A method and system for measuring intracranial pressure based on ultrasound technology

Abstract

The present invention relates to the field of ultrasound technology, and discloses a method and system for measuring intracranial pressure based on ultrasound technology, including: collecting sound wave signal data in the intracranial pressure, performing data enhancement processing on the sound wave signal data, and obtaining sound wave signal enhancement data; extracting sound wave signal feature data of the sound wave signal enhancement data, performing data screening on the sound wave signal feature data, and obtaining sound wave signal screening data; detecting the sound wave signal screening data, and obtaining intracranial pressure measurement data; performing error analysis on the measurement data and the actual intracranial pressure data, and obtaining measurement error data, and generating measurement data feedback semantics according to the measurement error data; determining the detection data of the intracranial pressure according to a preset detection threshold, and calculating the best measurement data of the intracranial pressure according to the measurement data feedback semantics and the detection data. The present invention can improve the accuracy of the measurement of intracranial pressure.

Description

Technical Field

[0001] The present invention relates to the field of ultrasonic technology, and in particular to a method and system for measuring intracranial pressure based on ultrasonic technology. Background Art

[0002] Ultrasound technology is a non-invasive imaging technology that generates images or performs measurements through the propagation and reflection of sound waves in human tissues. In the medical field, ultrasound technology is widely used in many fields such as cardiovascular, obstetrics and gynecology, and liver, and in recent years, it has also begun to explore its application in intracranial pressure measurement.

[0003] The existing intracranial pressure measurement technology provides a certain degree of intracranial pressure information through electroencephalogram (EEG), and infers the intracranial pressure through changes in brain electrical activity. In practical applications, EEG cannot directly measure the intracranial pressure value, but indirectly infers the intracranial pressure by observing changes in brain electrical activity. The indirectness may lead to uncertainty and error in the inference results, thus reducing the accuracy of intracranial pressure measurement. Summary of the invention

[0004] The present invention provides a method and system for measuring intracranial pressure based on ultrasound technology, the main purpose of which is to solve the problem of low accuracy in measuring intracranial pressure.

[0005] To achieve the above object, the present invention provides a method for measuring intracranial pressure based on ultrasound technology, comprising:

[0006] Collecting sound wave signal data in intracranial pressure, performing data enhancement processing on the sound wave signal data, and obtaining sound wave signal enhanced data;

[0007] Extracting acoustic signal characteristic data of the acoustic signal enhancement data, and performing data screening on the acoustic signal characteristic data according to a preset characteristic data algorithm to obtain acoustic signal screening data;

[0008] Detecting the sound wave signal screening data to obtain measurement data of intracranial pressure;

[0009] Performing error analysis on the measurement data and preset actual intracranial pressure data to obtain measurement error data, and generating measurement data feedback semantics according to the measurement error data;

[0010] The detection data of the intracranial pressure is determined according to a preset detection threshold, and the optimal measurement data of the intracranial pressure is calculated according to the feedback semantics of the measurement data and the detection data.

[0011] Optionally, collecting the sound wave signal data within the intracranial pressure includes:

[0012] Obtain the scanning area and scanning depth within the intracranial pressure;

[0013] Determine the data location within the intracranial pressure according to the scanning area and the scanning depth;

[0014] Configure ultrasound acquisition parameters through preset ultrasound algorithms;

[0015] Data is collected at the data positioning position according to the ultrasonic collection parameters to obtain sound wave signal data.

[0016] Optionally, performing data enhancement processing on the sound wave signal data to obtain sound wave signal enhanced data includes:

[0017] Using a preset filtering algorithm to remove noise from the sound wave signal data to obtain first sound wave signal data;

[0018] Performing signal adjustment on the first sound wave signal data to obtain second sound wave signal data;

[0019] Performing smoothing processing on the second sound wave signal data to obtain third sound wave signal data;

[0020] Correction processing is performed on the third sound wave signal data to obtain sound wave signal enhancement data.

[0021] Optionally, the extracting the acoustic wave signal feature data of the acoustic wave signal enhancement data includes:

[0022] Extracting time domain features and frequency domain features of the acoustic signal enhancement data;

[0023] Extracting statistical features and morphological features of the acoustic signal enhancement data;

[0024] The time domain features, the frequency domain features, the statistical features and the morphological features are fused using the following preset multi-dimensional feature algorithm to obtain acoustic signal feature data:

[0025] f = { 1 p 1 ,x 2 p 2 ,x 3 p 3 ,x 4 p 4}

[0026] Wherein, f is the characteristic data of the acoustic wave signal, and x 1 is the time domain weight of the time domain feature, p 1 is the time domain feature, χ 2 is the frequency domain weight of the frequency domain feature, p 2 is the frequency domain feature, χ 3 is the statistical weight of the statistical feature, p3 is the statistical feature, χ 4 is the morphological weight of the morphological feature, p 4 For the morphological characteristics.

