Universal ultrasonic equipment interface system

Abstract

The invention relates to the technical field of ultrasonic equipment, in particular to a universal ultrasonic equipment interface system. Comprising a universal interface module which is used for being adaptively connected with various types of ultrasonic equipment; the interface identification module is connected with the universal interface module and is used for identifying the model of the ultrasonic equipment and outputting configuration parameters according to the characteristics of the ultrasonic equipment; and the data processing module is connected with the universal interface module and is used for receiving the ultrasonic data of the ultrasonic equipment and converting the ultrasonic data into a data format supported by an external artificial intelligence system. The method can be compatible with ultrasonic equipment of different brands and models, simplifies the equipment access process, simplifies the data transmission and processing process, facilitates the access of the ultrasonic equipment to the artificial intelligence system, improves the compatibility and analysis efficiency of the artificial intelligence system, and reduces the development process of the artificial intelligence system.

Description

Technical Field

[0001] The present invention relates to the technical field of ultrasonic equipment, and in particular to a universal ultrasonic equipment interface system. Background Art

[0002] Most of the ultrasound equipment interfaces currently on the market are proprietary interfaces and lack unified standards. The interfaces of ultrasound equipment of different brands and models are incompatible, and the data formats output by ultrasound equipment vary, such as DICOM, JPEG, BMP, etc. Even different models of equipment of the same brand may use different data formats, which results in the need for the subsequent artificial intelligence system to develop a separate data processing method for each device when analyzing ultrasound data, increasing the complexity of data processing. In addition, the communication protocols used by each ultrasound device are not unified, and the artificial intelligence system has communication compatibility issues when receiving ultrasound data. Integrating ultrasound equipment with the AI ​​system requires a lot of development work, which is time-consuming and costly. Developers need to adapt and develop each device separately, resulting in decentralized and inefficient development work. Summary of the invention

[0003] The purpose of the present invention is to provide a universal ultrasound equipment interface system to solve the above technical problems;

[0004] The technical problem solved by the present invention can be achieved by adopting the following technical solutions:

[0005] A universal ultrasound equipment interface system, comprising:

[0006] Universal interface module, used to adapt and connect various types of ultrasonic equipment;

[0007] An interface identification module, connected to the universal interface module, used to identify the model of the ultrasonic device and output configuration parameters according to the characteristics of the ultrasonic device;

[0008] A data processing module is connected to the universal interface module and is used to receive the ultrasonic data of the ultrasonic device and convert the ultrasonic data into a data format supported by an external artificial intelligence system.

[0009] Preferably, the data processing module includes a pre-trained support vector machine recognition model, which is used to classify and predict the feature vectors of the ultrasound data and output the data format category of the ultrasound data. The data processing module converts the ultrasound data into a data format supported by the artificial intelligence system based on the data format category recognition result.

[0010] Preferably, the universal interface module is provided with a replaceable adapter or interface board for adapting and connecting to various models of the ultrasonic equipment. The universal interface module also includes a communication protocol conversion unit for being compatible with the communication protocol of the ultrasonic equipment.

[0011] Preferably, the interface identification module includes an identification chip and a sensor, the sensor is used to obtain physical information of the ultrasound device, and the identification chip identifies the model of the ultrasound device based on the physical information and outputs the configuration parameters.

[0012] Preferably, the data processing module includes:

[0013] A data format recognition unit, wherein the support vector machine recognition model is pre-trained in the data format recognition unit;

[0014] a data format conversion unit, connected to the data format identification unit, and configured to convert the ultrasound data into a data format supported by the artificial intelligence system;

[0015] The data cleaning unit is used to clean the ultrasonic data after the format conversion.

[0016] Preferably, the data format recognition unit includes a feature extraction subunit for extracting features from the ultrasound data and constructing the feature vector, and using the feature vector as an input of the support vector machine recognition model.

[0017] Preferably, the feature extraction subunit extracts features including a histogram of the ultrasound data, file header information, file size and file extension.

