation threshold is compared with the standard deviation of the measured blood pressure difference to determine the dispersion of the non-invasive device's measurement results; exceeding the threshold indicates insufficient measurement stability. The mean and standard deviation of the measured blood pressure difference are the average and standard deviation obtained by statistically analyzing all measured blood pressure differences within the detection period. The maximum measurement difference is the largest absolute value among all measured blood pressure differences. The maximum measurement error threshold is a preset upper limit for judging the maximum measurement difference. Based on the ratio of the maximum measurement error threshold to the maximum measurement difference, it is ensured that the device does not exhibit a significant deviation exceeding the safety threshold at any time, and a final rejection judgment is executed.

[0046] In this embodiment, the blood pressure difference is used as the basic data. The mean and standard deviation of the blood pressure difference reflect the systematic and random errors of the non-invasive blood pressure measurement device. The blood pressure error threshold and the blood pressure error standard deviation threshold can provide statistical judgment criteria. The maximum measurement deviation and the corresponding threshold are used to make the final judgment, ensuring that the average performance of the device meets the standards and there are no single point of huge error, thus comprehensively judging the accuracy of the device.

[0047] The working principle and beneficial effects of this embodiment are as follows: Based on the invasive blood pressure prediction value, the difference between the measured blood pressure and the non-invasive blood pressure detection value is calculated. The mean, standard deviation, and blood pressure error threshold of the deviation values ​​are compared for numerical judgment. Finally, the maximum measurement difference and the maximum measurement error threshold are used for ultimate judgment, comprehensively determining the accuracy of the equipment. This triple-indicator progressive judgment of mean, standard deviation, and maximum measurement difference assesses both the overall accuracy and stability of the equipment and captures extreme deviations, avoiding the risk of good average performance but fatal single errors, thus ensuring the comprehensiveness and safety of the equipment accuracy assessment.

[0048] Example 6: The method provided in this embodiment of the present disclosure generates a test report for a non-invasive blood pressure monitoring device, including evaluation conclusions, according to the accuracy of the device and a preset report template, comprising: If the device is in good condition, then the first report template in the preset report template library is invoked; If the device is inaccurate, then the second report template in the preset report template library is invoked. The second report template includes an area indicating the reason for non-compliance.

[0049] In this embodiment, the first report template is the report format framework invoked when the equipment is accurate. It includes a basic equipment information area, a detailed indicator data area, and a compliance assessment conclusion area. When the equipment passes all the assessments, a standardized compliance report is generated, intuitively presenting the equipment's performance compliance status. The second report template is the report format framework invoked when the equipment is inaccurate. It additionally includes a non-compliance reason annotation area. When the equipment fails the assessment, a differentiated report containing problem location information is generated, facilitating the tracing of the root cause of the non-compliance. The preset report template library is a set of templates storing the first and second report templates. It dynamically matches the corresponding template based on the equipment's accuracy, achieving automated and standardized report generation. The non-compliance reason annotation area is a reserved fillable field in the second report template, used to annotate specific reasons such as exceeding the mean limit, standard deviation limit, or maximum deviation limit. This can intuitively display the specific root cause of the equipment's non-compliance, providing a clear direction for subsequent calibration or repair.

[0050] The working principle and beneficial effects of this embodiment are as follows: Based on the determination of the accuracy of the equipment, the corresponding template is dynamically retrieved from the preset report template library; when the equipment is accurate, the first report template is called; when the equipment is inaccurate, the second report template containing the area indicating the reason for non-compliance is called, generating a differentiated test report. This achieves a clear presentation of the evaluation conclusion. When the equipment is inaccurate, the root cause of non-compliance is intuitively displayed through the reason annotation area, facilitating rapid problem location and improving the information completeness and clinical operability of the test report.

[0051] To achieve the above embodiments, this disclosure also proposes a non-invasive blood pressure measurement device testing device.