[0027] Optionally, the performing data screening on the acoustic wave signal characteristic data to obtain acoustic wave signal screening data includes:

[0028] The acoustic signal characteristic data is compared with the preset acoustic signal characteristic threshold data according to the following preset characteristic data algorithm to obtain a data comparison symbol value:

[0029]

[0030] Wherein, q is the data comparison symbol value, p i is the characteristic value of the i-th characteristic in the acoustic signal characteristic data, w i is the characteristic value of the i-th characteristic in the acoustic signal characteristic threshold data;

[0031] generating a data symbol sequence according to the data comparison symbol value;

[0032] The data symbol sequence is screened according to a preset target numerical quantity to obtain acoustic wave signal screening data.

[0033] Optionally, the detecting the acoustic wave signal screening data to obtain the intracranial pressure measurement data includes:

[0034] Performing vector conversion on the sound wave signal screening data to obtain a sound wave signal vector;

[0035] Extracting global signal features corresponding to the acoustic wave signal vector using a feature layer of a pre-trained deep learning model;

[0036] The global signal feature is input into the output layer of the pre-trained deep learning model to output the measurement data of intracranial pressure.

[0037] Optionally, performing error analysis on the measurement data and preset actual intracranial pressure data to obtain measurement error data includes:

[0038] Determining a measurement data interval of the measurement data according to preset actual intracranial pressure data;

[0039] The measurement data is subjected to error analysis according to the measurement data interval to obtain measurement error data, wherein the measurement error data calculation formula is:

[0040]

[0041] Wherein, r is the measurement error data, c is the measurement data, xk is the kth actual intracranial pressure data in the measurement data interval, and the number of interval data is n.

[0042] Optionally, generating measurement data feedback semantics according to the measurement error data includes:

[0043] Determining a feedback information type according to the measurement error data and a preset measurement error threshold;

[0044] Generate a feedback sentence template according to the feedback information type;

[0045] The measurement data feedback semantics is generated according to the feedback sentence template.

[0046] Optionally, the calculating the optimal measurement data of intracranial pressure according to the measurement data feedback semantics and the detection data includes:

[0047] Determining target measurement data of intracranial pressure according to the measurement data feedback semantics;

[0048] The target measurement data and the detection data are averaged to obtain optimal measurement data of intracranial pressure.

[0049] In order to solve the above problems, the present invention also provides a system for measuring intracranial pressure based on ultrasound technology, the system comprising:

[0050] The acoustic signal data enhancement module is used to collect the acoustic signal data in the intracranial pressure, perform data enhancement processing on the acoustic signal data, and obtain acoustic signal enhanced data;

[0051] An acoustic signal characteristic data module, used for extracting the acoustic signal characteristic data of the acoustic signal enhancement data, and performing data screening on the acoustic signal characteristic data to obtain acoustic signal screening data;

[0052] An acoustic wave signal screening data module, used to detect the acoustic wave signal screening data to obtain intracranial pressure measurement data;

[0053] A measurement data feedback semantics generating module, used for performing error analysis between the measurement data and preset actual intracranial pressure data to obtain measurement error data, and generating measurement data feedback semantics according to the measurement error data;

[0054] The optimal measurement data calculation module is used to determine the detection data of the intracranial pressure according to a preset detection threshold, and calculate the optimal measurement data of the intracranial pressure according to the feedback semantics of the measurement data and the detection data.

[0055] The embodiment of the present invention collects intracranial acoustic wave signal data through an ultrasonic algorithm, which is conducive to providing accurate input, and then enhances the data, thereby improving data quality, enhancing signal characteristics, and facilitating the accuracy of subsequent feature extraction and detection; extracting useful feature information from the acoustic wave signal is conducive to revealing the potential laws of the data, removing useless information through data screening, reducing noise interference, and improving the efficiency and accuracy of subsequent processing; detecting intracranial pressure data through a pre-trained deep learning model is conducive to improving the accuracy and stability of intracranial pressure measurement; by comparing with actual data and performing error analysis, the accuracy and reliability of the measurement results can be evaluated, providing a reference for subsequent improvements; combining multiple measurement values ​​and data feedback semantics, more accurate intracranial pressure measurement data can be obtained. Therefore, the method and system for measuring intracranial pressure based on ultrasonic technology proposed in the present invention can solve the problem of low accuracy when measuring intracranial pressure. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] Figure 1 A schematic diagram of a flow chart of a method for measuring intracranial pressure based on ultrasound technology provided by an embodiment of the present invention;

[0057] Figure 2 A schematic diagram of a process for screening acoustic signal characteristic data provided by an embodiment of the present invention;

[0058] Figure 3 A schematic diagram of a process for training a deep learning model provided by an embodiment of the present invention;

[0059] Figure 4 A functional module diagram of a system for measuring intracranial pressure based on ultrasound technology provided by an embodiment of the present invention;

[0060] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0061] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0062] The embodiment of the present application provides a method for measuring intracranial pressure based on ultrasound technology. The execution subject of the method for measuring intracranial pressure based on ultrasound technology includes but is not limited to at least one of the electronic devices such as a server and a terminal that can be configured to execute the method provided in the embodiment of the present application. In other words, the method for measuring intracranial pressure based on ultrasound technology can be executed by software or hardware installed on a terminal device or a server device, and the software can be a blockchain platform. The server includes but is not limited to: a single server, a server cluster, a cloud server or a cloud server cluster, etc. The server can be an independent server, or it can be a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (Content Delivery Network, CDN), and big data and artificial intelligence platforms.