[0018] Preferably, the data cleaning unit adopts a median filtering method to clean the ultrasonic data after the conversion format. The data cleaning unit is provided with a filtering window that can slide on the ultrasonic data. The filtering window sorts all pixel values ​​within the window by size and replaces the pixel value at the center of the window with the sorted median.

[0019] Preferably, the data processing module further includes:

[0020] A data preprocessing unit, connected to the data cleaning unit, and configured to perform image enhancement and separation processing on the cleaned ultrasonic data to generate the preprocessed ultrasonic data;

[0021] A data compression unit is connected to the data preprocessing unit and is used to compress the preprocessed ultrasonic data to generate compressed ultrasonic data for transmission to the artificial intelligence system.

[0022] Preferably, it also includes,

[0023] A communication module, through which the artificial intelligence system is connected to the data processing module to receive the ultrasound data;

[0024] A database is connected to the data processing module and is used to store and manage the ultrasound data.

[0025] Beneficial effects of the present invention: Due to the adoption of the above technical solution, the present invention is compatible with ultrasound equipment of different brands and models, simplifies the equipment access process as well as the data transmission and processing process, facilitates the access of ultrasound equipment to the artificial intelligence system, thereby improving the compatibility and analysis efficiency of the artificial intelligence system and reducing the development process of the artificial intelligence system. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 is an architecture diagram of a universal ultrasound equipment interface system in an embodiment of the present invention;

[0027] Figure 2 This is a schematic diagram of the architecture of a universal interface module in an embodiment of the present invention;

[0028] Figure 3 Schematic diagram of the data processing module architecture in an embodiment of the present invention.

[0029] In the attached figure: 1. Universal interface module; 11. Communication protocol conversion unit; 12. Adapter; 13. Interface board; 2. Interface identification module; 3. Data processing module; 31. Data format identification unit; 32. Data format conversion unit; 33. Data cleaning unit; 34. Data preprocessing unit; 35. Data compression unit; 4. Communication module; 5. Database; 6. Ultrasonic equipment; 7. Artificial intelligence system. DETAILED DESCRIPTION

[0030] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0031] It should be noted that, in the absence of conflict, the embodiments of the present invention and the features in the embodiments may be combined with each other.

[0032] The present invention will be further described below in conjunction with the accompanying drawings and specific embodiments, but they are not intended to limit the present invention.

[0033] A universal ultrasound equipment interface system, such as Figure 1 As shown, including,

[0034] A universal interface module 1, used for adapting and connecting to various types of ultrasonic equipment 6;

[0035] The interface identification module 2 is connected to the universal interface module 1 and is used to identify the model of the ultrasonic device 6 and output configuration parameters according to the characteristics of the ultrasonic device 6;

[0036] The data processing module 3 is connected to the universal interface module 1 and is used to receive the ultrasonic data from the ultrasonic device 6 and convert the ultrasonic data into a data format supported by the external artificial intelligence system 7.

[0037] Specifically, the present invention aims to provide an intelligent ultrasound equipment interface system that is compatible with multiple brands and models, which can be compatible with ultrasound equipment 6 of different brands and models, simplify the equipment access process and simplify the data transmission and processing process, and facilitate the ultrasound equipment 6 to access the artificial intelligence (AI) system, thereby improving the compatibility and analysis efficiency of the artificial intelligence system 7 and reducing the development process of the artificial intelligence system.

[0038] The artificial intelligence system 7 connected to the present invention is used to analyze ultrasound image data, for example, to identify lesion features in ultrasound data and mark the location of lesion points through a machine learning model. The artificial intelligence system 7 of the present invention can simply and conveniently receive ultrasound data from ultrasound devices 6 of different models, thereby improving the computational compatibility and analysis efficiency of the artificial intelligence system 7.

[0039] In a preferred embodiment, the data processing module 3 includes a pre-trained support vector machine recognition model, which is used to classify and predict the feature vectors of the ultrasound data and output the data format category of the ultrasound data. The data processing module 3 converts the ultrasound data into a data format supported by the artificial intelligence system 7 based on the data format category recognition result.

