A remote ultrasound diagnosis support system based on cloud computing

Abstract

The present invention relates to the field of telemedicine technology, and specifically to a remote ultrasound diagnosis support system based on cloud computing. The system includes an echo feature analysis module, a dynamic lesion screening module, a case image comparison module, a lesion trend classification module, and a remote diagnosis support module. In the present invention, by analyzing the echo signal in the remote ultrasound image data stream and calculating the echo change rate between image frames, the ability to capture dynamic features of the lesion area is improved. Combining with the calculation of the gray gradient change value, suspected lesion areas that meet the stable interval of the change rate are screened. Using the contour parameters of the lesion shape edge curve, matching analysis is performed with the ultrasound case image database to enhance the accuracy of lesion category attribution. Analyze the morphological change trend of the lesion area, and combine with the calculation of the trend stability index to achieve lesion evolution classification, ensure the rationality of lesion type classification, and mark the progressive lesion area to make the remote diagnosis more intuitive and improve the auxiliary decision-making ability.

Description

Technical Field

[0001] The present invention relates to the technical field of telemedicine, and in particular, to a remote ultrasound diagnosis support system based on cloud computing. Background Art

[0002] The technical field of telemedicine includes using information and communication technologies to achieve remote interaction and data sharing of medical services. The core contents include remote diagnosis, remote surgery, remote monitoring, and health management, etc., aiming to overcome the problem of uneven geographical distribution of medical resources and improve the accessibility and efficiency of medical services. In a telemedicine system, technologies such as wireless communication, cloud computing, big data processing, and artificial intelligence analysis are usually adopted, enabling patients to receive real-time diagnosis and health monitoring from medical institutions in different locations. At the same time, this field involves multiple aspects such as medical image transmission, data encryption and privacy protection, remote control, and interactive diagnosis to ensure the security and real-time nature of medical data, and is widely applied to scenarios such as ultrasonic image diagnosis, electrocardiogram monitoring, and pathological analysis to support cross-regional medical image analysis and clinical decision-making assistance.

[0003] Among them, a remote ultrasound diagnosis support system based on cloud computing refers to an ultrasonic image acquisition, transmission, processing, and analysis system constructed relying on cloud computing technology. The system covers digital acquisition of ultrasonic signals, efficient encoding and transmission of ultrasonic images, cloud-based medical image storage and management, remote ultrasonic image analysis, and diagnosis support. Specifically, an ultrasonic probe is used to convert echo signals into digital data, and after reducing bandwidth occupancy through data compression methods, it is transmitted to the cloud. The cloud server stores and classifies ultrasonic images using a medical image storage format, and performs image analysis through feature extraction and template matching. Doctors can view ultrasonic images through a remote access interface and use image segmentation and pattern recognition methods to make a diagnostic judgment on the lesion area. In addition, the system uses a medical image archiving protocol for data management, supports multi-terminal real-time access and remote consultation to provide an auxiliary decision-making basis for ultrasonic diagnosis.

[0004] In the existing remote ultrasound diagnosis process, the identification of dynamic changes in lesions relies on single-frame images or simple inter-frame comparison, making it difficult to accurately obtain the minute dynamic features of the lesion area, resulting in some lesion areas being difficult to distinguish from normal tissues. The lesion screening method uses a fixed feature threshold, lacking refined calculation of the changing trend of lesions, resulting in early lesion areas not being effectively identified. Case image matching mainly relies on global features, lacking precise comparison of morphological edge contours, affecting the accuracy of lesion attribution. The time series analysis ability is insufficient, failing to fully quantify the evolution trend of lesions, making the assessment of the lesion development status uncertain. The remote medical terminal device lacks an intuitive progressive lesion annotation in the presentation of lesion classification results, making it difficult for doctors to quickly grasp the lesion development situation during remote diagnosis and affecting the diagnosis efficiency of ultrasonic images. Summary of the Invention

[0005] The object of the present invention is to solve the disadvantages existing in the prior art, and to propose a remote ultrasonic diagnosis support system based on cloud computing.

[0006] In order to achieve the above object, the present invention adopts the following technical solutions: A remote ultrasonic diagnosis support system based on cloud computing includes:

[0007] The echo feature analysis module obtains the echo signal data in the remote ultrasonic image data stream, extracts the echo intensity value of the lesion area in the image frame, calculates the echo change rate between consecutive image frames, analyzes the echo intensity at different angles through cloud computing, calculates the gradient change in the time series, screens the echo signal area whose change rate exceeds the dynamic echo gradient reference value, and outputs the echo change rate feature data;

[0008] The dynamic lesion screening module extracts the gray gradient change value of the lesion area in consecutive image frames based on the echo change rate feature data, calculates the standard deviation of the change rate, screens the suspected lesion area, and generates the dynamic screening area of the lesion;

[0009] The case image comparison module calculates the contour parameters of the lesion shape edge curve based on the dynamic screening area of the lesion, combines the ultrasonic case image database, analyzes the similarity of the morphological structure of the screening area, determines the attribution of the lesion category, and generates the lesion image matching record;

[0010] The lesion trend classification module analyzes the morphological change trend of the lesion area in the ultrasonic image time series according to the lesion image matching record, obtains the change trend slope within a time period, calculates the trend stability index, classifies the lesion type, and generates the lesion evolution classification result.

[0011] As a further solution of the present invention, the echo change rate feature data includes echo intensity distribution data, echo change gradient data, and abnormal echo area. The dynamic screening area of the lesion includes the suspected lesion boundary, gray gradient stable area, and standard deviation range of the change rate. The lesion image matching record includes lesion morphological feature parameters, edge curve contour matching degree, and ultrasonic case image similarity analysis result. The lesion evolution classification result includes lesion morphological change trend, trend stability index, and lesion type classification information.

[0012] As a further solution of the present invention, the echo feature analysis module includes:

[0013] The echo signal extraction sub-module obtains the echo signal data in the remote ultrasonic image data stream, detects the echo intensity value of the lesion area in the image frame, arranges the echo signals of consecutive image frames in a time series, and generates the echo signal data of the lesion area;

[0014] The echo change calculation sub-module, based on the echo signal data of the lesion area, uses the formula:

[0015] ;

[0016] Calculate the echo intensity change rate between adjacent image frames , and generate echo change rate data, where represents the echo intensity value of the th frame, represents the echo intensity value of the th frame, represents the echo intensity value of the th frame, represents the number of image frames;

[0017] The dynamic gradient screening sub-module, based on the echo change rate data, calculates the gradient change value in the time series, screens the echo signal area whose change rate exceeds the dynamic echo gradient reference value, and generates echo change rate feature data.