[0063] Reference Figure 1 FIG. 1 is a flow chart of a method for measuring intracranial pressure based on ultrasound technology provided by an embodiment of the present invention. In this embodiment, the method for measuring intracranial pressure based on ultrasound technology includes:

[0064] S1. Collecting sound wave signal data in intracranial pressure, performing data enhancement processing on the sound wave signal data, and obtaining sound wave signal enhanced data.

[0065] In the embodiment of the present invention, the sound wave signal data refers to the digitized data of the sound wave signal of intracranial pressure collected by the ultrasonic device. The sound wave signal data is recorded in digital form, including information such as the amplitude, frequency, and phase of the sound wave.

[0066] In the embodiment of the present invention, the step of collecting the acoustic wave signal data within the intracranial pressure includes:

[0067] Obtain the scanning area and scanning depth within the intracranial pressure;

[0068] Determine the data location within the intracranial pressure according to the scanning area and the scanning depth;

[0069] Configure ultrasound acquisition parameters through preset ultrasound algorithms;

[0070] Data is collected at the data positioning position according to the ultrasonic collection parameters to obtain sound wave signal data.

[0071] In detail, medical imaging technology (such as CT or MRI) is used to assist in determination to ensure that the intracranial area of ​​interest can be accurately covered, and then an appropriate scanning depth is set according to the position of the scanning area and the depth of the target structure to ensure that the sound waves can effectively penetrate the target area and capture the reflected signals. Based on the settings of the scanning area and depth, the specific data positioning position in the brain is determined, and the data positioning position will be used for subsequent sound wave signal acquisition to ensure the accuracy and reliability of the data.

[0072] Specifically, configure the parameters of the ultrasound device, including frequency, gain, depth, intensity, emission angle, etc., to ensure that the collected sound wave signal data meets the requirements, and then use the ultrasound device to collect sound wave signals at the determined data positioning position according to the preset ultrasound acquisition parameters. Keep the probe at the target position and collect data according to the equipment operating specifications. The ultrasound device will emit sound waves and receive signals reflected from intracranial tissues to obtain sound wave signal data, and save and record the collected sound wave signal data, including relevant patient information, acquisition time, etc.

[0073] Furthermore, in order to improve the data quality of the collected acoustic wave signal data, it is necessary to perform data preprocessing on the acoustic wave signal data to improve the data quality of the acoustic wave signal data, thereby improving the accuracy and reliability of intracranial pressure measurement.

[0074] In the embodiment of the present invention, the acoustic signal enhancement data refers to improved and optimized acoustic signal data obtained after a series of processing steps, thereby improving the quality, clarity and usability of the acoustic signal data.

[0075] In the embodiment of the present invention, the step of performing data enhancement processing on the sound wave signal data to obtain sound wave signal enhanced data includes:

[0076] Using a preset filtering algorithm to remove noise from the sound wave signal data to obtain first sound wave signal data;

[0077] Performing signal adjustment on the first sound wave signal data to obtain second sound wave signal data;

[0078] Performing smoothing processing on the second sound wave signal data to obtain third sound wave signal data;

[0079] Correction processing is performed on the third sound wave signal data to obtain sound wave signal enhancement data.

[0080] In detail, according to the characteristics of the sound wave signal and the type of noise (such as white noise, environmental noise, etc.), the sound wave signal data is subjected to noise removal to obtain the first sound wave signal data, and the filtering algorithm includes but is not limited to a low-pass filter, a high-pass filter, a band-pass filter and an adaptive filter, for example, a low-pass filter is used to remove high-frequency noise or a high-pass filter is used to remove low-frequency noise; the signal can be filtered by a time domain or frequency domain method to retain the effective signal and remove the noise part, and the original sound wave signal data is processed by the selected filtering algorithm to remove unnecessary frequency components or noise, so as to obtain a clearer first sound wave signal data after removing irrelevant noise from the sound wave signal data.

[0081] Specifically, the first sound wave signal data after removing irrelevant noise from the sound wave signal data is subjected to signal adjustment to ensure that the amplitude range of the signal is within an appropriate range, including but not limited to operations such as signal amplification, attenuation or normalization. Signal adjustment helps to highlight the signal features of interest and improve the signal-to-noise ratio (SNR), thereby obtaining the second sound wave signal data after signal adjustment; the second sound wave signal data is then smoothed to eliminate peaks and burrs in the signal, so that the signal is smoother and more continuous, which is conducive to subsequent feature extraction and analysis. The smoothing method includes but is not limited to moving average and weighted moving average; finally, the collected data is corrected to ensure the accuracy and reliability of the data, eliminate possible deviations or errors, ensure that the signal accurately reflects the actual physiological or physical conditions, and obtain the final sound wave signal enhancement data, thereby significantly improving the quality of the sound wave signal data, reducing noise interference, and enhancing signal characteristics.

[0082] Furthermore, it is necessary to extract rich feature information from the enhanced sound wave signal, which can more comprehensively reflect the state of the detected object, thereby improving the accuracy and reliability of analysis and judgment.

[0083] S2. Extracting the acoustic wave signal characteristic data of the acoustic wave signal enhancement data, and performing data screening on the acoustic wave signal characteristic data to obtain acoustic wave signal screening data.

