Lesion development early warning method and system based on ultrasonic image, medium and equipment
By extracting features such as the liver echo attenuation coefficient and liver-kidney ratio based on ultrasound images, the development rate and morphological changes of liver lesions are quantitatively assessed, which solves the subjective problem of early diagnosis of liver disease and achieves early detection and accurate intervention.
Patent Information
- Application Number
- CN202510801988.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-06-16
- Publication Date
- 2025-09-23
AI Technical Summary
In existing technologies, early diagnosis of liver disease relies on doctors' experience and judgment, which is subjective and makes it difficult to accurately monitor the development and changes of liver lesions in real time.
By extracting features such as the liver echo attenuation coefficient and liver-kidney ratio based on ultrasound images, the development rate and morphological changes of liver lesions are quantitatively evaluated, and early warning signals are triggered based on preset thresholds.
It achieves early detection and timely intervention of liver abnormalities, reduces errors in subjective judgment, and improves the accuracy and efficiency of diagnosis.
Smart Images
Figure CN120678470A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the technical field of artificial intelligence analysis of medical images and intelligent medical equipment, and in particular to a lesion development early warning method, system, medium, and equipment based on ultrasound imaging. Background Art
[0002] In the field of medical imaging, ultrasound imaging technology, with its numerous advantages such as non-invasiveness, convenience, and real-time imaging, has become an indispensable tool for clinical diagnosis and disease monitoring. Ultrasound imaging allows doctors to visually observe the morphology, structure, and hemodynamics of internal organs, providing key information for early disease screening, condition assessment, and monitoring of treatment effectiveness.
[0003] Ultrasound imaging also plays a crucial role in the monitoring and management of liver-related conditions. As a crucial metabolic organ in the human body, the health of the liver directly impacts the body's overall physiological function. However, early-stage liver disease often lacks obvious clinical symptoms. By the time noticeable discomfort appears, the condition may have already progressed to a more serious stage, posing a significant challenge to timely intervention and treatment. Currently, clinical assessment of liver condition largely relies on the physician's intuitive observation and empirical judgment of ultrasound images. Physicians observe the liver's morphology, size, echogenicity, and other characteristics, combining them with the patient's medical history and other test results to comprehensively determine whether the liver is abnormal. However, this assessment method is somewhat subjective, and the experience and judgment criteria of different physicians may vary, which can easily lead to deviations in the assessment results.
[0004] In addition, the development of liver disease is a dynamic process. It is difficult to grasp the development and changes of liver lesions in real time and accurately by relying solely on regular ultrasound imaging examinations and doctors' experience judgments. Summary of the Invention
[0005] In view of this, the present application provides a lesion development warning method, system, medium, and equipment based on ultrasound imaging. By extracting features based on ultrasound images and quantitatively evaluating the development rate of liver lesions and the results of morphological changes, it can trigger warnings in a timely manner, which is conducive to the early detection and intervention of abnormal liver conditions.
[0006] According to one aspect of the present application, a method for early warning of lesion development based on ultrasound imaging is provided, the method comprising:
[0007] Obtain ultrasound images that include the liver and kidneys;
[0008] Extracting primary ultrasound image features based on the ultrasound image, and determining a liver lesion development rate based on the extracted primary ultrasound image features, wherein the primary ultrasound image features include a liver echo attenuation coefficient and a liver-kidney ratio;
[0009] Extracting secondary ultrasound image features based on the ultrasound image, and determining a liver morphology change assessment result based on the extracted secondary ultrasound image features, wherein the secondary ultrasound image features include liver morphology, and the liver morphology change assessment result includes abnormal and normal;
[0010] When the determined liver lesion development rate is greater than or equal to a preset lesion development rate warning threshold, and / or the liver morphological change assessment result is abnormal, a lesion development warning signal is triggered.