[0052] Figure 2 This is a schematic diagram of the structure of a non-invasive blood pressure measurement device provided in an embodiment of this disclosure. The device 200 can be implemented by software and / or hardware, and is generally integrated into an electronic device. For example... Figure 2 As shown, the device 200 includes: an acquisition module 201, a prediction module 202, a determination module 203, and a generation module 204, wherein, The acquisition module 201 is used to acquire the test records of historical time periods and the test records of the test period, extract invasive blood pressure reference value features from the test records based on a preset blood pressure assignment model, and extract pulse wave morphological features from the test records. Prediction module 202 is used to obtain a multi-dimensional fusion feature vector by fusing the invasive blood pressure reference value features and the pulse wave morphology features based on the blood pressure assignment model, calculate the multi-dimensional fusion feature vector based on the blood pressure assignment model, and output the invasive blood pressure prediction value for the detection period. The determination module 203 is used to make a numerical determination on the non-invasive blood pressure detection value obtained during the detection period based on the invasive blood pressure prediction value, obtain a numerical determination result, and generate the device accuracy status of the non-invasive blood pressure detection device based on the numerical determination result. The generation module 204 is used to generate a test report for the non-invasive blood pressure monitoring device, which includes evaluation conclusions, according to the accuracy of the device and a preset report template.

[0053] The apparatus provided in this disclosure, wherein the acquisition module 201 is specifically used for: The test records and the test records are time-series aligned to generate associated data pairs; The invasive blood pressure waveform data in the test record is preprocessed to obtain a first preprocessing result. Based on the preset peak detection algorithm and the blood pressure assignment model, the systolic blood pressure peak point and diastolic blood pressure trough point in the first preprocessing result are identified. The invasive blood pressure reference value features are extracted based on the systolic blood pressure peak point and diastolic blood pressure trough point. The pulse wave signal in the detection record is subjected to a second preprocessing to obtain a second preprocessing result. Based on the blood pressure assignment model, the pulse wave morphological features are extracted from the second preprocessing result.

[0054] The apparatus provided in this disclosure, wherein the prediction module 202 is specifically used for: Based on the blood pressure assignment model and the associated data pair, the invasive blood pressure reference value features and the pulse wave morphological features are concatenated to obtain a multi-dimensional fused feature vector. Based on the blood pressure assignment model, the invasive blood pressure reference value features in the multi-dimensional fusion feature vector are used to calibrate and map the pulse wave morphology features to obtain the invasive blood pressure prediction waveform data corresponding to the detection period. Based on the invasive blood pressure prediction waveform data, the systolic blood pressure value and diastolic blood pressure value corresponding to the detection period are calculated using the preset peak detection algorithm, and the systolic blood pressure value and diastolic blood pressure value are determined as the invasive blood pressure prediction value for the detection period.

[0055] The apparatus provided in this disclosure, wherein the acquisition module 201 is specifically used for: The blood pressure assignment model includes an input layer, a multi-layer feature extraction layer, a feature fusion layer, a fully connected regression layer, and an output layer; The input layer of the blood pressure assignment model obtains the invasive blood pressure waveform data in the test record and the pulse wave signal in the detection record. Based on the multi-layer feature extraction layer of the blood pressure assignment model, the pulse wave signal and the invasive blood pressure waveform data are subjected to hierarchical feature extraction to obtain the pulse wave morphological features and the invasive blood pressure reference value features. Based on the feature fusion layer of the blood pressure assignment model, the pulse wave morphological features and the invasive blood pressure reference value features are fused to obtain a multi-dimensional fused feature vector. Based on the fully connected regression layer of the blood pressure assignment model, the multi-dimensional fused feature vector is subjected to multi-layer nonlinear transformation and regression mapping to obtain invasive blood pressure prediction waveform data. Based on the output layer of the blood pressure assignment model, the invasive blood pressure prediction waveform data is output.

[0056] The apparatus provided in this disclosure, wherein the determination module 203 is specifically used for: Non-invasive blood pressure readings were obtained based on the aforementioned detection period. Based on the invasive blood pressure prediction value, it is compared with the non-invasive blood pressure detection value, and the difference in measured blood pressure is calculated according to the comparison result; Obtain the preset blood pressure error threshold and blood pressure error standard deviation threshold, calculate the mean and standard deviation of the measured blood pressure difference based on the measured blood pressure difference during the detection period, and extract the measured blood pressure difference with the largest absolute difference in the calculation results as the maximum measured blood pressure difference; If the mean of the measured blood pressure difference is less than or equal to the blood pressure error threshold, and the standard deviation of the measured blood pressure difference is less than or equal to the blood pressure error standard deviation threshold, then the numerical determination result is passed. If the mean of the measured blood pressure difference is greater than the blood pressure error threshold, or the standard deviation of the measured blood pressure difference is greater than the blood pressure error standard deviation threshold, then the numerical judgment result is "not passed". If a preset maximum measurement error threshold is obtained, and the numerical judgment result is passed, and the maximum measurement difference is less than or equal to the maximum measurement error threshold, then the non-invasive blood pressure detection device is accurate. If the numerical result is "fail", then the non-invasive blood pressure monitoring device is inaccurate. The accuracy of the equipment includes both accurate and inaccurate equipment.