[0018] As a further solution of the present invention, the dynamic lesion screening module includes:

[0019] The gray-scale gradient calculation sub-module, based on the echo change rate feature data, detects the gray-scale value of the lesion area through consecutive image frames, calculates the gray-scale change amount between adjacent frames, obtains the gray-scale gradient change value of the lesion area in the image frame sequence, and calculates and outputs the gray-scale gradient change rate;

[0020] The change rate standard deviation calculation sub-module calculates the mean value of the gray-scale gradient change rate according to the gray-scale gradient change rate, statistically calculates the sum of the squared deviations of each image frame relative to the mean value, and calculates the change rate standard deviation;

[0021] The stable interval screening sub-module, based on the change rate standard deviation, sets the upper and lower limit thresholds of the stable interval, screens the lesion areas of all image frames, extracts the suspected lesion areas that meet the stable interval range, and uses the formula:

[0022] ;

[0023] Calculate the screening stability index , screen the suspected lesion areas that meet the stable rate interval, and generate the dynamic screening area of the lesion, where represents the gray-scale gradient change rate of the th frame, represents the mean value of the gray-scale gradient change rate, represents the standard deviation of the gray-scale gradient change rate, represents the weight coefficient of the current frame in the sequence, represents the number of image frames.

[0024] As a further solution of the present invention, the case image comparison module includes:

[0025] Based on the dynamic screening area of the lesion, the morphological edge calculation sub-module detects the edge contour of the lesion area, calculates the curvature change, contour closure degree and area ratio of the edge curve, and obtains the morphological edge parameters of the lesion;

[0026] The structural similarity analysis sub-module compares the morphological edge parameters of the lesion with the morphological parameters in the ultrasonic case image database, and uses the formula:

[0027] ;

[0028] Calculate and output the morphological matching degree of the lesion , where represents the morphological parameters of the th group of database lesions, represents the morphological parameters of the current lesion area, represents the curvature value of the lesion area, represents the curvature value of the database-matched lesion, represents the number of morphological parameter comparison items in the database, represents the number of lesion curvature comparison items;

[0029] Based on the morphological matching degree of the lesion, the lesion category attribution sub-module sets a category determination threshold, screens the lesion category with the highest matching degree, obtains the corresponding category attribution information, and establishes a lesion image matching record.

[0030] As a further solution of the present invention, the lesion trend classification module includes:

[0031] Based on the lesion image matching record, the lesion morphological trend calculation sub-module detects the morphological changes of the lesion area in the time series, calculates the lesion contour change rate for each time period, fits the morphological change trend curve of the lesion area, and obtains the change trend slope data;

[0032] The trend stability analysis sub-module calculates the standard deviation of the trend slopes of each segment in the time series according to the change trend slope, and uses the formula:

[0033] ;

[0034] Calculate the trend stability index , where represents the change trend slope of the th time period, represents the average value of the change trend slope, represents the lesion morphological change rate, Represents the matching change rate of the lesion morphology, Represents the number of time periods in the time series, Represents the number of lesion matching change items;

[0035] Based on the trend stability index, the lesion type division sub-module sets the change rate and stability division criteria, screens the lesion types that meet the specific trend pattern, and establishes the lesion evolution classification result.

[0036] As a further aspect of the present invention, the system further includes a remote diagnosis support module;

[0037] Based on the lesion evolution classification result, the remote diagnosis support module associates with the remote ultrasonic terminal device, pushes the classification result to the terminal device, marks the progressive change lesion area, and generates remote diagnosis reference data;

[0038] The remote diagnosis reference data includes the lesion progressive change marking result, the remote ultrasonic terminal classification push result, and the remote diagnosis auxiliary analysis information.

[0039] As a further aspect of the present invention, the remote diagnosis support module includes:

[0040] Based on the lesion evolution classification result, the classification result push sub-module detects the matching remote ultrasonic terminal device, extracts the unique identification information of the terminal device, pushes the lesion classification result to the target terminal device, and obtains the remote push data;

[0041] According to the remote push data, the lesion area marking sub-module detects the progressive change lesion area in the classification result, obtains the coordinate information of the lesion area, calculates the regional boundary characteristics, performs image annotation, and generates a progressive lesion mark;

[0042] Based on the progressive lesion mark, the remote diagnosis data integration sub-module extracts the historical classification information of the lesion area, matches the image data of the same type of lesion, counts the change frequency of the lesion area, integrates the image matching information, and establishes the remote diagnosis reference data.

[0043] Compared with the prior art, the advantages and positive effects of the present invention are as follows:

[0044] In the present invention, by analyzing the echo signals in the remote ultrasound image data stream, extracting the echo intensity values of the lesion areas, and calculating the echo change rate between image frames, the ability to capture the dynamic characteristics of the lesion areas is improved. Based on the multi-angle analysis of the echo intensity and the calculation of the gradient changes in the time series, areas with abnormal change rates can be screened out, making the identification of the lesion areas more targeted. Combining the calculation of the gray gradient change values, suspected lesion areas that meet the stable interval of the change rate are screened out to avoid false screening caused by individual differences or image noise. Using the contour parameters of the lesion shape edge curve to perform matching analysis with the ultrasound case image database enhances the accuracy of the lesion category attribution. Based on the analysis of the morphological change trend of the lesion areas in the ultrasound image time series and combining the calculation of the trend stability index, the evolution classification of the lesions is realized to ensure the rationality of the lesion type division. Combining the remote ultrasound terminal device to obtain the classification results and marking the progressive lesion areas makes the remote diagnosis more intuitive and improves the auxiliary decision-making ability of ultrasound diagnosis. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 is the system flowchart of the present invention;

[0046] Figure 2 is the flowchart of the echo feature analysis module of the present invention;

[0047] Figure 3 is the flowchart of the dynamic lesion screening module of the present invention;

[0048] Figure 4 is the flowchart of the case image comparison module of the present invention;

[0049] Figure 5 is the flowchart of the lesion trend classification module of the present invention;

[0050] Figure 6 is the flowchart of the remote diagnosis support module of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0051] In order to make the objectives, technical solutions and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. 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.

[0052] In the description of the present invention, it should be understood that the orientation or positional relationship indicated by the terms "length", "width", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc. is based on the orientation or positional relationship shown in the drawings. It is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and thus cannot be construed as a limitation on the present invention. In addition, in the description of the present invention, the meaning of "a plurality of" is two or more unless otherwise specifically defined.