[0011] According to another aspect of the present application, there is provided a lesion development early warning system based on ultrasound imaging, the system comprising:
[0012] An ultrasound image acquisition module, used for acquiring ultrasound images including the liver and kidneys;
[0013] a primary image feature extraction module, configured to extract primary ultrasound image features based on ultrasound images, and determine the liver lesion development rate based on the extracted primary ultrasound image features, wherein the primary ultrasound image features include the liver echo attenuation coefficient and the liver-kidney ratio;
[0014] a secondary image feature extraction module, configured to extract secondary ultrasound image features based on the ultrasound image, and determine a liver morphology change assessment result based on the extracted secondary ultrasound image features, wherein the secondary ultrasound image features include liver morphology, and the liver morphology change assessment result includes abnormal and normal;
[0015] The lesion development warning module is used to trigger the generation of a lesion development warning signal when the determined liver lesion development rate is greater than or equal to a preset lesion development rate warning threshold, and / or the liver morphological change assessment result belongs to a preset warning change degree.
[0016] According to another aspect of the present application, a medium is provided, on which a computer program is stored. When the program is executed by a processor, the above-mentioned ultrasound image-based lesion development early warning method is implemented.
[0017] According to another aspect of the present application, a device is provided, including a medium, a processor, and a computer program stored on the medium and executable on the processor, wherein when the processor executes the program, the above-mentioned ultrasound image-based lesion development warning method is implemented.
[0018] Through the above-mentioned technical solution, the present application provides a lesion development warning method and system, medium, and equipment based on ultrasound imaging. By extracting features based on ultrasound images, the development rate of liver lesions and the results of morphological changes are quantitatively evaluated, which can trigger warnings in a timely manner and facilitate early detection and intervention of abnormal liver conditions.
[0019] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS
[0020] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:
[0021] Figure 1 A schematic diagram of a process for early warning of lesion development based on ultrasound imaging provided in an embodiment of the present application is shown;
[0022] Figure 2 A schematic diagram of a process for another ultrasound imaging-based lesion development early warning method provided in an embodiment of the present application is shown;
[0023] Figure 3 FIG2 shows a flow chart of another ultrasound imaging-based lesion development early warning method provided in an embodiment of the present application;
[0024] Figure 4 A schematic structural diagram of a lesion development early warning system based on ultrasound imaging provided in an embodiment of the present application is shown. DETAILED DESCRIPTION
[0025] The present application will be described in detail below with reference to the accompanying drawings and in combination with embodiments. It should be noted that, unless there is a conflict, the embodiments and features in the embodiments of the present application can be combined with each other.
[0026] In this embodiment, a method for early warning of lesion development based on ultrasound imaging is provided. Figure 1 As shown, the method includes:
[0027] Step 101: Acquire an ultrasound image containing the liver and kidneys.
[0028] In the above embodiments of the present application, a dynamic monitoring system overall architecture can be constructed to achieve early warning of lesion development. The dynamic monitoring system overall architecture can cover hardware modifications (for example, modification of the Philips IU22 device) and core algorithms. The core algorithm includes the mathematical expression of the drug-image association model, that is, the subsequent liver fat content calculation formula and the liver index decay model, which are used to characterize the degree of liver "decay", thereby preventing lesion risks in advance. Specifically, first, an ultrasound image can be obtained using an ultrasound machine. When obtaining, the patient needs to fast and drink water within 8 hours before the examination to avoid food and gas in the gastrointestinal tract from interfering with the ultrasound image and affecting the clarity of imaging of organs such as the liver and kidneys. Then, when using an ultrasound machine to obtain an ultrasound image, it is necessary to ensure that the various parameters of the ultrasound machine equipment are normal, such as image resolution, gain adjustment, focus depth and other parameters are set within the appropriate range to ensure that high-quality ultrasound images can be obtained. At the same time, an ultrasound probe suitable for abdominal examination can also be selected. For example, a convex array probe can be used, which has a wide field of view and is suitable for observing abdominal organs such as the liver and kidneys. When obtaining, the patient can lie on the examination bed in a supine position to fully expose the abdomen. If needed, the doctor can ask the patient to turn their body to the left or right to better expose the liver and kidney examination areas. The doctor then evenly applies a layer of ultrasound coupling agent to the surface of the ultrasound probe. The coupling agent removes air from between the probe and the skin, reducing the attenuation of the ultrasound signal during transmission and improving the quality of the ultrasound image. The doctor then holds the ultrasound probe and gently places it on the patient's abdominal skin, scanning from different angles and directions. Furthermore, for liver scans, the doctor can start from below the xiphoid process and slowly move the probe downward along the right midclavicular line to observe the size, shape, and margins of the liver. The liver is located in the right upper abdomen and is largely covered by the costal arches. The doctor can adjust the angle and pressure of the probe to minimize obstruction by the ribs and obtain a clear image of the liver. For example, an intercostal oblique cut allows visualization of the right and left internal lobes of the liver, while a transverse section allows visualization of the liver's left-right and anteroposterior diameters.