[0057] The apparatus provided in this disclosure embodiment, wherein the generation module 204 is specifically used for: If the device is in good condition, then the first report template in the preset report template library is invoked; If the device is inaccurate, then the second report template in the preset report template library is invoked. The second report template includes an area indicating the reason for non-compliance.

[0058] The non-invasive blood pressure measurement device testing device provided in this disclosure can execute the non-invasive blood pressure measurement device testing method provided in any embodiment of this disclosure, and has the corresponding functional modules and beneficial effects for executing the method.

[0059] To implement the above embodiments, this disclosure also proposes a computer program product, including a computer program / instructions, which, when executed by a processor, implements the non-invasive blood pressure measurement device detection method in the above embodiments.

[0060] Figure 3 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this disclosure.

[0061] The following is a detailed reference. Figure 3 The diagram illustrates a structural schematic suitable for implementing the electronic device in the embodiments of this disclosure. The electronic device in the embodiments of this disclosure may include, but is not limited to, mobile terminals such as mobile phones, laptops, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), in-vehicle terminals (e.g., in-vehicle navigation terminals), and fixed terminals such as digital TVs and desktop computers. Figure 3 The electronic device shown is merely an example and should not be construed as limiting the functionality and scope of the embodiments disclosed herein.

[0062] like Figure 3 As shown, the electronic device may include a processor (e.g., a central processing unit, a graphics processing unit, etc.) 301, which can perform various appropriate actions and processes according to a program stored in read-only memory (ROM) 302 or a program loaded from memory 308 into random access memory (RAM) 303. The RAM 303 also stores various programs and data required for the operation of the electronic device. The processor 301, ROM 302, and RAM 303 are interconnected via a bus 304. An input / output (I / O) interface 305 is also connected to the bus 304.

[0063] Typically, the following devices can be connected to I / O interface 305: input devices 306 including, for example, touchscreens, touchpads, keyboards, mice, cameras, microphones, accelerometers, gyroscopes, etc.; output devices 307 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; memory devices 308 including, for example, magnetic tapes, hard disks, etc.; and communication devices 309. Communication device 309 allows electronic devices to communicate wirelessly or wiredly with other devices to exchange data. Although Figure 3Electronic devices with various devices are shown, but it should be understood that it is not required to implement or have all of the devices shown. More or fewer devices may be implemented or have alternatively.

[0064] In particular, according to embodiments of this disclosure, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of this disclosure include a computer program product comprising a computer program carried on a non-transitory computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device 309, or installed from a memory 308, or installed from a ROM 302. When the computer program is executed by the processor 301, it performs the functions defined in the non-invasive blood pressure measurement device detection method of embodiments of this disclosure.

[0065] It should be noted that the computer-readable medium described in this disclosure can be a computer-readable signal medium or a computer-readable storage medium, or any combination thereof. A computer-readable storage medium can be, for example,—but not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this disclosure, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In this disclosure, a computer-readable signal medium can include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium can be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to: wires, optical fibers, RF (radio frequency), etc., or any suitable combination thereof.

[0066] In some implementations, clients and servers can communicate using any currently known or future-developed network protocol such as HTTP (Hypertext Transfer Protocol) and can interconnect with digital data communication (e.g., communication networks) of any form or medium. Examples of communication networks include local area networks (“LANs”), wide area networks (“WANs”), the Internet (e.g., the Internet of Things), and peer-to-peer networks (e.g., ad hoc peer-to-peer networks), as well as any currently known or future-developed networks.

[0067] The aforementioned computer-readable medium may be included in the aforementioned electronic device; or it may exist independently and not assembled into the electronic device.

[0068] The aforementioned computer-readable medium carries one or more programs, which, when executed by the electronic device, cause the electronic device to perform the aforementioned non-invasive blood pressure measurement device detection method.

[0069] Electronic devices can be programmed with computer program code in one or more programming languages ​​or combinations thereof to perform the operations of this disclosure. These programming languages ​​include, but are not limited to, object-oriented programming languages ​​such as Java, Smalltalk, and C++, as well as conventional procedural programming languages ​​such as "C" or similar languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0070] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0071] The units described in the embodiments of this disclosure can be implemented in software or hardware. The names of the units are not, in some cases, intended to limit the specific unit.

