Portable ultrasonic imaging system and method integrating AI and motion sensor

By integrating AI and motion sensors on portable ultrasound devices, real-time ultrasound image analysis and operation navigation is achieved, the problem that traditional devices cannot provide real-time operation guidance is solved, and the accuracy and efficiency of diagnosis is improved. It is suitable for primary medical institutions.

CN118873166APending Publication Date: 2024-11-01SHANGHAI SOUNDWISE TECHNOLOGY CO LTD
View PDF 0 Cites 1 Cited by

Patent Information

Application Number
CN202410953402.7
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-16
Publication Date
2024-11-01

AI Technical Summary

Technical Problem

Traditional portable ultrasound devices cannot provide real-time operational guidance and diagnostic support, resulting in missed or misdiagnosed during operation in primary medical institutions or inexperienced doctors.

Method used

Using a portable ultrasound imaging system that combines AI and motion sensors, data is collected through motion sensors on the ultrasound probe, combined with image processing models and data analysis models, providing real-time operation navigation suggestions and diagnostic support, including image recognition, operation navigation instructions and diagnostics hint.

Benefits of technology

It improves the stability and reliability of ultrasound examination, reduces the missed diagnosis rate, improves the accuracy and efficiency of diagnosis, and provides a portable and easy-to-use intelligent assist system suitable for various medical environments.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118873166A_ABST
    Figure CN118873166A_ABST
Patent Text Reader

Abstract

The invention provides a portable ultrasonic imaging system and method fusing AI and a motion sensor. A motion sensor module arranged on an ultrasonic probe collects motion data and position data of the ultrasonic probe; processing the collected motion data and position data to obtain motion feature information and position feature information of the ultrasonic probe; preprocessing the ultrasonic echo signal to obtain an ultrasonic image; processing the ultrasonic image based on a pre-trained image processing model and obtaining an image recognition result; and comprehensively analyzing the motion feature information, the position feature information and the image recognition result based on a pre-trained data analysis model, and outputting an operation navigation suggestion of the portable ultrasonic equipment. A portable and easy-to-use intelligent auxiliary system is provided, and the system is suitable for various medical environments, especially basic medical institutions.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the fields of portable ultrasound technology and ultrasound image processing technology, and in particular to a portable ultrasound imaging system and method integrating AI and motion sensors. Background Art

[0002] Ultrasound imaging technology, as a non-invasive and real-time medical imaging diagnostic tool, has occupied an indispensable position in clinical practice. It uses high-frequency sound waves to penetrate human tissues and construct images of internal organs and structures through reflected wave signals, thereby providing doctors with important information about the patient's health status. Although it has many advantages in diagnosis, such as high safety, relatively low cost, and real-time monitoring. In contemporary medical practice, portable ultrasound devices, as representatives of non-invasive instant imaging technology, are becoming increasingly important. This technology uses high-frequency sound waves to penetrate biological tissues to achieve non-invasive visualization of the internal structure of the human body, especially in emergency treatment, bedside examinations, and medical care in remote areas. Although portable ultrasound has greatly broadened the boundaries of medical diagnosis with its small size and strong mobility, it still faces several technical and application limitations. Traditional ultrasound imaging relies on the experience and skills of the operating doctor, and has a certain degree of subjectivity and variability, especially when it is performed by primary medical institutions or inexperienced doctors, which may lead to missed or misdiagnoses. In addition, most existing ultrasound imaging systems require complex post-processing steps and cannot provide real-time operation guidance and diagnostic support. Summary of the invention

[0003] Based on the above problems, the present invention provides a portable ultrasound imaging system and method integrating AI and motion sensors, which aims to solve the technical problems in the prior art that portable ultrasound equipment cannot provide real-time operation guidance and diagnostic support.

[0004] A portable ultrasound imaging system integrating AI and motion sensors, comprising:

[0005] An ultrasonic probe, used for transmitting ultrasonic signals and receiving ultrasonic echo signals, wherein the ultrasonic probe is provided with a motion sensor module, and the motion sensor module is used for collecting motion data and position data of the ultrasonic probe;

[0006] A motion processing module, connected to the motion sensor module, is used to process the collected motion data and position data to obtain motion feature information and position feature information of the ultrasonic probe;

[0007] An image forming module, connected to an ultrasonic probe, for preprocessing ultrasonic echo signals to obtain ultrasonic images;

[0008] An image processing module, connected to the image forming module, for processing the ultrasound image based on a pre-trained image processing model and obtaining an image recognition result;

[0009] The data analysis module is connected to the motion processing module and the image processing module respectively, and is used to comprehensively analyze the motion feature information, the position feature information and the image recognition result based on the pre-trained data analysis model, and output the operation navigation suggestion of the portable ultrasound device, and the operation navigation suggestion includes the operation navigation instruction of the ultrasound probe;

[0010] The display processing module is connected to the image processing module and the data analysis module respectively, and is used for real-time display of ultrasound images, image recognition results and operation navigation instructions.