[0084] In an embodiment of the present invention, the acoustic signal characteristic data refers to data extracted after analyzing and processing the acoustic signal, which can represent the essential attributes and characteristics of the acoustic signal. The characteristic data is usually obtained from the original acoustic signal after a series of preprocessing, enhancement and feature extraction steps, and reflects the key information of the acoustic signal in multiple dimensions such as time domain, frequency domain, statistics and morphology.

[0085] In the embodiment of the present invention, the step of extracting the acoustic signal characteristic data of the acoustic signal enhancement data includes:

[0086] Extracting time domain features and frequency domain features of the acoustic signal enhancement data;

[0087] Extracting statistical features and morphological features of the acoustic signal enhancement data;

[0088] The time domain features, the frequency domain features, the statistical features and the morphological features are fused using the following preset multi-dimensional feature algorithm to obtain acoustic signal feature data:

[0089] f = { 1 p 1 ,x 2 p 2 ,x 3 p 3 ,x 4 p 4}

[0090] Wherein, f is the characteristic data of the acoustic wave signal, and x 1 is the time domain weight of the time domain feature, p 1 is the time domain feature, χ 2 is the frequency domain weight of the frequency domain feature, p 2 is the frequency domain feature, χ 3 is the statistical weight of the statistical feature, p 3 is the statistical feature, χ 4 is the morphological weight of the morphological feature, p 4 For the morphological characteristics.

[0091] In detail, time domain features are extracted in the time domain. The time domain features reflect the characteristics of the sound wave signal in the time dimension, such as amplitude, energy, zero crossing rate, etc.; frequency domain features analyze the frequency components of the sound wave signal by performing Fourier transform or wavelet transform on the sound wave signal, and extract features such as spectrum, bandwidth, frequency center, etc.; statistical features analyze the sound wave signal from a statistical perspective, and extract statistical quantities such as mean, variance, skewness, kurtosis, etc. Statistical features can reveal the distribution and characteristics of the signal and play a certain role in the classification and identification of the signal; morphological features describe the shape and structure of the sound wave signal waveform, such as peak value, valley value, rise time, fall time, etc., which are helpful for morphological analysis and identification of the signal.

[0092] Specifically, the time domain features, frequency domain features, statistical features and morphological features extracted above are fused using a preset multidimensional feature algorithm to obtain comprehensive acoustic signal feature data, which can comprehensively describe all aspects of the signal and provide more information and reference for subsequent analysis and application. The feature weights in the multidimensional feature algorithm can indicate that some features may reflect the information we care about better than other features. By assigning higher weights to these features, their influence in the final analysis results can be emphasized, so that the results are more in line with actual application needs. In the absence of weights, the contribution of each feature to the final result is the same, which may cause some less important features to have too much influence on the result. By reasonably allocating weights, the contribution of each feature to the final result can be balanced to ensure the rationality and accuracy of the result. Therefore, through feature fusion, the characteristics of the acoustic signal in different dimensions can be comprehensively considered, making the final feature data more comprehensive and representative.

[0093] Furthermore, in order to extract the most relevant and useful information from a large amount of raw or basic feature data, it is necessary to screen the acoustic signal feature data, which is conducive to tasks such as identification or classification being based on more accurate and high-quality data sets, which can improve processing efficiency and analysis accuracy while ensuring data quality.

[0094] In the embodiment of the present invention, the acoustic wave signal screening data refers to the data after normal data and abnormal data are screened to screen out the data feature set with the most information.

[0095] In the embodiment of the present invention, referring to Figure 2 As shown, the data screening of the acoustic wave signal characteristic data to obtain acoustic wave signal screening data includes:

[0096] S21, comparing the acoustic signal characteristic data with preset acoustic signal characteristic threshold data according to the following preset characteristic data algorithm to obtain a data comparison symbol value:

[0097]

[0098] Wherein, q is the data comparison symbol value, p i is the characteristic value of the i-th characteristic in the acoustic signal characteristic data, w i is the characteristic value of the i-th characteristic in the acoustic signal characteristic threshold data;

[0099] S22, generating a data symbol sequence according to the data comparison symbol value;

[0100] S23. Screening the data symbol sequence according to a preset target numerical quantity to obtain sound wave signal screening data.

[0101] In detail, the extracted acoustic signal feature data is compared with the preset acoustic signal feature threshold data one by one. If the feature in the acoustic signal feature data is greater than the feature value corresponding to the feature in the acoustic signal feature threshold data, the data comparison symbol value corresponding to this feature is marked as 1; if the feature in the acoustic signal feature data is equal to the feature value corresponding to the feature in the acoustic signal feature threshold data, the data comparison symbol value corresponding to this feature is marked as 0; when the feature in the acoustic signal feature data is less than the feature value corresponding to the feature in the acoustic signal feature threshold data, the data comparison symbol value corresponding to this feature is marked as -1, thereby generating a data symbol sequence of only 1, 0, and -1 according to the data comparison symbol value, and the data symbol sequence reflects the result of comparing all feature values ​​with their corresponding thresholds. The data symbol sequence can be regarded as a set that summarizes all feature comparison results. Finally, according to the preset target number of values ​​(that is, the number of feature values ​​that you want to filter out to meet the conditions), the data symbol sequence is screened, and the feature values ​​that best meet the set conditions are selected from all features to form acoustic signal screening data.