[0053] Please refer to Figure 1 , a remote ultrasonic diagnosis support system based on cloud computing includes:

[0054] The echo feature analysis module obtains the echo signal data in the remote ultrasonic image data stream, extracts the echo intensity values of the lesion areas in the image frames, calculates the echo change rate between consecutive image frames, analyzes the echo intensities at different angles through cloud computing, calculates the gradient change in the time series, screens the echo signal areas whose change rates exceed the dynamic echo gradient reference value, and outputs the echo change rate feature data;

[0055] The dynamic lesion screening module extracts the gray gradient change values of the lesion areas in consecutive image frames based on the echo change rate feature data, calculates the standard deviation of the change rate, screens the suspected lesion areas that meet the stable change rate interval, and generates the dynamic screening area for lesions;

[0056] The case image comparison module calculates the contour parameters of the lesion morphology edge curve based on the dynamic screening area for lesions, combines the ultrasonic case image database, analyzes the morphological structure similarity of the screening area, determines the attribution of the lesion category for the screening area, and generates the lesion image matching record;

[0057] The lesion trend classification module analyzes the morphological change trend of the lesion area in the ultrasonic image time series according to the lesion image matching record, obtains the change trend slope within the time period, calculates the trend stability index, divides the lesion types according to the change rate and the trend stability index, and generates the lesion evolution classification result;

[0058] The remote diagnosis support module associates with the remote ultrasonic terminal device based on the lesion evolution classification result, pushes the classification result to the terminal device, marks the progressive change lesion area, and generates the remote diagnosis reference data.

[0059] The echo change rate characteristic data includes echo intensity distribution data, echo change gradient data, and abnormal echo regions. The lesion dynamic screening regions include suspected lesion boundaries, gray gradient stable regions, and the range of standard deviation of change rates. The lesion image matching records include lesion morphological characteristic parameters, edge curve contour matching degrees, and ultrasonic case image similarity analysis results. The lesion evolution classification results include lesion morphological change trends, trend stability indices, and lesion type classification information. The remote diagnosis reference data includes lesion progressive change marking results, remote ultrasonic terminal classification push results, and remote diagnosis auxiliary analysis information.

[0060] Please refer to Figure 2 , the echo feature analysis module includes:

[0061] The echo signal extraction sub-module obtains the echo signal data in the remote ultrasonic image data stream, detects the echo intensity values of the lesion regions in the image frames, arranges the echo signals of consecutive image frames in a time series, and generates the echo signal data of the lesion regions.

[0062] During the ultrasonic diagnosis process, first, it is necessary to obtain the echo signal data in the remote ultrasonic image data stream. The ultrasonic probe emits ultrasonic waves at a certain frequency (such as 3.5 MHz). The ultrasonic waves propagate in the human tissue and generate reflections when encountering tissue interfaces with different densities, forming echo signals. These echo signals are received by the probe and converted into electrical signals. Subsequently, the analog signals are converted into digital signals through an analog-to-digital converter to form an ultrasonic image data stream. Suppose that during a single scan, the system obtains 1000 image frames, each frame contains 1024×768 pixel points, and each pixel point corresponds to an echo intensity value, representing the tissue characteristics at that position. After obtaining the image data, it is necessary to detect the echo intensity values of the lesion regions in the image frames. First, an image segmentation method is used to distinguish the lesion regions in the image frames from the normal tissue regions. Suppose that in a certain image frame, the lesion region occupies 200×150 pixel points. Next, the echo intensity values of each pixel point in this region are extracted to form a matrix containing 200×150 elements, representing the echo intensity distribution of the lesion region. To make the echo signal data more stable, signal denoising processing is required. For example, median filtering is used to smooth the echo signals and reduce the interference of random noise. Suppose the echo intensity values of a certain pixel point in 5 image frames are 0.78, 0.81, 0.80, 0.79, and 0.82 respectively. Then the echo intensity value after median filtering is 0.80, thereby improving the data stability. Finally, the echo signal data of the lesion regions is generated.

[0063] The echo change calculation sub-module is based on the echo signal data of the lesion regions and uses the formula:

[0064] ;

[0065] Calculate the change rate of echo intensity between adjacent image frames , and generate echo change rate data, where represents the echo intensity value of the th frame, represents the echo intensity value of the th frame, represents the echo intensity value of the th frame, represents the number of image frames;

[0066] Based on the echo signal data of the lesion area, calculate the change rate of echo intensity between adjacent image frames, and construct a time series echo gradient change model. Specifically, for the lesion area in each frame of the image, calculate the difference in echo intensity of the corresponding pixel points between adjacent frames. Assume that the echo intensities of a certain pixel point in the th frame and the th frame are and respectively, then the change rate of echo intensity of this pixel point can be expressed as:

[0067] ;

[0068] Calculate the average value of the change rates of echo intensity of all pixel points in the lesion area to obtain the average echo change rate of this frame of image. Assume that in a certain frame, the average echo change rate in the lesion area is 0.05.

[0069] To enhance the accuracy of the calculation, second-order change gradient calculation is introduced to measure the dynamic change trend of echo intensity, and the echo change rate is calculated using the formula.

[0070] To verify the practical application of the calculation method, assume that the echo intensity changes of a certain lesion area in consecutive image frames are as follows:

[0071] Table 1.1 Echo intensity change data of the lesion area in the image frame (unit: dB)

[0072]

[0073] According to the data in Table 1.1, calculate the echo change rate:

[0074] ;

[0075] ;

[0076] The calculation results show that the echo change rate of this lesion area is 0.0475 dB, which can be used for further analysis of the dynamic evolution of the lesion, and finally generate echo change rate data.

[0077] The dynamic gradient screening sub-module calculates the gradient change values in the time series based on the echo change rate data, screens the echo signal regions where the change rate exceeds the dynamic echo gradient reference value, and generates echo change rate feature data.

[0078] Based on the echo change rate data, calculate the gradient change in the time series to screen out the echo signal regions where the change rate exceeds the dynamic echo gradient reference value. Specifically, for each analysis angle, calculate the gradient of the echo intensity change over time. Assuming that at the 0° angle, the echo intensity values at the 1st second, 2nd second, and 3rd second are 0.80, 0.82, and 0.85 respectively, the gradient change calculation is as follows:

[0079] ;

[0080] To screen out the regions with abnormal echo intensity changes, a dynamic echo gradient reference value needs to be set. The setting of this reference value is based on the large-scale statistical analysis of ultrasonic image data. By extracting the echo gradient change ranges of normal tissue regions and lesion regions, calculating their mean values and standard deviations, the reference value is finally determined. Specifically, select 1000 cases of normal ultrasonic image data, extract echo signals in different tissue regions (such as muscle, fat, liver tissue, etc.), calculate the gradient change rate within the time series, and obtain the mean value and standard deviation , in normal tissues, the mean value of the echo gradient change rate is usually between 0.012 and 0.018, and the standard deviation is about 0.002. Combining the statistical analysis, the reference value is set to , that is:

[0081] ;

[0082] The change trend of this value is related to the tissue type and imaging parameters. If the frequency of the ultrasonic probe increases, for example, from 3.5 MHz to 7.5 MHz, then due to the improved resolution, the local tissue gradient change will be more obvious, and the reference value may need to be adjusted; on the contrary, if the image resolution decreases or the echo signal of the tissue interface is weak, the reference value may be correspondingly reduced to 0.018.