[0072] The functions described above in this document can be performed at least in part by one or more hardware logic components. For example, exemplary types of hardware logic components that can be used, without limitation, include: field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), system-on-a-chip (SoCs), complex programmable logic devices (CPLDs), and so on.

[0073] In the context of this disclosure, a machine-readable medium can be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can be, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0074] The above description is merely a preferred embodiment of this disclosure and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of this disclosure is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the above-described concept. For example, technical solutions formed by substituting the above features with (but not limited to) technical features disclosed in this disclosure that have similar functions.

[0075] Furthermore, while the operations are described in a specific order, this should not be construed as requiring these operations to be performed in the specific order shown or in a sequential order. In certain environments, multitasking and parallel processing may be advantageous. Similarly, while several specific implementation details are included in the above discussion, these should not be construed as limiting the scope of this disclosure. Certain features described in the context of individual embodiments may also be implemented in combination in a single embodiment. Conversely, various features described in the context of a single embodiment may also be implemented individually or in any suitable sub-combination in multiple embodiments.

[0076] Although the subject matter has been described using language specific to structural features and / or methodological logic, it should be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or actions described above. Rather, the specific features and actions described above are merely illustrative examples of implementing the claims.

Claims

1. A detection method for a non-invasive blood pressure measurement device, characterized in that, include: Acquire test records from historical time periods and test records from the test period; extract invasive blood pressure reference value features from the test records based on a preset blood pressure assignment model; and extract pulse wave morphological features from the test records. Based on the blood pressure assignment model, the invasive blood pressure reference value features and the pulse wave morphological features are fused to obtain a multi-dimensional fused feature vector. Based on the blood pressure assignment model, the multi-dimensional fused feature vector is calculated to output the invasive blood pressure prediction value for the detection period. Based on the invasive blood pressure prediction value, the non-invasive blood pressure detection value obtained during the detection period is numerically determined to obtain the numerical determination result, and the accuracy status of the non-invasive blood pressure detection device is generated based on the numerical determination result. Based on the accuracy of the device, a test report containing the evaluation conclusions of the non-invasive blood pressure monitoring device is generated according to a preset report template.

2. The method according to claim 1, characterized in that, Acquire test records from historical time periods and test records from the test period; extract invasive blood pressure reference value features from the test records based on a preset blood pressure assignment model; and extract pulse wave morphological features from the test records, including: The test records and the test records are time-series aligned to generate associated data pairs; The invasive blood pressure waveform data in the test record is preprocessed to obtain a first preprocessing result. Based on the preset peak detection algorithm and the blood pressure assignment model, the systolic blood pressure peak point and diastolic blood pressure trough point in the first preprocessing result are identified. The invasive blood pressure reference value features are extracted based on the systolic blood pressure peak point and diastolic blood pressure trough point. The pulse wave signal in the detection record is subjected to a second preprocessing to obtain a second preprocessing result. Based on the blood pressure assignment model, the pulse wave morphological features are extracted from the second preprocessing result.

3. The method according to claim 2, characterized in that, Based on the blood pressure assignment model, a multi-dimensional fused feature vector is obtained by fusing the invasive blood pressure reference value features and the pulse wave morphological features. The multi-dimensional fused feature vector is then calculated based on the blood pressure assignment model to output the predicted invasive blood pressure value for the detection period, including: Based on the blood pressure assignment model and the associated data pair, the invasive blood pressure reference value features and the pulse wave morphological features are concatenated to obtain a multi-dimensional fused feature vector. Based on the blood pressure assignment model, the invasive blood pressure reference value features in the multi-dimensional fusion feature vector are used to calibrate and map the pulse wave morphology features to obtain the invasive blood pressure prediction waveform data corresponding to the detection period. Based on the invasive blood pressure prediction waveform data, the systolic blood pressure value and diastolic blood pressure value corresponding to the detection period are calculated using the preset peak detection algorithm, and the systolic blood pressure value and diastolic blood pressure value are determined as the invasive blood pressure prediction value for the detection period.