[0102] For example, the characteristic value of the feature in the wave signal characteristic data is (A, B, C, D), and the characteristic value of the feature in the sound wave signal characteristic threshold data is like The data symbol comparison value is -1. The data symbol comparison value is 0. With this type, the possible data symbol sequence is (-1, 0, 1, 1). Then the number of -1 in the data symbol sequence is counted. As long as the number of -1 is greater than or equal to 1, the data symbol sequence is regarded as abnormal data, otherwise it is regarded as normal data.

[0103] Furthermore, the acoustic wave signal screening model can be tested through the pre-trained model to obtain more accurate intracranial pressure measurement data and more reliable measurement results.

[0104] S3. Detect the sound wave signal screening data to obtain measurement data of intracranial pressure.

[0105] In an embodiment of the present invention, the deep learning model is a machine learning model based on an artificial neural network structure, which learns the complex features and patterns of input data by forming a network of multi-level neurons, including but not limited to multi-layer perceptron (MLP), convolutional neural network (CNN), recurrent neural network (RNN) and transformer (Transformer), and can continuously adjust the weights and parameters between layers through the back propagation algorithm, so that the model can gradually understand and learn the complex features of the input data and make accurate predictions or classifications.

[0106] In the embodiment of the present invention, before the acoustic wave signal screening data is detected to obtain the measurement data of intracranial pressure, the method further includes:

[0107] Acquire historical sound wave signal data of intracranial pressure, and perform data annotation on the historical sound wave signal data to obtain a sound wave signal data set;

[0108] Inputting the acoustic wave signal data set into a preset learning model for training to obtain a model output value;

[0109] Calculating a loss value between the model output value and a preset actual intracranial pressure value;

[0110] When the loss value is less than a preset loss threshold, the learning model is output as a deep learning model.

[0111] In detail, a large amount of historical sound wave signal data is collected and acquired. These data will be used as input for training the deep learning model. Data related to intracranial pressure are collected, including ultrasound images, clinical records and other relevant data. The historical data contains intracranial pressure measurement data under various circumstances, so as to train a representative model and obtain a sound wave signal data set. The acquired historical sound wave signal data is then labeled, that is, the actual intracranial pressure value corresponding to each data point or sample mark is labeled. The labeled data will serve as the training target in supervised learning to help the model learn the correct association. The labeled historical sound wave signal data constitutes a data set for model training. The data set will include sound wave signal features and corresponding intracranial pressure values. The constructed sound wave signal data set is input into the preset learning model for training. The labeling can be achieved through manual labeling by medical experts or using automated methods.

[0112] Specifically, the deep learning model will continuously adjust parameters through the back-propagation algorithm to minimize the loss function, thereby learning the complex mapping relationship between the acoustic wave signal and the intracranial pressure. During the training process, the loss value between the model output value and the actual intracranial pressure value is calculated. The loss calculation includes but is not limited to indicators such as the mean squared error to measure the deviation between the model output and the true value. When the loss value is less than the preset loss threshold, it means that the model has achieved a certain accuracy. It can be considered that the model training is sufficient, and the learning model can be output as the final deep learning model for intracranial pressure measurement of acoustic wave signal screening data.

[0113] Furthermore, the use of deep learning models for acoustic signal detection and data processing can realize an automated measurement process, reduce manual intervention, and improve measurement efficiency. Deep learning models can learn complex patterns and features from large amounts of historical data, and therefore can provide more precise and accurate intracranial pressure measurement data.

[0114] In the embodiment of the present invention, the intracranial pressure measurement data refers to the numerical value and related information of the patient's intracranial pressure, which is very important for evaluating the patient's brain health status, monitoring the fluctuation of intracranial pressure and guiding treatment.

[0115] In the embodiment of the present invention, refer to Figure 3 As shown, the detection of the acoustic wave signal screening data to obtain the measurement data of intracranial pressure includes:

[0116] S31, performing vector conversion on the sound wave signal screening data to obtain a sound wave signal vector;

[0117] S32, using the feature layer of the pre-trained deep learning model to extract the global signal features corresponding to the acoustic wave signal vector;

[0118] S33. Input the global signal feature into the output layer of the pre-trained deep learning model, and output the measurement data of intracranial pressure.

[0119] In detail, after screening and preprocessing, the acoustic signal data can be represented in a vector form, in which each element represents the amplitude value of the acoustic signal at a certain time point or frequency. The acoustic signal vector can be used as the input of the deep learning model. The pre-trained deep learning model usually contains multiple layers, in which the first few layers are used to extract low-level features of the data, while the later layers can extract more abstract and global features. The acoustic signal vector passes through the feature extraction layer of the deep learning model to obtain the corresponding global signal feature representation, that is, the convolution operation is used to extract the local features of the acoustic signal, and higher-level features can be gradually extracted by stacking multiple convolution layers. The pooling operation is used to reduce the dimension of the feature map and retain important feature information. The feature map obtained by the pooling layer is flattened. The global signal feature can capture important patterns and structural information in the acoustic signal. Finally, the extracted global signal feature will be sent to the output layer of the deep learning model. The output layer can be trained to map the global feature to the measurement data of intracranial pressure, that is, the output layer usually uses a linear or nonlinear activation function to produce the final prediction result. For intracranial pressure measurement, the output layer can be a single node, representing the predicted pressure value.