[0083] Assume that the calculation result of a certain region is 0.025. Since , this region is determined to be an echo signal region with an excessive change rate. Finally, echo change rate feature data is generated.

[0084] Please refer to Figure 3 , the dynamic lesion screening module includes:

[0085] Based on the echo change rate feature data, the gray-scale gradient calculation sub-module detects the gray-scale values of the lesion area through consecutive image frames, calculates the gray-scale change amount between adjacent frames, obtains the gray-scale gradient change value of the lesion area in the image frame sequence, and calculates and outputs the gray-scale gradient change rate.

[0086] When analyzing consecutive image frames, it is necessary to first obtain the gray-scale values of the lesion area. For a sequence containing multiple image frames, it is first necessary to extract the gray-scale values for each frame, and set the image frame sequence , where represents the image data of the -th frame. Each image frame is composed of a pixel point matrix, and each pixel point has a gray-scale value . In practical applications, such as the analysis of ultrasonic images, the gray-scale value range of each pixel point is usually , where 0 represents the darkest area and 255 represents the brightest area. For example, in a lung CT image, the lesion area may have the characteristic of high density, and its gray-scale value is often higher than that of the surrounding healthy tissues. Therefore, by detecting the gray-scale values of the pixel points in the lesion area frame by frame, a gray-scale data matrix can be established.

[0087] Calculate the gray-scale change amount between adjacent frames using the formula:

[0088] ;

[0089] where represents the gray-scale difference of the pixel point between two adjacent frames. For example, in a set of image sequences, assume that the gray-scale value of a certain pixel point in the -th frame is 120, and in the -th frame is 135, then the gray-scale change amount of this pixel point . If the gray-scale change amounts of all pixel points are continuously calculated, the gray-scale change data set of the entire image frame sequence can be obtained.

[0090] To further extract the gray-scale gradient change value of the lesion area, it is necessary to calculate the average gray-scale gradient of the entire lesion area:

[0091] ;

[0092] where represents the total number of pixel points in the lesion area. For example, if the lesion area contains 100 pixel points, and the calculated gray-scale change amounts are respectively , then its average value is:

[0093] ;

[0094] This mean value can be used for subsequent analysis to finally obtain the gray-scale gradient change rate.

[0095] Table 2.1 Gray-scale change rate data of image frames

[0096]

[0097] As shown in Table 2.1, the gray-scale gradient change rate data of consecutive image frames are listed for calculating the standard deviation of the change rate and screening the stable interval.

[0098] The standard deviation calculation sub-module of the change rate calculates the mean value of the gray-scale gradient change rate according to the gray-scale gradient change rate, statistically calculates the sum of the squared deviations of each image frame relative to the mean value, and calculates the standard deviation of the change rate;

[0099] Call the previously calculated gray-scale gradient change rate and set a time series , corresponding to the acquisition time of the image frame, and each time point has a gray-scale gradient change rate . In order to analyze the degree of dispersion of the change rate, it is necessary to calculate the mean value of the gray-scale gradient change rate:

[0100] ;

[0101] where is the number of image frames. For example, if the gray-scale gradient change rates calculated in 10 image frames are respectively , then its mean value:

[0102] ;

[0103] Next, calculate the sum of the squared deviations of all frames relative to the mean value:

[0104] ;

[0105] In the above example:

[0106] ;

[0107] Calculate the standard deviation:

[0108] ;

[0109] Substitute the values:

[0110] ;

[0111] Finally, obtain the standard deviation of the change rate.

[0112] The stable interval screening sub-module sets the upper and lower threshold values of the stable interval based on the standard deviation of the change rate, screens the lesion areas of all image frames, extracts the suspected lesion areas within the stable interval range, and uses the formula:

[0113] ;

[0114] Calculate the screening stability index , screen the suspected lesion areas that meet the stable rate interval, and generate the dynamic screening area of the lesion. Among them, represents the gray gradient change rate of the th frame, represents the average value of the gray gradient change rate, represents the standard deviation of the gray gradient change rate, represents the weight coefficient of the current frame in the sequence, represents the number of image frames;

[0115] According to the standard deviation of the change rate calculated above, set the upper and lower threshold values of the stable interval, and the threshold range is:

[0116] ;

[0117] ;

[0118] Among them, is an empirically set factor, and its setting basis lies in the dispersion degree of the gray gradient change rate of the lesion area in the image frame sequence. Usually, it is set according to the standard normal distribution in statistics, that is, the data covers about 95.4% of the data points within the interval of the mean value up and down . If the variation degree of the data is large (that is, the standard deviation is high), then the value can be appropriately increased to expand the screening range. If the data distribution is relatively concentrated ( is small), then the value should be appropriately reduced to avoid over-screening.

[0119] In this example, the total number of image frames is 10 frames, and the calculated mean value , the standard deviation , when is set, the screening interval is calculated as follows:

[0120] ;

[0121] ;

[0122] The theoretical basis for this setting is that the gray gradient change rates of most lesion areas in the image frame sequence should be within this range. If the gray gradient change rate of an individual frame is lower than or higher than , it indicates that the frame may be affected by external interference or belongs to an unstable lesion area. Screen all the image frame data that meet .

[0123] The gray gradient change rate within the range , and the image frames outside the threshold range are excluded .