[0011] Furthermore, the motion sensor module includes:

[0012] An accelerometer, used to collect acceleration data from the ultrasound probe;

[0013] A gyroscope, used to collect angular velocity data of the ultrasound probe;

[0014] A position sensor, used for collecting position data of the ultrasonic probe in three-dimensional space;

[0015] The motion processing module is used to fuse and correct the acceleration data, angular velocity data and position data to obtain the operation characteristic information and position characteristic information of the ultrasonic probe.

[0016] Furthermore, the image processing model includes an image feature extraction model and an image segmentation model;

[0017] The image processing module includes:

[0018] An image feature extraction unit, used for extracting image feature information of the ultrasound image based on the image feature extraction module;

[0019] An image segmentation unit, connected to the image feature extraction unit, is used to process the image feature information based on the image segmentation model to identify key anatomical structures and lesion features, and form an image recognition result;

[0020] An image annotation unit, connected to the image segmentation unit, for annotating image recognition results on the ultrasound image;

[0021] The display processing module is used to display the ultrasound image annotated with the image recognition result.

[0022] Furthermore, the data analysis module is also used to output diagnostic prompt information after comprehensive analysis using the data analysis model, and the diagnostic prompt information includes diagnostic questions;

[0023] The display processing module is also used to display diagnostic prompt information.

[0024] Furthermore, it also includes:

[0025] The medical consultation record module is used to record the answer information of the person being tested after the person is asked questions according to the medical consultation questions, and together with the questions asked, form the medical condition inquiry data.

[0026] Furthermore, it also includes:

[0027] The disease diagnosis module is connected to the motion processing module, the image processing module and the medical consultation record module respectively, and is used to comprehensively process the image recognition results, and / or the operation characteristic information, and / or the position characteristic information, and the disease inquiry data based on the pre-trained disease diagnosis model, and output a comprehensive diagnosis suggestion;

[0028] The display processing module is connected to the disease diagnosis module and is used to display comprehensive diagnosis suggestions.

[0029] Furthermore, it also includes:

[0030] A parameter optimization module, connected to the motion processing module and the image processing module respectively, for adaptively processing motion feature information, position feature information and image recognition results based on a pre-trained parameter optimization model to form ultrasound optimization parameters;

[0031] The scanning control module is connected to the parameter optimization module and the ultrasonic probe respectively, and is used to adjust the scanning parameters of the ultrasonic probe according to the ultrasonic optimization parameters, and drive the ultrasonic probe to transmit ultrasonic signals and receive ultrasonic echo signals according to the adjusted scanning parameters.

[0032] Furthermore, the display processing module superimposes the operation navigation instructions on the ultrasound image for display.

[0033] Furthermore, it also includes:

[0034] A voice receiving module is used to receive the voice answer information of the subject when the subject is asked questions according to the medical questions;

[0035] The medical consultation record module is connected to the voice receiving module and is used to record the voice answer information and form the medical condition inquiry data.

[0036] A portable ultrasound imaging method integrating AI and motion sensor, using the aforementioned portable ultrasound imaging system integrating AI and motion sensor, comprising:

[0037] Step A1, the ultrasonic probe transmits an ultrasonic signal and receives an ultrasonic echo signal, and a motion sensor module provided on the ultrasonic probe collects operation data and position data of the ultrasonic probe;

[0038] Step A2, processing the collected operation data and position data to obtain operation characteristic information and position characteristic information of the ultrasonic probe;

[0039] Step A3, preprocessing the ultrasonic echo signal to obtain an ultrasonic image;

[0040] Step A4, processing the ultrasound image based on the pre-trained image processing model and obtaining an image recognition result;

[0041] Step A5, comprehensively analyzing the motion feature information, the position feature information and the image recognition result based on the pre-trained data analysis model, and outputting operation navigation suggestions for the portable ultrasound device, wherein the operation navigation suggestions include operation navigation instructions for the ultrasound probe;

[0042] Step A6, displaying the ultrasound image, image recognition results and operation navigation instructions in real time.

[0043] The beneficial technical effects of the present invention are: the present invention uses sensor fusion technology to achieve real-time ultrasound image analysis and processing on portable ultrasound equipment, provide operation navigation and image optimization suggestions, improve system stability and reliability, automatically identify key anatomical structures and lesion characteristics in ultrasound images, provide real-time diagnosis support, reduce doctors' missed diagnosis rate, improve diagnosis accuracy, consistency and efficiency, and provide a portable, easy-to-use intelligent assistance system suitable for various medical environments, especially primary medical institutions. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1-4 This is a module schematic diagram of a portable ultrasound imaging system integrating AI and motion sensors according to the present invention;

[0045] Figure 5-8 This is a flowchart of the steps of a portable ultrasonic imaging method integrating AI and motion sensors of the present invention. DETAILED DESCRIPTION

[0046] 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.

[0047] 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.

[0048] 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.