[0040] Specifically, the present invention integrates a data format recognition algorithm model of a support vector machine (SVM), which can automatically recognize the data format output by the ultrasound device 6, such as DICOM, JPEG, BMP, etc.

[0041] Support vector machine is a supervised learning model mainly used for classification and regression analysis. Its basic idea is to classify data into different categories by finding an optimal hyperplane. For data format recognition, SVM can recognize different image and file formats through feature extraction and training.

[0042] In the present invention, the application steps of the support vector machine recognition model include:

[0043] Step 1, model training, using the training data set to train the support vector machine recognition model, find the best hyperplane to distinguish different data formats. The data set in the present invention includes multiple data formats, and the data samples of each format are marked with their true categories.

[0044] Step 2, data preprocessing, extracting features from the ultrasound data to be identified output by the ultrasound device 6, such as image histogram, file header information, file size, file extension, etc., and constructing the extracted features into feature vectors as input of the SVM;

[0045] Step 3, classification prediction, input the feature vector into the trained support vector machine recognition model, perform classification prediction, and output the data format category.

[0046] The support vector machine recognition model has good classification ability in high-dimensional space, can accurately identify different data formats, and has strong robustness to noise and abnormal data, which is suitable for processing the complex data format recognition task of the present invention. After training with the collected data set again, the support vector machine recognition model can be better generalized to the new data format, and it needs subsequent expansion, and has strong adaptability.

[0047] In a preferred embodiment, the universal interface module 1 is provided with a replaceable adapter 12 or interface board 13 for adapting to and connecting to various models of ultrasonic devices 6. The universal interface module 1 also includes a communication protocol conversion unit 11 for being compatible with the communication protocol of the ultrasonic device 6.

[0048] Specifically, the present invention provides a universal interface module 1 that supports multiple physical connection modes, such as USB, HDMI, Ethernet, Wi-Fi, etc. The universal interface module 1 is connected to the ultrasound device 6 via a pluggable adapter 12 or a multi-function interface board 13 to achieve compatibility with ultrasound devices 6 of different brands and models.

[0049] The communication protocol conversion unit 11 supports multiple communication protocols, such as USB, HDMI, Ethernet, Wi-Fi, etc., to achieve seamless connection with different ultrasound devices 6.

[0050] The universal interface module 1 of the present invention further includes a power management unit for providing a stable power supply to ensure that the ultrasonic device 6 can operate normally.

[0051] In a preferred embodiment, the interface identification module 2 includes an identification chip and a sensor, the sensor is used to obtain physical information of the ultrasonic device 6, and the identification chip identifies the model of the ultrasonic device 6 based on the physical information and outputs configuration parameters.

[0052] Specifically, the sensor uses, for example, a current sensor, a voltage sensor, etc. that identifies the connection status, to obtain current information and voltage information, and thereby identify the model of the ultrasound device 6. The present invention can store the physical information corresponding to the model of each ultrasound device 6 and the working parameters corresponding to the model of the ultrasound device 6 in the database 5, and substitute the physical information into the database 5 through the identification chip to match and identify the model of the ultrasound device 6. For example, if it is detected that the type of the ultrasound device 6 is cardiac ultrasound, the corresponding working parameters are further matched to configure the ultrasound device 6, thereby improving the configuration efficiency of the ultrasound device 6.

[0053] In a preferred embodiment, the data processing module 3 includes:

[0054] A data format recognition unit 31, wherein a support vector machine recognition model is pre-trained in the data format recognition unit 31;

[0055] A data format conversion unit 32, connected to the data format identification unit 31, for converting the ultrasound data into a data format supported by the artificial intelligence system 7;

[0056] The data cleaning unit 33 is used to clean the ultrasonic data after the format conversion.