[0029] After completing the liver scan, the doctor moves the probe to the patient's waist area. For example, they can first locate the outline of the kidneys, starting with a coronal section to understand the kidney's long diameter and morphology. Then, by rotating the probe, they obtain transverse and sagittal sections of the kidneys, comprehensively observing structures such as the renal cortex and medulla. During the scan, the doctor can adjust probe parameters such as depth and gain as needed to obtain the optimal image quality.
[0030] When the doctor observes a clear, complete ultrasound image of the liver and kidneys on the ultrasound system's screen, they press the capture button on the device to capture the currently displayed image. The captured image can be either a static image or a dynamic video clip, allowing for a more comprehensive observation of organ movement and blood flow. The captured ultrasound image is automatically stored in the ultrasound system's storage device. The doctor can also transfer the image to the hospital's Picture Archiving and Communication System (PACS) via the network for subsequent review, analysis, and diagnosis.
[0031] In particular, the final acquired ultrasound image data, ie, the standardized ultrasound images of the liver and kidneys, may have fixed parameters set to a depth of 15 cm, a mechanical index of 1.3, and a total gain of 85%.
[0032] Optionally, the ultrasonic image is acquired through an ultrasonic instrument, which corresponds to a probe pressure sensor and an ultrasonic probe. The probe pressure sensor is integrated in the ultrasonic probe. When the ultrasonic probe is used to acquire the ultrasonic image, the scanning pressure of the ultrasonic probe is monitored based on the probe pressure sensor, and the ultrasonic image initially acquired by the ultrasonic probe is compensated for image deformation based on the monitored scanning pressure, and then the final ultrasonic image is output.
[0033] In the above-mentioned embodiment of the present application, during the ultrasound image acquisition process, the probe pressure sensor is integrated into the ultrasound probe, and provides key data support for image deformation compensation by real-time monitoring of the pressure changes at the contact surface between the probe and the skin. When the doctor holds the ultrasound probe for scanning, the probe pressure sensor continuously collects pressure information and converts it into an electrical signal for transmission to the ultrasound device. The device's built-in algorithm dynamically adjusts the image acquisition parameters based on the pressure data. For example, when it is detected that the probe pressure is too high, resulting in excessive compression of the tissue, the system will automatically reduce the gain compensation intensity to reduce image deformation; conversely, if the pressure is insufficient, the signal compensation will be enhanced to ensure uniform image brightness.
[0034] Taking liver ultrasound examination as an example, when the probe is scanning obliquely between the right ribs, the doctor may unconsciously increase the pressure due to the obstruction of the ribs. At this time, the probe pressure sensor will immediately feedback the pressure value, and the equipment (ultrasound machine) will correct the image of the edge area of the liver through the deformation compensation algorithm to avoid blurred liver contours or distortion of lesion morphology due to uneven pressure. The final output ultrasound image has clearer tissue boundaries and more accurate lesion display, providing a reliable basis for subsequent lesion development rate calculation and morphological evaluation. To this end, through pressure monitoring and real-time compensation, the quality consistency of ultrasound images can be improved, the impact of human operation differences on diagnostic results can be reduced, and the inspection process can be optimized, improving diagnostic efficiency and accuracy.