[0049] See also Figure 1 The present invention provides a portable ultrasonic imaging system integrating AI and motion sensor, comprising:

[0050] An ultrasonic probe (1) is used to transmit ultrasonic signals and receive ultrasonic echo signals, the ultrasonic probe is provided with a motion sensor module (2), and the motion sensor module (2) is used to collect motion data and position data of the ultrasonic probe;

[0051] A motion processing module (3) is connected to the motion sensor module (2) and is used to process the collected motion data and position data to obtain motion feature information and position feature information of the ultrasonic probe;

[0052] An image forming module (4), connected to the ultrasonic probe (1), and used for preprocessing the ultrasonic echo signal to obtain an ultrasonic image;

[0053] An image processing module (5), connected to the image forming module (4), for processing the ultrasound image based on a pre-trained image processing model and obtaining an image recognition result;

[0054] A data analysis module (6) is connected to the motion processing module (3) and the image processing module (5) respectively, and is used to comprehensively analyze the motion feature information, the position feature information and the image recognition result based on a pre-trained data analysis model, and output an operation navigation suggestion of the portable ultrasound device, wherein the operation navigation suggestion includes an operation navigation instruction for the ultrasound probe;

[0055] The display processing module (7) is connected to the image processing module (5) and the data analysis module (6) respectively, and is used to display the ultrasound image, image recognition results and operation navigation instructions in real time. The ultrasound probe (1) is used to transmit and receive ultrasound waves and collect ultrasound images inside the human body. The probe has a built-in micro motion sensor module (2) that can monitor the movement and position of the ultrasound probe in real time.

[0056] The image processing module (5) uses an integrated high-performance processor and / or a dedicated image processing chip to process ultrasound images in real time. The image processing module (5) uses a modular design to facilitate upgrades and maintenance.

[0057] The display processing module (7) is a high-resolution touch screen display, which is used to display ultrasound images, image recognition results and operation navigation instructions in real time. The display processing module (7) has a multi-touch function and supports gesture operation, which is convenient for doctors to interact during the examination process.

[0058] The present invention uses sensor fusion technology to achieve real-time ultrasound image analysis and processing on portable ultrasound equipment, provide real-time operation navigation and image optimization suggestions, improve system stability and reliability, automatically identify key anatomical structures and lesion characteristics in ultrasound images, provide real-time diagnosis support, reduce doctors' missed diagnosis rate, improve diagnosis accuracy, consistency and efficiency, and provide a portable, easy-to-use intelligent assistance system suitable for various medical environments, especially primary medical institutions.

[0059] See also Figure 2 , further, the motion sensor module (2) comprises:

[0060] An accelerometer (21) for collecting acceleration data of the ultrasonic probe;

[0061] A gyroscope (22), used for collecting angular velocity data of the ultrasonic probe;

[0062] A position sensor (23), used for collecting position data of the ultrasonic probe in three-dimensional space;

[0063] The motion processing module (3) is used to fuse and correct the acceleration data, angular velocity data and position data to obtain the operation characteristic information and position characteristic information of the ultrasonic probe.

[0064] The accelerometer detects the acceleration change of the probe to help determine the moving direction and speed of the ultrasound probe. The gyroscope measures the angular velocity of the ultrasound probe to help determine the rotation angle and direction of the probe. The position sensor accurately locates the position of the ultrasound probe in three-dimensional space, and combines the data of the accelerometer and gyroscope to provide complete ultrasound probe motion trajectory information. Data fusion: The motion processing module (3) fuses and corrects the data of each sensor through a certain data fusion algorithm to generate accurate probe motion characteristics and position characteristics to provide operation navigation and feedback, guide the movement and adjustment of the probe, and ensure the best image acquisition angle and position.

[0065] By fusing data from accelerometers, gyroscopes, and position sensors, the system can accurately monitor the movement and position of the ultrasound probe, providing more accurate operational navigation and feedback. This sensor fusion technology improves system stability and reliability and reduces operational errors.

[0066] See also Figure 3 , further, the image forming module (4) comprises:

[0067] An image forming unit (41), used for processing the ultrasonic echo signal to form an initial image;

[0068] An image denoising unit (42), connected to the image forming unit (41), and used for performing denoising processing on the initial image;

[0069] An image enhancement unit (43), connected to the image denoising unit (42), for performing enhancement processing on the initial image after denoising processing;

[0070] The standardization unit (44) is connected to the image enhancement unit (43) and is used to perform standardization processing on the enhanced initial image to form an ultrasonic image.

[0071] The image denoising unit (42) uses a filtering algorithm to remove noise from the ultrasound image and improve the image quality. The image enhancement unit (43) uses contrast enhancement, edge enhancement and other technologies to highlight the details in the image. The standardization unit (44) performs standardization processing on the image to ensure the uniformity of the input image and facilitate subsequent analysis.

[0072] See also Figure 4 ,Furthermore, the image processing model includes an image feature extraction model and an image ,segmentation model;

[0073] The image processing module (5) comprises:

[0074] An image feature extraction unit (51), used for extracting image feature information of an ultrasound image based on an image feature extraction module;

[0075] An image segmentation unit (52), connected to the image feature extraction unit (51), is used to process the image feature information based on the image segmentation model to identify key anatomical structures and pathological features, and form an image recognition result;

[0076] An image annotation unit (53), connected to the image segmentation unit (52), for annotating the image recognition result on the ultrasound image;

[0077] The display processing module (7) is used to display the ultrasonic image annotated with the image recognition result.