[0057] Specifically, the data processing module 3 is used to perform data format conversion and data cleaning preprocessing on the ultrasound data transmitted by the ultrasound device 6. The data format conversion unit 32 converts the identified different data formats into the standard format required by the artificial intelligence system 7, such as DICOM and JPEG formats, to ensure the consistency of the data for analysis and processing by the artificial intelligence system 7.

[0058] In a preferred embodiment, the data format recognition unit 31 includes a feature extraction subunit for extracting features from ultrasound data and constructing feature vectors, and using the feature vectors as inputs of a support vector machine recognition model.

[0059] Specifically, the data processing module 3 includes a pre-trained support vector machine recognition model, which is used to classify and predict the feature vectors of the ultrasound data and output the data format category of the ultrasound data. The data processing module 3 converts the ultrasound data into a data format supported by the artificial intelligence system 7 based on the data format category recognition result.

[0060] Specifically, the present invention integrates a data format recognition algorithm model of a support vector machine (SVM), which can automatically recognize the data format output by the ultrasound device 6, such as DICOM, JPEG, BMP, etc.

[0061] In a preferred embodiment, the features extracted by the feature extraction subunit include a histogram of ultrasound data, file header information, file size, and file extension.

[0062] The present invention uses a training data set to train a support vector machine recognition model to find the best hyperplane to distinguish different data formats. The data set in the present invention includes multiple data formats, and the data samples of each format are labeled with their true categories. Features are extracted from the ultrasound data to be recognized output by the ultrasound device 6, such as the histogram of the image, file header information, file size, file extension, etc., and the extracted features are constructed into a feature vector as the input of the SVM. The feature vector is input into the trained support vector machine recognition model, and classification prediction is performed to output the data format category.

[0063] In a preferred embodiment, the data cleaning unit 33 uses a median filtering method to clean the converted ultrasonic data. The data cleaning unit 33 is provided with a filtering window that can slide on the ultrasonic data. The filtering window sorts all pixel values ​​within the window by size and replaces the pixel value at the center of the window with the sorted median.

[0064] Specifically, the data cleaning unit 33 uses a median filtering method to clean the data, remove redundant and noisy data in the ultrasound data, and ensure data quality. Median filtering is a nonlinear filtering technology that effectively removes noise by replacing each pixel value or data point in the ultrasound data with the median value in its neighborhood. The specific steps are as follows:

[0065] Step 1: preset a filter window of a fixed size, such as 3x3 or 5x5, and slide the filter window on each pixel point of the ultrasound data.

[0066] Step 2: sort all pixel values ​​within the filter window by size.

[0067] Step 3: Replace the pixel value at the center of the window with the sorted median value.

[0068] The present invention inputs the ultrasonic data output by the ultrasonic device 6 into the data cleaning unit 33, selects a filter window of size 3x3 or 5x5 according to the image resolution and the noise level, slides the filter window on the ultrasonic data, and performs median calculation on each pixel point; replaces the pixel value at the center of the window with the calculated median, and outputs the filtered ultrasonic data.

[0069] Preferably, the present invention uses median filtering to effectively remove salt and pepper noise in ultrasonic data and can effectively remove isolated noise points in the image. Compared with linear filtering methods such as mean filtering, the present invention uses median filtering to better retain the image edge details of ultrasonic data and will not cause ultrasonic image blur. In addition, the median filtering method is simple to calculate and is suitable for real-time processing, which can meet the real-time requirements of data processing of ultrasonic equipment 6.

[0070] In a preferred embodiment, the data processing module 3 further includes:

[0071] The data preprocessing unit 34 is connected to the data cleaning unit 33 and is used to perform image enhancement and separation processing on the cleaned ultrasonic data to generate preprocessed ultrasonic data;

[0072] The data compression unit 35 is connected to the data preprocessing unit 34 and is used to compress the preprocessed ultrasound data to generate compressed ultrasound data to be transmitted to the artificial intelligence system 7 .

[0073] Specifically, the data preprocessing unit 34 performs necessary data preprocessing when transmitting the ultrasound data to the artificial intelligence system 7, such as image enhancement and segmentation, to provide high-quality input data for the artificial intelligence system 7 and improve the calculation accuracy of the artificial intelligence system 7; the data compression unit 35 is used to compress the data to reduce transmission and storage costs.