[0035] In particular, the probe pressure sensor can be integrated into the C5-1 probe. At the same time, the monitored scanning pressure range can be 0.5 to 2.5N to compensate for image deformation.
[0036] Step 102: extracting first-level ultrasound image features based on the ultrasound image, and determining the liver lesion development rate based on the extracted first-level ultrasound image features, wherein the first-level ultrasound image features include the liver echo attenuation coefficient and the liver-kidney ratio.
[0037] Next, the liver echo attenuation coefficient and liver-kidney ratio were extracted from the ultrasound images to calculate the rate of liver lesion progression. The liver echo attenuation coefficient reflects the attenuation of the ultrasound signal as it propagates through liver tissue, while the liver-kidney ratio is the ratio of the grayscale intensity of the liver and kidney cortical regions. Both features are quantitative indicators. Compared to the traditional method of relying on the physician's subjective judgment of the liver lesion in the image, quantitative features can more objectively and accurately reflect the characteristics of liver tissue, reducing errors caused by subjective factors such as physician experience and visual judgment.
[0038] Optionally, in step 102, “extracting first-level ultrasound image features based on the ultrasound image” specifically includes:
[0039] Step 1021: Determine the liver region in the ultrasound image and measurement parameters of the liver echo attenuation coefficient, wherein the measurement parameters include a starting depth and an ending depth, and the starting depth and the ending depth are used to ensure that the ultrasound signal used for detection can cover the liver region where the echo attenuation needs to be measured.
[0040] Step 1022 , automatically collecting echo signals based on the liver region and measurement parameters, and calculating the liver echo attenuation coefficient based on the intensity variation of the echo signals at different depths in the liver region, wherein the echo attenuation coefficient represents the attenuation degree of the ultrasound signal when propagating in the liver tissue.
[0041] Step 1023, based on the ultrasound image, determine the cortical areas of the liver and kidney respectively, calculate the average grayscale intensity of the cortical area of the liver according to the pixel grayscale values of the cortical area of the liver, and calculate the average grayscale intensity of the cortical area of the kidney according to the pixel grayscale values of the cortical area of the kidney.
[0042] Step 1024 , obtaining a liver-kidney ratio based on the ratio of the average grayscale intensity of the liver cortex region to the average grayscale intensity of the kidney cortex region.
[0043] In the above embodiment of the present application, in particular, the cortical regions of the liver and kidney, that is, the region of interest (ROI), the ROI size can be fixed to 30×30 dpi. When calculating the liver echo attenuation coefficient and the liver-kidney ratio, fixed parameters (for example, the depth is fixed at 15 cm, the mechanical index is fixed at 1.3, and the total gain is fixed at 85%) and a fixed time gain can be used to obtain sagittal ultrasound images of the liver and right kidney, and the ultrasound physician collects images of the regions of interest (the cortical regions of the liver and kidney, respectively), and uses NIH image analysis software to calculate the liver echo attenuation coefficient. At the same time, the grayscale intensity of the liver and kidney cortical regions is measured, and the ratio of the two is calculated as the liver-kidney ratio. Specifically:
[0044] To calculate the echo attenuation coefficient of the liver, first ensure that the NIH Image Analysis Software is correctly installed and functioning properly. Then, open the software and create a new project or open an existing one (if one already exists). Import an ultrasound image containing the liver and kidneys into the software. This can be done by selecting the appropriate ultrasound image file from the "File" menu, "Import" or "Open." In the software interface, use selection tools (such as the rectangular selection tool or polygon selection tool) to accurately select the liver region on the ultrasound image. Avoid structures such as blood vessels and bile ducts when selecting the region to ensure accurate measurement. Then, set the echo attenuation coefficient measurement parameters. While different versions of the NIH Image Analysis Software may have slightly different parameter settings, generally, you need to specify the measurement depth range and echo signal acquisition method. For example, set the starting and ending depths to cover the area of the liver where echo attenuation is to be measured. The software automatically acquires echo signals based on the selected liver region and the set measurement parameters. The echo attenuation coefficient of the liver is calculated by analyzing the intensity variations of the echo signals at different depths. The echo attenuation coefficient (EAC) reflects the attenuation of ultrasound signals as they propagate through liver tissue. Its calculation formula is based on the relationship between the logarithmic change in echo signal intensity and depth. Once the calculation is complete, the software displays the liver EAC result on the interface, which the user can record or save.