4. The method according to claim 2, characterized in that, The blood pressure assignment model includes an input layer, a multi-layer feature extraction layer, a feature fusion layer, a fully connected regression layer, and an output layer; Based on the blood pressure assignment model, the multi-dimensional fused feature vector is calculated, and the predicted invasive blood pressure value for the detection period is output, including: The input layer of the blood pressure assignment model obtains the invasive blood pressure waveform data in the test record and the pulse wave signal in the detection record. Based on the multi-layer feature extraction layer of the blood pressure assignment model, the pulse wave signal and the invasive blood pressure waveform data are subjected to hierarchical feature extraction to obtain the pulse wave morphological features and the invasive blood pressure reference value features. Based on the feature fusion layer of the blood pressure assignment model, the pulse wave morphological features and the invasive blood pressure reference value features are fused to obtain a multi-dimensional fused feature vector. Based on the fully connected regression layer of the blood pressure assignment model, the multi-dimensional fused feature vector is subjected to multi-layer nonlinear transformation and regression mapping to obtain invasive blood pressure prediction waveform data. Based on the output layer of the blood pressure assignment model, the invasive blood pressure prediction waveform data is output.

5. The method according to claim 1, characterized in that, Based on the invasive blood pressure prediction value, the non-invasive blood pressure measurement value obtained during the detection period is numerically determined to obtain a numerical determination result. Based on the numerical determination result, the accuracy status of the non-invasive blood pressure measurement device is generated, including: Non-invasive blood pressure readings were obtained based on the aforementioned detection period. Based on the invasive blood pressure prediction value, it is compared with the non-invasive blood pressure detection value, and the difference in measured blood pressure is calculated according to the comparison result; Obtain the preset blood pressure error threshold and blood pressure error standard deviation threshold, calculate the mean and standard deviation of the measured blood pressure difference based on the measured blood pressure difference during the detection period, and extract the measured blood pressure difference with the largest absolute difference in the calculation results as the maximum measured blood pressure difference; If the mean of the measured blood pressure difference is less than or equal to the blood pressure error threshold, and the standard deviation of the measured blood pressure difference is less than or equal to the blood pressure error standard deviation threshold, then the numerical determination result is passed. If the mean of the measured blood pressure difference is greater than the blood pressure error threshold, or the standard deviation of the measured blood pressure difference is greater than the blood pressure error standard deviation threshold, then the numerical judgment result is "not passed". If a preset maximum measurement error threshold is obtained, and the numerical judgment result is passed, and the maximum measurement difference is less than or equal to the maximum measurement error threshold, then the non-invasive blood pressure detection device is accurate. If the numerical result is "fail", then the non-invasive blood pressure monitoring device is inaccurate. The accuracy of the equipment includes both accurate and inaccurate equipment.

6. The method according to claim 1, characterized in that, Based on the accuracy of the device, a test report for the non-invasive blood pressure monitoring device, including evaluation conclusions, is generated according to a preset report template, including: If the device is in good condition, then the first report template in the preset report template library is invoked; If the device is inaccurate, then the second report template in the preset report template library is invoked. The second report template includes an area indicating the reason for non-compliance.

7. A non-invasive blood pressure measurement device, the device comprising: The acquisition module is used to acquire the test records of historical time periods and the test records of the test period, extract invasive blood pressure reference value features from the test records based on a preset blood pressure assignment model, and extract pulse wave morphological features from the test records. The prediction module is used to obtain a multi-dimensional fusion feature vector by fusing the invasive blood pressure reference value features and the pulse wave morphology features based on the blood pressure assignment model, calculate the multi-dimensional fusion feature vector based on the blood pressure assignment model, and output the invasive blood pressure prediction value for the detection period. The determination module is used to make a numerical determination on the non-invasive blood pressure detection value obtained during the detection period based on the invasive blood pressure prediction value, obtain a numerical determination result, and generate the device accuracy status of the non-invasive blood pressure detection device based on the numerical determination result. The generation module is used to generate a test report for the non-invasive blood pressure monitoring device, which includes evaluation conclusions, according to the accuracy of the device and a preset report template.

8. An electronic device, characterized in that, include: Memory; processor; as well as Computer programs; The computer program is stored in the memory and configured to be executed by the processor to implement the steps of the method as described in any one of claims 1-6.

9. A computer-readable storage medium, characterized in that, It stores a computer program / instruction thereon, which, when executed by a processor, implements the steps of the method described in any one of claims 1-6.

10. A computer program product comprising a computer program / instructions, characterized in that, When the computer program / instructions are executed by the processor, they implement the steps of the method according to any one of claims 1-6.

Images