[0074] In a preferred embodiment, it also includes:

[0075] The communication module 4, the artificial intelligence system 7 is connected to the data processing module 3 through the communication module 4 to receive the ultrasound data;

[0076] The database 5 is connected to the data processing module 3 and is used to store and manage ultrasound data.

[0077] Specifically, the communication module 4 in the present invention is responsible for data transmission with the artificial intelligence system 7. The communication module 4 of the present invention supports multiple communication protocols to ensure high-speed and stable data transmission. The method execution steps of the communication module 4 are as follows:

[0078] Step 1, selecting a communication protocol according to the requirements of the artificial intelligence system 7;

[0079] Step 2, packaging the ultrasound data into a format suitable for transmission;

[0080] Step 3, transmitting the ultrasound data to the artificial intelligence system 7 via the selected communication protocol.

[0081] The ultrasonic equipment interface system of the present invention provides a software interface to facilitate the integration of the ultrasonic equipment 6 and the artificial intelligence system 7, and realizes the connection identification, data acquisition, format conversion, and data transmission functions of the ultrasonic equipment 6. The ultrasonic equipment interface system is also provided with a status monitoring module, which is used to monitor the operating status of the system or the ultrasonic equipment 6 and provide alarm and log functions.

[0082] The database 5 of the present invention is used to store and manage ultrasound data, including:

[0083] A data storage unit, used for storing ultrasound data in different formats;

[0084] Data index unit, used to establish data index for quick retrieval;

[0085] The data backup unit is used to regularly back up data to ensure data security.

[0086] In a specific embodiment, the present invention needs to build a test environment when deployed to simulate ultrasound equipment 6 of different brands and models to verify the compatibility and stability of the hardware interface and software interface of the present invention. By performing functional tests on the hardware interface and software interface, various functions are verified. The specific test content includes:

[0087] Equipment connection test, used to test the connection of different brands and models of ultrasonic equipment 6;

[0088] Data transmission test, used to test the speed, stability and reliability of data transmission;

[0089] Data processing test, used to test the performance of data processing module 3 and verify the effects of format conversion, data cleaning, data compression and other functions of data processing module 3;

[0090] The test environment also includes performance testing of the ultrasound equipment interface system to evaluate the performance of the ultrasound equipment interface system under high load conditions. The specific test contents include:

[0091] Data transmission speed test: Test the data transmission speed of the ultrasound equipment interface system under high concurrency conditions.

[0092] System stability test, testing the stability of the ultrasonic equipment interface system under long-term operation.

[0093] In a specific embodiment, the ultrasound device 6 of this embodiment uses a handheld ultrasound, and is equipped with an ultrasound probe for collecting ultrasound image data inside the human body, and transmits the ultrasound image data collected by the ultrasound probe to the data processing module 3 by connecting to the universal interface module 1;

[0094] The data format recognition unit 31 of the data processing module 3 is provided with a feature extraction subunit. The feature extraction subunit adopts MCU / FPGA and can extract feature information of ultrasound image data in real time. The features extracted by the feature extraction subunit include the histogram of ultrasound data, file header information, file size and file extension.

[0095] The data processing module 3 uses the trained support vector machine recognition model to find the best hyperplane to distinguish different data formats, and constructs the extracted features into feature vectors as the input of the support vector machine recognition model. The feature vectors are input into the trained support vector machine recognition model for classification prediction and output of the data format category.

[0096] The data format conversion unit 32 converts the identified different data formats into the standard format required by the artificial intelligence system 7, such as DICOM and JPEG formats, to ensure the consistency of the data for analysis and processing by the artificial intelligence system 7.