[0078] The image processing module (5) integrates AI processing functions, image feature extraction models and image segmentation models, for example, using a pre-trained convolutional neural network (CNN) model. The model has been trained with a large number of medical image data sets and can perform real-time analysis while collecting ultrasound images, extract multi-level feature information, and accurately identify key anatomical structures and lesion features.

[0079] The image feature extraction unit (51) uses a pre-trained image feature extraction model to extract multi-level features and capture the details and global information of the image. The model includes multiple convolutional layers, pooling layers and fully connected layers, and can extract spatial features and semantic information of the image layer by layer. Through multi-scale feature extraction technology, image information at different scales is captured to improve the robustness and accuracy of feature extraction.

[0080] The multi-level feature information includes texture features, edge features, shape features, etc. The calculation parameters include echo intensity, tissue density, etc.

[0081] The image segmentation unit (52) uses a deep learning model based on a convolutional neural network to automatically identify key anatomical structures in ultrasound images, such as the heart, liver, and kidneys. The model is trained with a large amount of labeled data and has high recognition accuracy and robustness. It uses an image segmentation algorithm to segment and identify different anatomical structures in ultrasound images. Through the deep learning algorithm, lesion features such as tumors, cysts, and thrombi are extracted from ultrasound images. The model can identify information such as the morphology, size, and location of the lesions. The image annotation unit (53) annotates the identified key anatomical structures on the ultrasound image and annotates the identified lesion features, including the lesion area, morphology, size, and location, etc., to generate a visual anatomical structure diagram.

[0082] Identification of key anatomical structures can identify organ types, organ boundaries, and specific anatomical landmarks.

[0083] For example:

[0084] Organ type: such as liver, kidney, heart, etc.

[0085] Identify organ boundaries: such as the boundaries of the liver, kidneys, heart, etc.

[0086] Identify specific anatomical landmarks: blood vessels, gall bladder, heart valves, etc.

[0087] Lesion feature recognition includes precise positioning of lesion types such as tumors, cysts, and thrombi, as well as the size, shape, and location of the lesion area.

[0088] When the ultrasound image is a dynamic ultrasound image, the image segmentation unit (52) can also analyze the dynamic changes of key anatomical structures, such as the contraction and relaxation of the heart, blood flow velocity, etc.

[0089] Furthermore, the data analysis module (6) is also used to output diagnostic prompt information after comprehensive analysis using the data analysis model;

[0090] The display processing module (7) is also used to display diagnostic prompt information.

[0091] The diagnostic prompt information can be fed back to the doctor through the interactive interface of the display processing module or the voice module to assist the doctor in making diagnostic decisions.

[0092] The data analysis module (6) integrates AI (artificial intelligence) processing functions. The data analysis model adopts a pre-trained convolutional neural network (CNN) model. The model has been trained with a large number of medical image data sets and can perform real-time analysis while collecting ultrasound images and generate corresponding operation navigation suggestions and diagnostic prompt information.

[0093] The interactive interface of the display processing module (7) can be designed to be simple and easy to operate, allowing the doctor to view real-time images, receive operation navigation suggestions and diagnostic prompt information during the examination process. The interactive interface also supports touch operation and voice commands, which is convenient for the doctor to operate quickly.

[0094] Furthermore, the diagnostic prompt information includes a diagnostic question, and the system further includes:

[0095] The medical consultation record module (8) is used to record the answer information of the subject after the subject is asked questions according to the medical consultation questions, and together with the question questions, form the medical consultation data.

[0096] Furthermore, it also includes:

[0097] The disease diagnosis module (9) is connected to the motion processing module (3), the image processing module (5) and the medical consultation record module (8) respectively, and is used to comprehensively process the image recognition results, and / or the operation characteristic information, and / or the position characteristic information, and the disease inquiry data based on the pre-trained disease diagnosis model, and output a comprehensive diagnosis suggestion;

[0098] The display processing module (7) is connected to the disease diagnosis module (9) and is used to display comprehensive diagnosis suggestions.

[0099] It should be noted that the disease diagnosis model can also be combined with other inspection and test data, and processed together with image recognition results, and / or operation characteristic information, and / or location characteristic information, and disease inquiry data to form a comprehensive diagnosis recommendation.

[0100] The operation characteristic information and location characteristic information are integrated into the sensor characteristic information. The multimodal data such as image recognition results, disease inquiry data and sensor characteristic information are integrated to generate comprehensive diagnosis suggestions. The system can have a built-in medical knowledge base, including disease characteristics, diagnostic criteria, treatment plans and other information, to assist in generating comprehensive diagnosis suggestions.

[0101] Comprehensive diagnostic recommendations include but are not limited to disease diagnosis results, further examination recommendations, treatment plan recommendations, etc.