[0120] Furthermore, by comparing the intracranial pressure measurement data output by the deep learning model with the actual intracranial pressure data, the accuracy and precision of the measurement results can be evaluated.

[0121] S4. Perform error analysis on the measurement data and preset actual intracranial pressure data to obtain measurement error data, and generate measurement data feedback semantics according to the measurement error data.

[0122] In the embodiment of the present invention, the measurement error data refers to the difference or degree of deviation between the actual measured value and the true value, that is, the gap between the measured value and the actual intracranial pressure.

[0123] In the embodiment of the present invention, performing error analysis on the measurement data and preset actual intracranial pressure data to obtain measurement error data includes:

[0124] Determining a measurement data interval of the measurement data according to preset actual intracranial pressure data;

[0125] The measurement data is subjected to error analysis according to the measurement data interval to obtain measurement error data, wherein the measurement error data calculation formula is:

[0126]

[0127] Wherein, r is the measurement error data, c is the measurement data, x k is the kth actual intracranial pressure data in the measurement data interval, and the number of interval data is n.

[0128] In detail, by screening the measurement data according to the actual intracranial pressure data corresponding to the historical data, the measurement data interval corresponding to the measurement data can be obtained. For example, if the actual intracranial pressure data is 30 and the measurement data is 35, then the measurement data interval is an integer between 30 and 35, that is, 30, 31, 32, 33, 34, 35. Then, the difference between the measurement data and each data in the measurement interval is calculated one by one to obtain measurement error data, and then the feedback data of the measurement data is determined according to the measurement error data, and the measurement data is continuously updated according to the feedback data.

[0129] Furthermore, by converting measurement error data into feedback semantics, the accuracy of monitoring data can be understood more intuitively, thereby improving the accuracy and reliability of the monitoring process, ensuring the accuracy of monitoring data, and reducing adverse effects caused by errors.

[0130] In an embodiment of the present invention, the data feedback semantics refers to converting data into a language form that is easy to understand and communicate. The data feedback semantics includes providing feedback through language expression based on the accuracy, degree of deviation, etc. of the measured data to provide a clear explanation and prompt of the meaning of the monitoring data.

[0131] In the embodiment of the present invention, the generating of measurement data feedback semantics according to the measurement error data includes:

[0132] Determining a feedback information type according to the measurement error data and a preset measurement error threshold;

[0133] Generate a feedback sentence template according to the feedback information type;

[0134] The measurement data feedback semantics is generated according to the feedback sentence template.

[0135] In detail, based on the measurement error data, define the type of information that needs to be fed back to the user, such as "high error", "low error", etc., and set the corresponding error threshold according to the statistical characteristics of the error data and the specific application scenario to determine the type of feedback information. For example, if the error value is greater than a certain threshold, it is judged as "high error", otherwise it is "low error". Then, according to the type of feedback information, write the corresponding feedback language template, such as "normal situation: the measurement result is accurate; slight deviation: the measurement result has a slight deviation, please note that there may be an error; medium deviation: the measurement result has a certain deviation, it is recommended to recheck the data; serious deviation: the measurement result has a large error, please take immediate measures to check the equipment or data, etc. According to the threshold and template, the actual error data is substituted into the template to generate the corresponding feedback statement. For example, for a measurement result with an error value of 0.1, if the threshold is set to 0.2, a feedback statement similar to "the measurement result is relatively accurate and the error is small" can be generated, and the generated feedback statement is output to the user or other relevant parties to prompt the accuracy or error of the measurement result.

[0136] Furthermore, through the preset detection threshold and measurement data feedback semantics, the errors and omissions in intracranial pressure monitoring can be effectively reduced, and the reliability and accuracy of the monitoring results can be improved.

[0137] S5. Determine the detection data of the intracranial pressure according to a preset detection threshold, and calculate the optimal measurement data of the intracranial pressure according to the feedback semantics of the measurement data and the detection data.

[0138] In an embodiment of the present invention, the detection data refers to multiple detection data obtained by multiple detections of acoustic wave signal data using a deep learning model, wherein multiple measurements are performed using a deep learning model according to a configured detection threshold, and multiple intracranial pressure detection data output by the deep learning model are obtained.

[0139] Furthermore, through the measurement data feedback semantics, the intracranial pressure monitoring data can be effectively evaluated and fed back, which is conducive to discovering possible measurement errors or abnormal data and ensuring the accuracy of the monitoring data.

[0140] In the embodiment of the present invention, the optimal measurement data refers to the most accurate and reliable data obtained by comprehensively considering the target measurement data and the actual detection data.

[0141] In an embodiment of the present invention, the step of calculating the optimal measurement data of intracranial pressure according to the measurement data feedback semantics and the detection data includes:

[0142] Determining target measurement data of intracranial pressure according to the measurement data feedback semantics;

[0143] The target measurement data and the detection data are averaged to obtain optimal measurement data of intracranial pressure.