[0124] Use the formula to further calculate the stability index. Among them, the weight coefficient is set according to the time interval during the acquisition of the frame sequence. For example:

[0125] ;

[0126] Substitute the above data for calculation:

[0127] ;

[0128] The calculation result is:

[0129] ;

[0130] In order to determine whether the image frame sequence belongs to the stable rate interval, it is necessary to set a screening stability threshold , and this value is used to distinguish between stable and unstable lesion areas. The setting of is based on the statistical characteristics of the overall gray change of the image frame sequence. Usually, a value is selected to keep the proportion of the stable lesion area at about 90% or more. The selection of this value needs to be analyzed in combination with multiple groups of image frame data. In this example, by analyzing 10 groups of image frame data of different lesion areas, the average screening stability index of different areas is calculated, and the screening results of the lesion areas under different

[0131] Table 2.2 Screening results under different settings

[0132]

[0133] It can be seen from Table 2.2 that when , 80% of the image frames pass the screening, and only 2 frames are excluded, which indicates that this threshold can ensure a sufficient number of frame data while excluding noise frames, meeting the actual screening requirements. Therefore, in this example, it is set as:

[0134] ;

[0135] Under this threshold, if the calculated , it means that the gray gradient of the image frame sequence changes stably and can be judged as a suspected lesion area. , it means that the image frame sequence may contain many abnormal frames or noise, and the screening conditions need to be further adjusted.

[0136] Finally, remove (Grayscale change is too small, may belong to the background area) and above The frames with grayscale changes too fast may belong to unstable lesion areas or image noise, and the frame data that meet the stability requirements are summarized to generate a dynamic lesion screening area.

[0137] See also Figure 4 , the case image comparison module includes:

[0138] The morphological edge calculation submodule detects the edge contour of the lesion area based on the dynamic lesion screening area, calculates the curvature change, contour closure and area ratio of the edge curve, and obtains the lesion morphological edge parameters;

[0139] Based on the dynamic screening area of the lesion, it is first necessary to extract the edge contour of the lesion area from the continuous image frames, obtain the change trend of the grayscale distribution, use the gradient calculation method to identify the edge pixel point set, and smooth the edge curve to reduce noise interference. Taking a certain ultrasound image data as an example, an image with a resolution of 512×512 pixels is selected, of which the lesion area accounts for about 16%. The edge point coordinate set is extracted by the gradient calculation method, and the contour closure is calculated based on the coordinate data. The contour closure can be defined as:

[0140] ;

[0141] in, is the length of the closed contour, is the total length of the lines between all edge points. If the calculation result , it is considered that the contour of the lesion is close to being closed. Further, calculate the curvature change of the edge curve, set a number of measurement points, measure the change rate of the normal vectors of adjacent pixels, and then obtain the curvature distribution characteristics. Taking a lesion area with a diameter of 20 mm as an example, among 50 sampling points, if the curvature change of 30 points is less than 0.1, the lesion shape tends to be regular. If the curvature change of more than 50% of the points is greater than 0.3, it indicates that the edge of the lesion is irregular. This setting is based on the smoothness of the normal tissue boundary. Usually, the boundary of a benign lesion changes smoothly with a small curvature change, while a malignant lesion often has a lobulated or spiculated structure, resulting in local curvature mutations. For the criterion that the curvature change of 30 points is less than 0.1, it is based on the statistical distribution of the curvature of the normal tissue boundary. More than 90% of the benign lesions in the database meet this criterion, and the criterion that the curvature change of more than 50% of the points is greater than 0.3 is based on the statistical characteristics of malignant lesions. More than 85% of the malignant lesions have this characteristic. Finally, calculate the area ratio, that is, the ratio of the lesion area to its convex hull area:

[0142] ;

[0143] Among them, is the area of the lesion area, is its convex hull area. If the area ratio is less than 0.75, the lesion shape may have irregular expansion characteristics. This setting is based on the analysis of the morphological parameters statistically in the database. The area ratio of benign lesions usually ranges between 0.80 - 0.95, while for malignant lesions, due to the infiltrative growth characteristics, the ratio is less than 0.75. In the statistics of 200 cases, among the lesions with an area ratio < 0.75, 78% were diagnosed as malignant lesions, while among the lesions with a ratio > 0.8, only 12% were diagnosed as malignant. Finally, the morphological edge parameters of the lesion are obtained.

[0144] The structure similarity analysis sub-module compares the morphological edge parameters of the lesion with the morphological parameters in the ultrasound case image database using the formula:

[0145] ;

[0146] Calculate and output the morphological matching degree of the lesion , among which, represents the morphological parameters of the th group of database lesions, represents the morphological parameters of the current lesion area, represents the curvature value of the lesion area, represents the curvature value of the database-matched lesion, represents the number of morphological parameter comparison items in the database, represents the number of lesion curvature comparison items;

[0147] Call the lesion morphological edge parameters, extract the lesion data with high morphological similarity from the ultrasonic case image database, compare the similarity of the contour morphology, and calculate the morphological structure similarity of the matching lesions. It is assumed that the database contains 3000 ultrasonic lesion images, and 10 key morphological parameters are extracted from each image, and the Euclidean distance is used to calculate the similarity between the morphological parameters.

[0148] Taking the lesion morphological parameters as an example, assume that the closure degree C of a certain lesion is 0.82, and the closest data point in the database sample is matched. Assume that its average closure degree = 0.85, then the closure degree deviation of this lesion is:

[0149] ;

[0150] Similarly, if the average curvature of the lesion area is 0.12 and the average curvature of the matching lesion in the database is 0.11, then calculate:

[0151] ;

[0152] Then the morphological matching degree Can be calculated as:

[0153] ;

[0154] Set the morphological matching degree threshold Indicates that the lesions are highly similar, Indicates that the lesions are partially similar, Indicates that there are obvious differences in morphology. This setting is based on the analysis of 1000 image comparisons. The statistical results show that for the lesions with morphological matching degree , 95% have the same pathological category, For the lesions, about 70% have the same pathological category, while For the lesions, only 35% belong to the same category. To ensure rationality, further statistical analysis of the image results of 30 lesions with high matching degree ( ) is compared with the pathological diagnosis data, and the result shows that the accuracy rate of the matching category reaches 96%. In a specific case, if the morphological parameters of the matching lesions in the database are as shown in the following table:

[0155] Table 3.1 Calculation results of morphological matching degree

[0156]

[0157] As shown in Table 3.1, the lesions with a matching degree lower than 0.1 can be regarded as highly similar samples, and the lesions with a matching degree higher than 0.2 may belong to different types. Finally, the morphological matching degree of the lesions is obtained.