[0102] The interactive interface of the display processing module (7) can display comprehensive diagnostic suggestions, which can also be fed back to the operator on site, such as a doctor, through the voice module feedback function, so that the doctor can obtain diagnostic information in real time during the operation, thereby further improving the diagnostic efficiency.

[0103] Furthermore, it also includes:

[0104] A parameter optimization module (10) is connected to the motion processing module (3) and the image processing module (5) respectively, and is used to adaptively process the motion feature information, the position feature information and the image recognition result based on a pre-trained parameter optimization model to form an ultrasound optimization parameter;

[0105] The scanning control module (11) is respectively connected to the parameter optimization module (10) and the ultrasonic probe (1), and is used to adjust the scanning parameters of the ultrasonic probe according to the ultrasonic optimization parameters, and drive the ultrasonic probe to transmit ultrasonic signals and receive ultrasonic echo signals according to the adjusted scanning parameters.

[0106] Based on real-time image analysis results and sensor data, the scanning parameters of the ultrasound device, such as gain, focus, depth, etc., are adaptively adjusted to optimize image quality.

[0107] The parameter optimization model can also use reinforcement learning algorithms to continuously optimize the image parameter adjustment strategy based on feedback information under different inspection conditions to ensure the best image quality under various conditions.

[0108] The types of scanning parameters that can be automatically adjusted include but are not limited to:

[0109] Gain: Automatically adjust the gain according to the brightness and contrast of the image to ensure image clarity.

[0110] Focus: Dynamically adjust the focus position based on the probe's position and depth information to optimize image quality at a specific depth.

[0111] Depth: Automatically adjusts the scanning depth based on the identified anatomical structure and lesion location to ensure visibility of important areas.

[0112] Frequency: Automatically select the optimal ultrasound frequency based on the scanning area and image requirements, balancing resolution and penetration.

[0113] Frame Rate: Adjust the frame rate according to the dynamic changes of the image to ensure the smoothness of real-time display.

[0114] For example, when the probe moves to the deep area of ​​the liver, the system detects that the image brightness is insufficient. It automatically adjusts the gain to increase the image brightness and adjusts the focus position to ensure clarity in the deep area.

[0115] For example, after identifying a suspected lesion in the liver, the system automatically adjusts the scanning depth to better observe the lesion. According to the characteristics of the lesion, the system automatically selects the appropriate ultrasound frequency to improve the resolution to obtain a more detailed image.

[0116] For example, when performing color Doppler ultrasound examination, the system automatically increases the frame rate to ensure smooth display of dynamic changes in blood flow.

[0117] Furthermore, the display processing module (7) superimposes the operation navigation instructions on the ultrasound image for display.

[0118] For example, the operation navigation instructions can be arrow instructions and / or text prompts to indicate the position and angle adjustment of the ultrasound probe. By superimposing arrow instructions and / or text prompts on the image, the operation navigation instructions are provided to the doctor in real time to guide the operator to adjust the ultrasound probe according to the instructions to ensure the best image acquisition angle and position. The operation navigation instructions include the probe movement direction instructions, such as:

[0119] Anterior-posterior movement: "Move the probe forward to continue scanning the front of the liver" or "Move the probe backward to cover the entire liver area."

[0120] Left-right movement: "Move the probe to the left to cover the left lobe of the liver" or "Move the probe to the right to scan the right lobe."

[0121] Operational navigation instructions include probe rotation angle instructions, such as:

[0122] Lateral rotation: “Rotate the probe 5 degrees clockwise” to better view a particular anatomical structure.

[0123] Longitudinal rotation: "Rotate the probe 10 degrees counterclockwise" to obtain the best section view.

[0124] Operational navigation instructions include probe tilt angle instructions, such as:

[0125] Up and down tilt: "Tilt the probe down 15 degrees" to better view deep structures.

[0126] Left-Right Tilt: "Tilt the probe 10 degrees to the left" to optimize image quality.

[0127] Operational guidance instructions include scanning speed and path instructions, such as:

[0128] Scan Speed: "Slow down the scan speed to get a clearer image".

[0129] Scan path: "Move the probe steadily from top to bottom along the specified path."

[0130] Operational navigation instructions include position feedback adjustments, such as:

[0131] Image quality feedback: "The current image quality is poor, it is recommended to adjust the probe position."

[0132] Furthermore, it also includes:

[0133] A voice receiving module (12) is used to receive voice answer information of the subject when the subject is asked questions according to the medical questions;

[0134] The medical consultation recording module (8) is connected to the voice receiving module (12) and is used to record the voice answer information and form the medical condition inquiry data.

[0135] The voice receiving module (12) not only receives the voice question information of the questioner, but also records the voice answer information of the subject, and together they form the condition inquiry data. For example, a doctor asks the subject questions according to the diagnosis prompt, and the question and answer are received by the voice receiving module. The consultation recording module (8) includes a voice processing module, which processes the voice information to form the condition inquiry data and records it.