[0144] In detail, according to the feedback semantics of the measurement data, combined with the patient's clinical condition and monitoring needs, the target measurement data of the intracranial pressure is determined, and the determined target measurement data and the actually detected data are averaged. The average of the two values ​​can be obtained by adding the target measurement data and the actual detection data and then dividing by 2. The value obtained by the mean calculation is the optimal measurement data of the intracranial pressure. The optimal measurement data comprehensively considers the target measurement data and the actual detection data, and can better reflect the patient's actual situation.

[0145] The embodiment of the present invention collects intracranial acoustic wave signal data through an ultrasonic algorithm, which is conducive to providing accurate input, and then enhances the data, thereby improving data quality, enhancing signal characteristics, and facilitating the accuracy of subsequent feature extraction and detection; extracting useful feature information from the acoustic wave signal is conducive to revealing the potential laws of the data, removing useless information through data screening, reducing noise interference, and improving the efficiency and accuracy of subsequent processing; detecting intracranial pressure data through a pre-trained deep learning model is conducive to improving the accuracy and stability of intracranial pressure measurement; by comparing with actual data and performing error analysis, the accuracy and reliability of the measurement results can be evaluated, providing a reference for subsequent improvements; combining multiple measurement values ​​and data feedback semantics, more accurate intracranial pressure measurement data can be obtained. Therefore, the method and system for measuring intracranial pressure based on ultrasonic technology proposed in the present invention can solve the problem of low accuracy when measuring intracranial pressure.

[0146] like Figure 4 , is a functional module diagram of a system for measuring intracranial pressure based on ultrasound technology provided by an embodiment of the present invention.

[0147] The intracranial pressure measurement system 100 based on ultrasound technology of the present invention can be installed in an electronic device. According to the functions implemented, the intracranial pressure measurement system 100 based on ultrasound technology can include an acoustic signal data enhancement module 101, an acoustic signal feature data module 102, an acoustic signal screening data module 103, a measurement data feedback semantic generation module 104 and an optimal measurement data calculation module 105. The module of the present invention can also be referred to as a unit, which refers to a series of computer program segments that can be executed by an electronic device processor and can complete fixed functions, which are stored in the memory of the electronic device.

[0148] In this embodiment, the functions of each module / unit are as follows:

[0149] The acoustic signal data enhancement module 101 is used to collect acoustic signal data in intracranial pressure, perform data enhancement processing on the acoustic signal data, and obtain acoustic signal enhanced data;

[0150] The acoustic signal characteristic data module 102 is used to extract the acoustic signal characteristic data of the acoustic signal enhancement data, and perform data screening on the acoustic signal characteristic data to obtain acoustic signal screening data;

[0151] The acoustic wave signal screening data module 103 is used to detect the acoustic wave signal screening data to obtain the measurement data of the intracranial pressure;

[0152] The measurement data feedback semantics generating module 104 is used to perform error analysis between the measurement data and preset actual intracranial pressure data to obtain measurement error data, and generate measurement data feedback semantics according to the measurement error data;

[0153] The optimal measurement data calculation module 105 is used to determine the detection data of the intracranial pressure according to a preset detection threshold, and calculate the optimal measurement data of the intracranial pressure according to the measurement data feedback semantics and the detection data.

[0154] In detail, each module described in the system 100 for measuring intracranial pressure based on ultrasound technology in the embodiment of the present invention is used in the same manner as described above. Figures 1 to 3 The same technical means as the method for measuring intracranial pressure based on ultrasound technology described in the text can produce the same technical effects and will not be repeated here.

[0155] In the several embodiments provided by the present invention, it should be understood that the disclosed systems and methods can be implemented in other ways. For example, the system embodiments described above are only illustrative, for example, the division of the modules is only a logical function division, and there may be other division methods in actual implementation.

[0156] The modules described as separate components may or may not be physically separated, and the components shown as modules may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0157] In addition, each functional module in each embodiment of the present invention may be integrated into one processing unit, each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of hardware plus software functional modules.

[0158] It is obvious to those skilled in the art that the present invention is not limited to the details of the above exemplary embodiments, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0159] Therefore, no matter from which point of view, the embodiments should be regarded as illustrative and non-restrictive, and the scope of the present invention is not limited only according to the above description, and it is intended that all changes within the meaning and scope of equivalent elements within the scope of protection are included in the present invention.

[0160] The embodiments of the present application can acquire and process relevant data based on artificial intelligence technology. Among them, artificial intelligence (AI) is the theory, method, technology and application system that uses digital computers or machines controlled by digital computers to simulate, extend and expand human intelligence, perceive the environment, acquire knowledge and use knowledge to obtain the best results.

[0161] In addition, it is obvious that the word "comprising" does not exclude other units or steps, and the singular does not exclude the plural. The multiple units or systems stated in the system embodiment can also be implemented by one unit or system through software or hardware. The words first, second, etc. are used to indicate names, and do not indicate any specific order.

[0162] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solution of the present invention can be modified or replaced by equivalents without departing from the spirit and scope of the technical solution of the present invention.