[0045] Similarly, when measuring the grayscale intensity of the liver and kidney cortical regions, the liver cortical region can be accurately selected on the imported ultrasound image using the selection tool. The liver cortical region typically has specific echogenicity characteristics on ultrasound images, allowing for selection based on these characteristics. The software analyzes the pixel grayscale values of the selected liver cortical region and calculates the average grayscale intensity of that region. Grayscale intensity reflects the brightness of different regions in the ultrasound image and typically ranges from 0 to 255 (for 8-bit grayscale images). After the measurement is complete, the software displays the average grayscale intensity value of the liver cortical region, which the user can record. Similarly, when measuring the grayscale intensity of the kidney cortical region, the selection tool can be used to accurately select the kidney cortical region on the ultrasound image. The kidney cortical region also has unique echogenicity characteristics on ultrasound images, making it easier to select accurately. The software analyzes the pixel grayscale values of the selected kidney cortical region and calculates the average grayscale intensity of that region. After the measurement is complete, the software displays the average grayscale intensity value of the kidney cortical region, which the user can record. The liver-kidney ratio is calculated by dividing the measured mean grayscale intensity of the liver cortex by the mean grayscale intensity of the kidney cortex. For example, if the mean grayscale intensity of the liver cortex is 150 and the mean grayscale intensity of the kidney cortex is 100, the liver-kidney ratio is 150 ÷ 100 = 1.5.
[0046] To this end, the calculated liver-kidney ratio can be recorded and used as one of the primary ultrasound imaging features to subsequently determine the development rate of liver lesions and make judgments in the overall risk warning method.
[0047] Alternatively, as Figure 2 As shown, in step 102, "determining the liver lesion development rate based on the extracted first-level ultrasound image features" specifically includes:
[0048] Step 1025: Obtain the historical medication duration of the patient captured by the ultrasound image.
[0049] Step 1026: Determine the liver fat content based on the liver fat content calculation formula, the liver echo attenuation coefficient, and the liver-kidney ratio.
[0050] In step 1027, the determined liver fat content is used as the starting liver fat content in the liver index decay process. The change in liver fat content over time is determined in combination with the historical medication duration and the liver index decay model. The determined change is used as the liver lesion progression rate. The liver index decay model is obtained by correlating the influence of the historical medication duration and the starting liver fat content on the change in liver fat content over time. The liver fat content calculation formula is:
[0051] Liver fat content = 62.592 × liver-kidney ratio + 168.076 × liver echo attenuation coefficient - 27.863;
[0052] The liver exponential decay model is:
[0053] Q(t)=Q0+α(1-e -βt ),
[0054] Where Q(t) is the liver fat content of the liver that changes with time t, which is used to characterize the liver lesion development rate, Q0 is the initial liver fat content, α is the preset long-term steady-state value, β is the preset metabolic rate constant, e is the Euler number, and e -βt Used to characterize the attenuation degree of liver fat content over time t.