[0136] See also Figure 5 The present invention also provides a portable ultrasonic imaging method integrating AI and motion sensor, using the aforementioned portable ultrasonic imaging system integrating AI and motion sensor, comprising:

[0137] Step A1, the ultrasonic probe transmits an ultrasonic signal and receives an ultrasonic echo signal, and a motion sensor module provided on the ultrasonic probe collects operation data and position data of the ultrasonic probe;

[0138] Step A2, processing the collected operation data and position data to obtain operation characteristic information and position characteristic information of the ultrasonic probe;

[0139] Step A3, preprocessing the ultrasonic echo signal to obtain an ultrasonic image;

[0140] Step A4, processing the ultrasound image based on the pre-trained image processing model and obtaining an image recognition result;

[0141] Step A5, comprehensively analyzing the motion feature information, the position feature information and the image recognition result based on the pre-trained data analysis model, and outputting operation navigation suggestions for the portable ultrasound device, wherein the operation navigation suggestions include operation navigation instructions for the ultrasound probe;

[0142] Step A6, displaying the ultrasound image, image recognition results and operation navigation instructions in real time.

[0143] Furthermore, in step A1, the collected operation data and position data of the ultrasonic probe include:

[0144] Acceleration data of the ultrasound probe collected using an accelerometer;

[0145] Angular velocity data of the ultrasound probe collected using a gyroscope;

[0146] Position data of the ultrasound probe in three-dimensional space collected using a position sensor;

[0147] In step A2, the acceleration data, angular velocity data and position data are fused and corrected to obtain the operation characteristic information and position characteristic information of the ultrasonic probe.

[0148] See also Figure 6 , further, the step A3 comprises:

[0149] Step A31, processing the ultrasonic echo signal to form an initial image;

[0150] Step A32, performing denoising processing on the initial image;

[0151] Step A33, performing enhancement processing on the initial image after denoising;

[0152] Step A34, performing standardization processing on the enhanced initial image to form an ultrasonic image.

[0153] See also Figure 7 , further, the step A4 comprises:

[0154] Step A41, extracting image feature information of the ultrasound image based on an image feature extraction module;

[0155] Step A42, processing the image feature information based on the image segmentation model to identify key anatomical structures and lesion features, and forming an image recognition result;

[0156] Step A43, marking the image recognition result on the ultrasound image;

[0157] In step A6, the ultrasound image annotated with the image recognition result is displayed.

[0158] Further, in step A5, after comprehensive analysis using the data analysis model, diagnostic prompt information is output;

[0159] In step A6, diagnosis prompt information is also displayed.

[0160] See also Figure 8 Further, the diagnostic prompt information includes a diagnostic question, and the method further includes:

[0161] Step A7, recording the answer information of the subject after asking the subject questions according to the medical questions, forming the medical condition inquiry data.

[0162] Step A8, comprehensively processing the image recognition results, and / or the operation characteristic information, and / or the location characteristic information, and the disease inquiry data based on the pre-trained disease diagnosis model, and outputting a comprehensive diagnosis suggestion;

[0163] Step A9, display comprehensive diagnosis suggestions.

[0164] Furthermore, in step A7, the voice answer information of the subject when the subject is asked questions according to the medical questions is received to form the medical condition inquiry data.

[0165] Specific operation navigation suggestions and guidance cases: Identify the liver and perform operation navigation

[0166] Start the device and place the probe: The operator starts the portable ultrasound device and places the probe on the patient's right upper abdomen.

[0167] Initial scanning and identification of the liver: Processing ultrasound images in real time to identify the boundaries and location of the liver;

[0168] Generate operational navigation suggestions: Generate operational navigation instructions based on the recognition results and motion sensor data: "Move the probe forward 2 cm to continue scanning the front of the liver", "Rotate the probe 5 degrees clockwise for a better view", "Tilt the probe down 10 degrees to view the deep liver structure". If the image quality deteriorates, the system will issue a position suggestion: "The current image quality is poor, it is recommended to move the probe 1 cm to the left and slow down the scanning speed." If an area suspected of a tumor is identified in the liver, the operational navigation suggestion also indicates: "Move the probe back 1 cm to better cover the suspected lesion area, and rotate the probe counterclockwise 3 degrees to obtain the best section image."

[0169] Finally, based on the analysis, comprehensive diagnostic recommendations are provided, including the disease diagnosis result: "A suspected tumor in the liver was identified, and a detailed examination is recommended.", and specific examination recommendations: "Color Doppler ultrasound is recommended to evaluate blood flow."

[0170] Application scenario 1: Abdominal ultrasound examination

[0171] During an abdominal ultrasound examination, the doctor uses a portable device to connect to the ultrasound probe and collect ultrasound images in real time. The system monitors the movement and position of the probe through the motion sensor module, generates operation navigation suggestions based on the image features, and prompts the doctor to adjust the angle and position of the probe. At the same time, the system prompts the doctor to ask the patient whether he has symptoms such as abdominal pain and nausea, and records the patient's answer. The system automatically adjusts the image parameters to ensure the best image quality. Finally, the system combines ultrasound images, sensor data, and condition inquiry data (and can also be combined with other laboratory test results) to generate comprehensive diagnostic suggestions (such as hepatitis, cholecystitis, etc.), and feedback to the doctor through the interface or voice prompts.