Claims

1. A method for measuring intracranial pressure based on ultrasound technology, characterized in that: The method comprises: Collecting sound wave signal data in intracranial pressure, performing data enhancement processing on the sound wave signal data, and obtaining sound wave signal enhanced data; Extracting the acoustic signal characteristic data of the acoustic signal enhancement data, and performing data screening on the acoustic signal characteristic data to obtain acoustic signal screening data; The acoustic wave signal screening data is detected to obtain the measurement data of the intracranial pressure, including: performing vector conversion on the acoustic wave signal screening data to obtain the acoustic wave signal vector; using the feature layer of the pre-trained deep learning model to extract the global signal feature corresponding to the acoustic wave signal vector; inputting the global signal feature into the output layer of the pre-trained deep learning model to output the measurement data of the intracranial pressure; The measurement data and preset actual intracranial pressure data are subjected to error analysis to obtain measurement error data, including: determining a measurement data interval of the measurement data according to the preset actual intracranial pressure data; and performing error analysis on the measurement data according to the measurement data interval to obtain measurement error data, wherein the measurement error data calculation formula is: Wherein, r is the measurement error data, c is the measurement data, x k is the kth actual intracranial pressure data in the measurement data interval, the number of n interval data; Generating measurement data feedback semantics according to the measurement error data, including: determining a feedback information type according to the measurement error data and a preset measurement error threshold; generating a feedback statement template according to the feedback information type; generating measurement data feedback semantics according to the feedback statement template; The detection data of the intracranial pressure is determined according to a preset detection threshold, and the optimal measurement data of the intracranial pressure is calculated according to the feedback semantics of the measurement data and the detection data.

2. The method for measuring intracranial pressure based on ultrasound technology according to claim 1, characterized in that: The collecting of sound wave signal data in intracranial pressure includes: Obtain the scanning area and scanning depth within the intracranial pressure; Determine the data location within the intracranial pressure according to the scanning area and the scanning depth; Configure ultrasound acquisition parameters through preset ultrasound algorithms; Data is collected at the data positioning position according to the ultrasonic collection parameters to obtain sound wave signal data.

3. The method for measuring intracranial pressure based on ultrasound technology as claimed in claim 1, characterized in that: The performing data enhancement processing on the sound wave signal data to obtain sound wave signal enhanced data includes: Using a preset filtering algorithm to remove noise from the sound wave signal data to obtain first sound wave signal data; Performing signal adjustment on the first sound wave signal data to obtain second sound wave signal data; Performing smoothing processing on the second sound wave signal data to obtain third sound wave signal data; Correction processing is performed on the third sound wave signal data to obtain sound wave signal enhancement data.

4. The method for measuring intracranial pressure based on ultrasound technology according to claim 1, characterized in that: The step of extracting the acoustic signal characteristic data of the acoustic signal enhancement data comprises: Extracting time domain features and frequency domain features of the acoustic signal enhancement data; Extracting statistical features and morphological features of the acoustic signal enhancement data; The time domain features, the frequency domain features, the statistical features and the morphological features are fused using the following preset multi-dimensional feature algorithm to obtain acoustic signal feature data: f={χ1p1,χ2p2,χ3p3,χ4p4} Among them, f is the characteristic data of the acoustic wave signal, χ1 is the time domain weight of the time domain feature, p1 is the time domain feature, χ2 is the frequency domain weight of the frequency domain feature, p2 is the frequency domain feature, χ3 is the statistical weight of the statistical feature, p3 is the statistical feature, χ4 is the morphological weight of the morphological feature, and p4 is the morphological feature.

5. The method for measuring intracranial pressure based on ultrasound technology according to claim 1, characterized in that: The step of screening the acoustic signal characteristic data to obtain acoustic signal screening data includes: The acoustic signal characteristic data is compared with the preset acoustic signal characteristic threshold data according to the following preset characteristic data algorithm to obtain a data comparison symbol value: Wherein, q is the data comparison symbol value, p i is the characteristic value of the i-th characteristic in the acoustic signal characteristic data, w i is the characteristic value of the i-th characteristic in the acoustic signal characteristic threshold data; generating a data symbol sequence according to the data comparison symbol value; The data symbol sequence is screened according to a preset target numerical quantity to obtain acoustic wave signal screening data.

6. A system for measuring intracranial pressure based on ultrasound technology, characterized in that: Used to implement the method for measuring intracranial pressure based on ultrasound technology as described in any one of claims 1 to 5, the system comprises: The acoustic signal data enhancement module is used to collect the acoustic signal data in the intracranial pressure, perform data enhancement processing on the acoustic signal data, and obtain acoustic signal enhanced data; An acoustic signal characteristic data module, used for extracting the acoustic signal characteristic data of the acoustic signal enhancement data, and performing data screening on the acoustic signal characteristic data to obtain acoustic signal screening data; An acoustic wave signal screening data module, used to detect the acoustic wave signal screening data to obtain intracranial pressure measurement data; A measurement data feedback semantics generating module, used for performing error analysis between the measurement data and preset actual intracranial pressure data to obtain measurement error data, and generating measurement data feedback semantics according to the measurement error data; The optimal measurement data calculation module is used to determine the detection data of the intracranial pressure according to a preset detection threshold, and calculate the optimal measurement data of the intracranial pressure according to the feedback semantics of the measurement data and the detection data.

Images