[0158] Based on the lesion shape matching degree, the lesion category attribution sub-module sets a category determination threshold, screens the lesion category with the highest matching degree, obtains the corresponding category attribution information, and establishes a lesion image matching record;

[0159] Based on the lesion shape matching degree, compare the classified lesion images in the database and set a matching threshold As the category attribution standard. Suppose there are 50 different categories of lesions marked in the database, and each category contains 50 - 200 case images. According to the lesion shape matching degree, select the top 5 most similar database lesions, count the category proportion, and determine the lesion category using the maximum probability attribution principle. For example, if 3 out of 5 matching lesions belong to category A and 2 belong to category B, then the current lesion is classified as A. If there is a tie, a weighted voting method is used for decision-making. Suppose the matching situation of a certain lesion with the database is as follows:

[0160]

[0161] In this case, the proportion of category A is 3 / 5 and the proportion of category B is 2 / 5. Then the final determination of the lesion category attribution is A. Finally, a lesion image matching record is established.

[0162] Please refer to Figure 5 , the lesion trend classification module includes:

[0163] The lesion shape trend calculation sub-module, based on the lesion image matching record, detects the morphological changes of the lesion area in the time series, calculates the change rate of the lesion contour for each time period, fits the morphological change trend curve of the lesion area, and obtains the change trend slope data;

[0164] Based on the lesion image matching record, call the historical image data and extract the contour features of the lesion area. Sort the image data according to the time series, select the image frames with uniform time intervals, measure the contour parameters of the lesion area, including area, perimeter, concavity, etc., and use the piecewise linear regression method to calculate the lesion morphological change rate in each time period to obtain the change trend slope.

[0165] To refine the calculation process, select the lesion image matching record of a certain case. The area of this lesion at the initial moment is 45.3 mm², and the perimeter is 28.1 mm. At the second time node, the area increases to 50.8 mm² and the perimeter increases to 30.5 mm. Assuming the time interval is 30 days, the morphological change rate is calculated as follows:

[0166] ;

[0167] ;

[0168] The calculated morphological change rates of the lesion are 0.1833 mm² / day and 0.08 mm / day respectively. The morphological change rates for all time periods are fitted to generate the slope of the change trend.

[0169] The trend stability analysis sub-module calculates the standard deviation of the trend slopes for each segment within the time series based on the slope of the change trend, and uses the formula:

[0170] ;

[0171] Calculate the trend stability index , where represents the slope of the change trend for the th time period, represents the mean of the slopes of the change trend, represents the morphological change rate of the lesion, represents the matching change rate of the lesion morphology, represents the number of time periods in the time series, represents the number of lesion matching change items;

[0172] Based on the slope of the change trend, calculate the standard deviation of the trend slopes for each segment within the time series, obtain the mean deviation of the slope changes to measure the stability of the trend changes, and calculate using the formula.

[0173] Suppose a total of H = 6 time points are collected in the time series of a certain lesion, and calculate its slope changes: , , , , , .

[0174] Calculate the mean slope:

[0175] ;

[0176] ;

[0177] ;

[0178] ;

[0179] ;

[0180] ;

[0181] Set the data of the morphological change rate of the lesion:

[0182] , ;

[0183] , ;

[0184] , ;

[0185] , ;

[0186] , ;

[0187] , ;

[0188] Calculate:

[0189] ;

[0190] ;

[0191] ;

[0192] ;

[0193] ;

[0194] ;

[0195] ;

[0196] ;

[0197] Calculate the final trend stability index

[0198] ;

[0199] Table 4.1: Lesion trend slope calculation table

[0200]

[0201] Table 4.2: Lesion morphology change rate calculation table

[0202]

[0203] As shown in Table 4.2, calculate the morphology change rate at each time point, and calculate the trend stability index based on the change trend slope.

[0204] The lesion type division sub-module sets the change rate and stability division criteria based on the trend stability index, screens the lesion types that meet the specific trend patterns, and establishes the lesion evolution classification results;

[0205] Set the classification criteria for lesion types. These criteria are based on the long-term trend data analysis of 500 lesion samples in the ultrasound image database and are divided in combination with the correlation between the image change rate and the clinicopathological diagnosis:

[0206] Stable type: Trend stability index , for lesions of this type, in the follow-up data of ultrasound images, compared with the lesion morphological parameters in the image matching record, the average change rate is less than 0.05 mm² / day, and the standard deviation is less than 0.015. The corresponding pathological test data shows that lesions of this type are mainly benign hyperplasia, cystic lesions or inflammatory reactions.

[0207] Slowly progressive type: , for such lesions, the average morphological change rate is between 0.05 mm² / day and 0.12 mm² / day, and the standard deviation is between 0.015 and 0.03. The matching lesions in its ultrasound image database are mostly low-grade malignancies or local proliferative lesions, such as focal nodules.

[0208] Rapidly progressive type: , for such lesions, the morphological change rate in the continuous image data is higher than 0.12 mm² / day, and the standard deviation is greater than 0.03. Comparing with the image database, more than 70% of the lesion samples meeting this standard are histopathologically confirmed as highly malignant, such as invasive canceration or rapidly spreading tumors.

[0209] Suppose the trend stability indices of a certain lesion at 6 time points are as follows:

[0210] ;

[0211] Calculate its mean value:

[0212] ;

[0213] Calculate the standard deviation:

[0214] ;

[0215] ;

[0216] ;

[0217] ;

[0218] Comparing with the classification criteria, the mean value of the trend stability index of this lesion is less than 0.02, and the standard deviation Less than 0.015, so the lesion is classified as stable, indicating that its morphological changes are small in the imaging time series, conforming to the imaging characteristics of benign hyperplasia or inflammatory reaction, stored in the database, and the lesion evolution classification result is generated.

[0219] Table 4.3: Lesion type classification criteria

[0220]

[0221] As shown in Table 4.3, by calculating the trend stability index and its standard deviation, the evolution category of the lesion is determined, and finally the lesion evolution classification result is established.

[0222] Please refer to Figure 6 , the remote diagnosis support module includes:

[0223] Based on the lesion evolution classification result, the classification result push sub-module detects the matching remote ultrasound terminal device, extracts the unique identification information of the terminal device, pushes the lesion classification result to the target terminal device, and obtains the remote push data;

[0224] Based on the lesion evolution classification result, to detect the matching remote ultrasound terminal device, first, specific lesion type data such as static lesions, progressive lesions, or regressive lesions need to be extracted from the lesion evolution classification result. Then, call the device management database in the telemedicine system to obtain the list of registered remote ultrasound terminal devices, and screen out the devices that meet the current lesion matching requirements. These devices mainly include portable handheld ultrasound devices, hospital fixed ultrasound scanners, remote robotic ultrasound devices, etc. According to specific clinical application requirements, the matching principles of these devices are also different. For example, portable handheld ultrasound devices are suitable for primary hospitals or community clinics, facilitating doctors to follow up suspicious lesions, while hospital fixed ultrasound scanners are mainly used in large medical institutions and are suitable for high-resolution detailed examinations. Remote robotic ultrasound devices are suitable for telemedicine scenarios, and the ultrasound scanning is performed by a remote doctor controlling the device.