[0172] Application scenario 2: Cardiac ultrasound examination

[0173] During a cardiac ultrasound examination, the doctor uses a portable device to connect to the ultrasound probe and collect ultrasound images in real time. The system monitors the movement and position of the probe through the motion sensor module, generates operation navigation suggestions based on the image features, and prompts the doctor to adjust the probe position and angle. At the same time, the system prompts the doctor to ask the patient whether he has symptoms such as chest pain and palpitations, and records the patient's answer. The system automatically adjusts the image parameters to ensure the best image quality. Finally, the system combines ultrasound images, sensor data, condition inquiries, and other electrocardiogram results to generate comprehensive diagnostic suggestions (such as myocardial infarction, valvular heart disease, etc.), and feedback to the doctor through the interface or voice prompts.

[0174] Application scenario 3: Thyroid ultrasound examination

[0175] During the thyroid ultrasound examination, the doctor uses a portable device to connect the ultrasound probe to collect ultrasound images of the thyroid area in real time. The system monitors the movement and position of the probe through the motion sensor module, generates operation navigation suggestions based on the real-time image analysis results, and prompts the doctor to adjust the angle and position of the probe to obtain the best inspection angle. The system automatically identifies the anatomical structure of the thyroid gland and detects nodules, cysts and other abnormal lesions. By combining the condition inquiry data (such as whether the patient has symptoms such as neck swelling and difficulty swallowing), the system generates comprehensive diagnostic suggestions (such as thyroid nodules, thyroiditis, etc.) and feedback to the doctor through the interface or voice prompts.

[0176] Application scenario 4: Carotid artery ultrasound examination

[0177] During the carotid ultrasound examination, the doctor uses a portable device to connect the ultrasound probe to collect ultrasound images of the carotid artery in real time. The system monitors the movement and position of the probe through the motion sensor module and guides the doctor to adjust the probe to obtain the best vascular imaging. The system automatically identifies the structure of the carotid artery and detects lesions such as plaques, stenosis and arteriosclerosis. Combined with the inquiries about the condition (such as whether the patient has symptoms such as dizziness and transient ischemic attack) and blood test results, the system generates comprehensive diagnostic suggestions (such as atherosclerosis, carotid stenosis, etc.) and feedback to the doctor through the interface or voice prompts.

[0178] Application scenario 5: Gynecological ultrasound examination

[0179] During a gynecological ultrasound examination, the doctor uses a portable device to connect the ultrasound probe to collect ultrasound images of the uterus and ovarian areas in real time. The system monitors the movement and position of the probe through the motion sensor module, generates operation navigation suggestions based on the real-time image analysis results, and prompts the doctor to adjust the angle and position of the probe to obtain the best inspection angle. The system automatically identifies anatomical structures such as the uterus and ovaries, and detects uterine fibroids, ovarian cysts and other abnormal lesions. Combined with the patient's inquiry data (such as abnormal menstruation, lower abdominal pain and other symptoms), the system generates comprehensive diagnostic suggestions (such as uterine fibroids, ovarian cysts, pelvic inflammatory disease, etc.), and feedback to the doctor through the interface or voice prompts.

[0180] The present invention significantly reduces the missed diagnosis rate and misdiagnosis rate during ultrasound examination by implementing real-time image processing, operation navigation and multi-sensor data fusion on portable devices. Combined with multimodal data fusion technology, it provides comprehensive diagnostic support and improves the comprehensiveness and accuracy of diagnosis. The system is portable and easy to use, and is suitable for various medical environments, especially primary medical institutions. Specific technical effects include:

[0181] 1. Real-time operation navigation and feedback are realized during ultrasound examination, which improves image quality and examination efficiency.

[0182] 2. Through sensor data fusion, provide more accurate operation navigation and reduce operation errors.

[0183] 3. Automatically adjust image parameters to ensure the best image quality under different scanning conditions.

[0184] 4. Combine multimodal data to provide comprehensive diagnostic suggestions and improve the comprehensiveness and accuracy of diagnosis.

[0185] 5. Portable design, suitable for various medical environments, especially primary medical institutions.

[0186] The above are only preferred embodiments of the present invention, and are not intended to limit the implementation methods and protection scope of the present invention. Those skilled in the art 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 portable ultrasound imaging system integrating AI and motion sensors, characterized in that: include: An ultrasonic probe, used for transmitting ultrasonic signals and receiving ultrasonic echo signals, wherein the ultrasonic probe is provided with a motion sensor module, and the motion sensor module is used for collecting motion data and position data of the ultrasonic probe; A motion processing module, connected to the motion sensor module, for processing the collected motion data and position data to obtain motion feature information and position feature information of the ultrasound probe; An image forming module, connected to the ultrasonic probe, and used for preprocessing the ultrasonic echo signal to obtain an ultrasonic image; An image processing module, connected to the image forming module, configured to process the ultrasound image based on the pre-trained image processing model and obtain an image recognition result; a data analysis module, connected to the motion processing module and the image processing module respectively, for comprehensively analyzing the motion feature information, the position feature information and the image recognition result based on a pre-trained data analysis model, and outputting an operation navigation suggestion of the portable ultrasound device, wherein the operation navigation suggestion includes an operation navigation instruction for the ultrasound probe; The display processing module is connected to the image processing module and the data analysis module respectively, and is used to display the ultrasound image, the image recognition result and the operation navigation instruction in real time.