[0097] The artificial intelligence system 7 connected to the present invention is used to analyze ultrasound image data, for example, to identify lesion features in ultrasound data and mark the location of lesions through a machine learning model. The artificial intelligence system 7 of the present invention can simply and conveniently receive ultrasound data from ultrasound devices 6 of different models, and can provide higher quality ultrasound data to the artificial intelligence system 7, thereby improving the analysis efficiency and calculation accuracy of the artificial intelligence system 7.

[0098] The above description is only a preferred embodiment of the present invention, and does not limit the implementation mode and protection scope of the present invention. For those skilled in the art, it should be aware that all solutions obtained by equivalent substitutions and obvious changes made using the description and illustrations of the present invention should be included in the protection scope of the present invention.

Claims

1. A universal ultrasonic equipment interface system, characterized in that: include, A universal interface module (1) for adapting and connecting to various types of ultrasonic equipment (6); An interface identification module (2), connected to the universal interface module (1), used to identify the model of the ultrasonic device (6) and output configuration parameters according to the characteristics of the ultrasonic device (6); A data processing module (3) is connected to the universal interface module (1) and is used to receive the ultrasound data from the ultrasound device (6) and convert the ultrasound data into a data format supported by the external artificial intelligence system (7).

2. The ultrasound device interface system according to claim 1, characterized in that: The data processing module (3) includes a pre-trained support vector machine recognition model, the support vector machine recognition model is used to classify and predict the feature vectors of the ultrasound data and output the data format category of the ultrasound data. The data processing module (3) converts the ultrasound data into a data format supported by the artificial intelligence system (7) based on the data format category recognition result.

3. The ultrasound device interface system according to claim 1, characterized in that: The universal interface module (1) is provided with a replaceable adapter (12) or an interface board (13) for adapting to and connecting to various models of the ultrasonic devices (6). The universal interface module (1) also includes a communication protocol conversion unit (11) for being compatible with the communication protocol of the ultrasonic devices (6).

4. The ultrasound equipment interface system according to claim 1, characterized in that: The interface identification module (2) comprises an identification chip and a sensor, wherein the sensor is used to obtain physical information of the ultrasonic device (6), and the identification chip identifies the model of the ultrasonic device (6) based on the physical information and outputs the configuration parameters.

5. The ultrasound device interface system according to claim 2, characterized in that: The data processing module (3) comprises: A data format recognition unit (31), wherein the support vector machine recognition model is pre-trained in the data format recognition unit (31); a data format conversion unit (32), connected to the data format identification unit (31), and configured to convert the ultrasound data into a data format supported by the artificial intelligence system (7); A data cleaning unit (33) is used to clean the ultrasonic data after the format conversion.

6. The ultrasound device interface system according to claim 5, characterized in that: The data format recognition unit (31) comprises a feature extraction subunit, which is used to extract features from the ultrasound data and construct the feature vector, and use the feature vector as the input of the support vector machine recognition model.

7. The ultrasound device interface system according to claim 6, characterized in that: The features extracted by the feature extraction subunit include a histogram of the ultrasound data, file header information, file size and file extension.

8. The ultrasound device interface system according to claim 5, characterized in that: The data cleaning unit (33) uses a median filtering method to clean the converted ultrasonic data. The data cleaning unit (33) is provided with a filtering window that can slide on the ultrasonic data. The filtering window sorts all pixel values ​​within the window by size and replaces the pixel value at the center of the window with the sorted median value.

9. The ultrasound device interface system according to claim 5, characterized in that: The data processing module (3) further comprises: A data preprocessing unit (34), connected to the data cleaning unit (33), for performing image enhancement and separation processing on the cleaned ultrasonic data to generate preprocessed ultrasonic data; A data compression unit (35) is connected to the data preprocessing unit (34) and is used to compress the preprocessed ultrasonic data to generate compressed ultrasonic data for transmission to the artificial intelligence system (7).

10. The ultrasound device interface system according to claim 1, characterized in that: Also includes, A communication module (4), wherein the artificial intelligence system (7) is connected to the data processing module (3) via the communication module (4) to receive the ultrasound data; A database (5) is connected to the data processing module (3) and is used to store and manage the ultrasound data.

Images