[0225] During the screening process, screening is carried out based on parameters such as the geographical location of the device, hospital level, device model, online status, etc. For example, if the lesion is from the long-term monitoring of a certain high-risk patient, the exclusive monitoring device of this patient is preferentially selected for pushing. If the lesion type is a suspicious lesion in the initial screening, it is pushed to all ultrasonic devices with diagnostic permissions in the region. After the screening is completed, the system structures the lesion data into standardized push information, which includes the lesion number, matching image index, evolution trend category, etc., and formats and converts the data. The converted data is encapsulated according to telemedicine standards such as HL7 or DICOM, and the remote communication protocol is called for data transmission. During the transmission process, the AES encryption algorithm is used to protect the security of the lesion data to ensure data integrity and confidentiality. Finally, after the target terminal device receives the pushed data, it verifies the data integrity and generates a push record at the device end to ensure that the data is successfully transmitted and received, and obtains the remotely pushed data.

[0226] Table 5.1 Classification of Remote Ultrasonic Terminal Devices

[0227]

[0228] As shown in Table 5.1, different remote ultrasonic terminal devices are applicable to different scenarios to ensure that the lesion classification results can be pushed and processed on suitable devices.

[0229] The lesion area marking sub-module detects the progressive change lesion area in the classification result according to the remotely pushed data, obtains the coordinate information of the lesion area, calculates the regional boundary features, performs image annotation, and generates progressive lesion marks.

[0230] According to the remotely pushed data, detect the progressive change lesion area in the classification results. First, the remote terminal device parses the pushed data, extracts the lesion number, image index, and their corresponding morphological change parameters, such as lesion area, edge curvature, morphological matching degree, etc. Subsequently, the terminal device calls the historical ultrasound images corresponding to the lesion from local storage or a remote database, compares the lesion contours in time series, and calculates the contour change rate of the lesion in each frame of the image. For progressive lesions, it is necessary to mark their change areas on the image. The specific marking method is to use pixel-level region segmentation technology. After extracting the lesion area, generate an isocurve and overlay a semi-transparent red area in the image for highlighting. At the same time, for the dynamic change of the lesion contour, it is presented in the form of a dynamic heat map in the image, setting color gradients corresponding to different degrees of change. For example, the area with a change rate in the range of 0.01 - 0.05 is marked yellow, the area of 0.05 - 0.1 is marked orange, and the area greater than 0.1 is marked red. In addition, the marked data also includes parameters such as the maximum diameter of the lesion and the decline rate of morphological similarity, so that doctors can quickly judge the change trend of the lesion. This marked data is stored in the terminal device database and can be transmitted back to the central database through a remote connection, finally generating a progressive lesion mark.

[0231] Table 5.2 Calculation results of the lesion area change rate

[0232]

[0233] As shown in Table 5.2, the contour change rates of different lesions correspond to different color marks, which are used for intuitive display on the image to assist doctors in quickly identifying the change trend of the lesions.

[0234] Based on the progressive lesion mark, the remote diagnosis data integration sub-module extracts the historical classification information of the lesion area, matches the image data of similar lesions, counts the change frequency of the lesion area, integrates the image matching information, and establishes remote diagnosis reference data;

[0235] Based on the progressive lesion mark, extract the historical classification information of the lesion area. First, call the storage system to query the historical classification records of this lesion at different time points, including whether it has ever been classified as a progressive, static, or regressive lesion. At the same time, match the image data of similar lesions and calculate the change frequency of the lesion area. The calculation method of the change frequency is to count the fluctuation range of the morphological matching degree of the lesion over time and calculate its standard deviation. If the standard deviation exceeds the set threshold, it is considered that the lesion has abnormal fluctuation characteristics. For example, for a certain lesion within the past 30 days, the standard deviation of its morphological matching degree is calculated as follows:

[0236] ;

[0237] where is the The morphological matching degree of the lesion over days, is the average matching degree, is the number of days in the time series. Assume the historical matching degree data of a lesion is as follows:

[0238] ;

[0239] Calculate its average value:

[0240] ;

[0241] Calculate the variance:

[0242] ;

[0243] ;

[0244] ;

[0245] ;

[0246] If the standard deviation is greater than 0.05, it is considered that the lesion has changed drastically; otherwise, it is considered that its change is stable. The setting basis of this threshold lies in the analysis of the evolution characteristics of the lesion, mainly referring to the fluctuation range of the imaging morphology of the lesion, and combining the morphological matching degree data of a large number of clinical cases to observe the change range of the matching degree over time. Clinical data shows that the standard deviation of the morphological matching degree of most stable lesions is usually between 0.02 and 0.05, while that of progressive lesions is concentrated between 0.06 and 0.12. Therefore, 0.05 is selected as the threshold for drastic change, and this value will be adjusted according to the imaging acquisition time interval, the lesion volume change rate, and the tissue elasticity parameters. For example, if the lesion volume growth rate exceeds 5% / month, the standard deviation threshold can be adjusted to 0.045; if the imaging acquisition time interval is short (such as daily acquisition), the threshold can be appropriately relaxed to 0.055 to adapt to higher-frequency data fluctuations.

[0247] As shown in Table 5.3, the calculation results of the morphological matching degree fluctuations of different lesions are given and used for subsequent remote diagnosis data integration.

[0248] Table 5.3 Calculation Table of Lesion Morphological Matching Degree Fluctuations

[0249]

[0250] As shown in Table 5.3, the standard deviation of the matching degree of different lesions determines whether they are classified as stable or fluctuating lesions. Stable lesions have less change and can be used as follow-up lesions, while fluctuating lesions need further analysis to determine whether there is a malignant change trend.

[0251] The above are only the preferred embodiments of the present invention, and do not limit the present invention in other forms. Any person skilled in the relevant art may use the technical content disclosed above to make changes or modifications into equivalent embodiments with equivalent changes and apply them to other fields. However, as long as it does not depart from the technical solution content of the present invention, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention still fall within the protection scope of the technical solution of the present invention.