2. The portable ultrasound imaging system integrating AI and motion sensor as claimed in claim 1, characterized in that: The motion sensor module comprises: An accelerometer, used to collect acceleration data of the ultrasonic probe; A gyroscope, used for collecting angular velocity data of the ultrasonic probe; A position sensor, used for collecting position data of the ultrasonic probe in three-dimensional space; The motion processing module is used to fuse and correct the acceleration data, angular velocity data and position data to obtain the operation characteristic information and position characteristic information of the ultrasonic probe.

3. The portable ultrasound imaging system integrating AI and motion sensor as claimed in claim 1, characterized in that: The image processing model includes an image feature extraction model and an image segmentation model; The image processing module comprises: An image feature extraction unit, configured to extract image feature information of the ultrasound image based on the image feature extraction module; An image segmentation unit, connected to the image feature extraction unit, for processing the image feature information based on the image segmentation model to identify key anatomical structures and lesion features, and forming the image recognition result; an image annotation unit, connected to the image segmentation unit, and configured to annotate the image recognition result on the ultrasound image; The display processing module is used to display the ultrasound image annotated with the image recognition result.

4. The portable ultrasound imaging system integrating AI and motion sensor as claimed in claim 1, characterized in that: The data analysis module is also used to output diagnostic prompt information after comprehensive analysis using the data analysis model; The display processing module is also used to display the diagnosis prompt information.

5. The portable ultrasonic imaging system integrating AI and motion sensor as claimed in claim 4, characterized in that: The diagnostic prompt information includes a diagnostic question, and the system further includes: The medical consultation record module is used to record the answer information of the subject after the subject is asked questions according to the medical consultation questions, and together with the question questions, form the medical condition inquiry data.

6. A portable ultrasonic imaging system integrating AI and motion sensor as claimed in claim 5, characterized in that: Also includes: a disease diagnosis module, connected to the motion processing module, the image processing module and the medical consultation record module, respectively, for comprehensively processing the image recognition result, and / or the operation characteristic information, and / or the position characteristic information, and the medical consultation data based on a pre-trained disease diagnosis model, and outputting a comprehensive diagnosis suggestion; A display processing module is connected to the disease diagnosis module and is used to display the comprehensive diagnosis suggestion.

7. The portable ultrasound imaging system integrating AI and motion sensor as claimed in claim 1, characterized in that: Also includes: A parameter optimization module, connected to the motion processing module and the image processing module respectively, for adaptively processing the motion feature information, the position feature information and the image recognition result based on a pre-trained parameter optimization model to form ultrasound optimization parameters; The scanning control module is respectively connected to the parameter optimization module and the ultrasonic probe, and is used to adjust the scanning parameters of the ultrasonic probe according to the ultrasonic optimization parameters, and drive the ultrasonic probe to transmit ultrasonic signals and receive ultrasonic echo signals according to the adjusted scanning parameters.

8. The portable ultrasound imaging system integrating AI and motion sensor as claimed in claim 1, characterized in that: The display processing module superimposes the operation navigation instruction on the ultrasound image for display.

9. The portable ultrasound imaging system integrating AI and motion sensor as claimed in claim 5, characterized in that: Also includes: A voice receiving module, used for receiving the voice answer information of the subject when the subject is asked questions according to the medical questions; The medical consultation recording module is connected to the voice receiving module and is used to record the voice answer information and form medical condition inquiry data.

10. A portable ultrasound imaging method integrating AI and motion sensor, characterized in that: A portable ultrasound imaging system integrating AI and motion sensor as claimed in any one of claims 1 to 9, comprising: Step A1, the ultrasonic probe transmits an ultrasonic signal and receives an ultrasonic echo signal, and a motion sensor module provided on the ultrasonic probe collects operation data and position data of the ultrasonic probe; Step A2, processing the collected operation data and the position data to obtain operation characteristic information and position characteristic information of the ultrasonic probe; Step A3, preprocessing the ultrasonic echo signal to obtain an ultrasonic image; Step A4, processing the ultrasound image based on the pre-trained image processing model and obtaining an image recognition result; Step A5, comprehensively analyzing the motion feature information, the position feature information and the image recognition result based on a pre-trained data analysis model, and outputting an operation navigation suggestion for the portable ultrasound device, wherein the operation navigation suggestion includes an operation navigation instruction for the ultrasound probe; Step A6: display the ultrasound image, the image recognition result and the operation navigation instruction in real time.

Citation Information

Cited By

  • B ultrasonic image data processing method and operation platform thereof

    CN120089322A