Claims

1. A remote ultrasound diagnosis support system based on cloud computing, characterized in that The system includes: The echo feature analysis module acquires the echo signal data in the remote ultrasonic image data stream, extracts the echo intensity values of the lesion areas in the image frames, calculates the echo change rate between consecutive image frames, analyzes the echo intensities at different angles through cloud computing, calculates the gradient changes in the time series, screens the echo signal areas where the change rate exceeds the dynamic echo gradient reference value, and outputs the echo change rate feature data; The dynamic lesion screening module, based on the echo change rate feature data, extracts the gray gradient change values of the lesion areas in consecutive image frames, calculates the standard deviation of the gray change rate, screens the suspected lesion areas, and generates the dynamic lesion screening areas; The case image comparison module, based on the dynamic lesion screening areas, calculates the contour parameters of the lesion shape edge curves, combines with the ultrasonic case image database, analyzes the similarity of the morphological structures of the screening areas, determines the attribution of the lesion categories, and generates the lesion image matching records; The lesion trend classification module, according to the lesion image matching records, analyzes the morphological change trends of the lesion areas in the ultrasonic image time series, obtains the change trend slope within a time period, calculates the trend stability index, classifies the lesion types, and generates the lesion evolution classification results.

2. The remote ultrasound diagnosis support system based on cloud computing according to claim 1, wherein The echo change rate feature data includes echo intensity distribution data, echo change gradient data, and abnormal echo areas. The dynamic lesion screening areas include suspected lesion boundaries, gray gradient stable areas, and the range of the standard deviation of the change rate. The lesion image matching records include lesion morphological feature parameters, edge curve contour matching degrees, and ultrasonic case image similarity analysis results. The lesion evolution classification results include lesion morphological change trends, trend stability indices, and lesion type classification information.

3. The remote ultrasound diagnosis support system based on cloud computing according to claim 1, characterized in that The echo feature analysis module includes: The echo signal extraction sub-module acquires the echo signal data in the remote ultrasonic image data stream, detects the echo intensity values of the lesion areas in the image frames, arranges the echo signals of consecutive image frames in a time series, and generates the echo signal data of the lesion areas; The echo change calculation sub-module, based on the echo signal data of the lesion areas, uses the formula: ； Calculate the change rate of echo intensity between adjacent image frames , and generate echo change rate data, where represents the echo intensity value of the th frame, represents the echo intensity value of the th frame, represents the echo intensity value of the th frame, represents the number of image frames; The dynamic gradient screening sub-module, based on the echo change rate data, calculates the gradient change values in the time series, screens the echo signal areas where the change rate exceeds the dynamic echo gradient reference value, and generates the echo change rate feature data.

4. The remote ultrasound diagnosis support system based on cloud computing according to claim 1, characterized in that, The dynamic lesion screening module includes: The gray gradient calculation sub-module, based on the echo change rate feature data, detects the gray values of the lesion areas through consecutive image frames, calculates the gray change amount between adjacent frames, obtains the gray gradient change values of the lesion areas in the image frame sequence, and calculates and outputs the gray gradient change rate; The standard deviation calculation sub-module of the change rate calculates the mean value of the gray gradient change rate according to the gray gradient change rate, statistically calculates the sum of the squared deviations of each image frame from the mean value, and calculates the standard deviation of the gray change rate; The stable interval screening sub-module, based on the standard deviation of the gray change rate, sets the upper and lower limit thresholds of the stable interval, screens the lesion areas of all image frames, extracts the suspected lesion areas that meet the stable interval range, and uses the formula: ； Calculate and screen stability indicators , screen suspected lesion areas that meet the stable rate interval, and generate a dynamic screening area for lesions, where represents the gray gradient change rate of the th frame, represents the average value of the gray gradient change rate, represents the standard deviation of the gray gradient change rate, represents the weight coefficient of the current frame in the sequence, represents the number of image frames.

5. The remote ultrasound diagnosis support system based on cloud computing according to claim 1, characterized in that The case image comparison module includes: The morphological edge computing sub-module detects the edge contour of the lesion area based on the dynamically screened area of the lesion, calculates the curvature change, contour closure degree and area ratio of the edge curve, and obtains the morphological edge parameters of the lesion; The structural similarity analysis sub-module compares the morphological edge parameters of the lesion with the morphological parameters in the ultrasonic case image database using the formula: ； Calculate the morphological matching degree of the output lesion , where represents the morphological parameters of the lesions in the nth group of the database, represents the morphological parameters of the current lesion area, represents the curvature value of the lesion area, represents the curvature value of the database-matched lesion, represents the number of morphological parameter comparison items in the database, represents the number of lesion curvature comparison items; The lesion category attribution sub-module sets a category determination threshold based on the lesion morphological matching degree, screens the lesion category with the highest matching degree, obtains the corresponding category attribution information, and establishes a lesion image matching record.

6. The remote ultrasound diagnosis support system based on cloud computing according to claim 1, characterized in that, The lesion trend classification module includes: The lesion morphological trend calculation sub-module detects the morphological changes of the lesion area in the time series based on the lesion image matching record, calculates the contour change rate of the lesion in each time period, fits the morphological change trend curve of the lesion area, and obtains the change trend slope data; The trend stability analysis sub-module calculates the standard deviation of the trend slopes in each segment of the time series according to the change trend slope and uses the formula: ； Calculate the trend stability index , where represents the change trend slope of the -th time period, represents the mean value of the change trend slope, represents the lesion morphology change rate, represents the lesion morphology matching change rate, represents the number of time periods in the time series, represents the number of lesion matching change terms; The lesion type division sub-module sets the change rate and stability division criteria based on the trend stability index, screens the lesion types that conform to a specific trend pattern, and establishes the lesion evolution classification result.

7. The remote ultrasound diagnosis support system based on cloud computing according to claim 1, characterized in that The system further includes a remote diagnosis support module; The remote diagnosis support module associates with the remote ultrasonic terminal device based on the lesion evolution classification result, pushes the classification result to the terminal device, marks the progressive change lesion area, and generates remote diagnosis reference data; The remote diagnosis reference data includes the lesion progressive change marking result, the remote ultrasonic terminal classification push result, and the remote diagnosis auxiliary analysis information.

8. The remote ultrasound diagnosis support system based on cloud computing according to claim 7, characterized in that The remote diagnosis support module includes: The classification result push sub-module detects the matching remote ultrasonic terminal device based on the lesion evolution classification result, extracts the unique identification information of the terminal device, pushes the lesion classification result to the target terminal device, and obtains the remote push data; The lesion area marking sub-module detects the progressive change lesion area in the classification result according to the remote push data, obtains the coordinate information of the lesion area, calculates the regional boundary features, performs image annotation, and generates a progressive lesion mark; The remote diagnosis data integration sub-module extracts the historical classification information of the lesion area based on the progressive lesion mark, matches the image data of the same type of lesion, counts the change frequency of the lesion area, integrates the image matching information, and establishes the remote diagnosis reference data.

Images