[0055] In the above embodiment of the present application, e is a natural constant (also known as Euler number). The natural constant e is an irrational number, and its value is approximately 2.71828. In the liver index decay model formula, e -βt It constitutes the exponential function part, which is used to describe the attenuation degree of liver fat content changing with time t. The exponential function is used in conjunction with other parameters (Q0, α, β) to determine the change of liver fat content over time. Specifically, a drug metabolism synchronization acquisition module can be constructed in the overall architecture of the dynamic monitoring system to connect with the hospital HIS system to obtain medication records (drug type, dosage, and duration) in real time, that is, to obtain the historical medication duration of the patient captured by ultrasound imaging. In particular, general information can also be obtained, such as gender, age, type of psychiatric medication, duration of medication, dosage, and serum marker data: total cholesterol (TC, mmol / L), triglycerides (TG, mmol / L), alanine aminotransferase (ALT, U / L) and aspartate aminotransferase (AST, U / L), which are used for subsequent analysis of the degree of liver damage caused by the patient taking psychiatric drugs (directly manifested as the incidence of hepatobiliary lesions), such as the derived fatty liver, gallstones, cholecystitis, liver cysts and other lesions, which can be achieved by comparing serum marker levels, that is, the inter-group comparison results of TC, TG, ALT, and AST. The TGC curve can also be automatically optimized based on BMI.
[0056] Step 103: extract secondary ultrasound image features based on the ultrasound image, and determine the liver morphology change assessment result based on the extracted secondary ultrasound image features, wherein the secondary ultrasound image features include liver morphology, and the liver morphology change assessment result includes abnormal and normal.
[0057] Next, changes in liver morphology in ultrasound images are a key characteristic of many liver diseases. By accurately extracting liver morphological features, it is possible to more intuitively observe any abnormal changes in the liver's size, contour, and margins.
[0058] Optionally, the ultrasound image is acquired by an ultrasound apparatus, and the ultrasound image corresponds to a fixed standard section selected during acquisition, wherein the fixed standard section includes at least one of a longitudinal section, a transverse section, a subcostal oblique section, and a right intercostal section for the liver and kidneys. Regarding step 103, "extracting secondary ultrasound image features based on the ultrasound image" specifically includes:
[0059] Step 1031: Determine the liver region in the ultrasound image.
[0060] Step 1032: extracting liver morphological features in the liver region, wherein the liver morphological features include at least one of the size, shape, and capsule integrity of the liver.
[0061] In the above-described embodiments of the present application, a two-dimensional, real-time ultrasound image of the liver can be acquired using a high-frequency ultrasound probe. To comprehensively observe the liver's morphology, multiple standard cross-sectional images can be acquired, including longitudinal, transverse, subcostal oblique, and right intercostal sections. These sections can display different anatomical structures of the liver. For example, the longitudinal section can visualize the left lateral lobe of the liver and the abdominal aorta, while the transverse section can reveal the structures of the first portal vein. After acquiring the ultrasound image, the image can be preprocessed to enhance the contrast between the liver and surrounding tissues. This step typically involves adjusting the image contrast and brightness to enhance the clarity of the liver region. Next, image segmentation techniques are used to accurately segment the liver region from the ultrasound image. Image segmentation is the basis for extracting morphological features, and its accuracy directly impacts subsequent feature extraction. Image segmentation methods can include threshold-based and region-growing methods. After liver region segmentation, morphological features are extracted. These features include liver size (e.g., length, width, thickness), shape (e.g., contour regularity and edge smoothness), and capsule integrity. For example, a normal liver is wedge-shaped, thicker on the right side and gradually thinner on the left lobe. A complete liver has a regular outline, a smooth capsule, and sharp edges. Measuring parameters such as the liver's long diameter, anteroposterior diameter, and transverse diameter can quantify liver size. By observing the liver's outline and edges, the liver's shape can be assessed for normality.
[0062] Furthermore, texture analysis can be combined to quantify subtle textural changes on the liver surface. Texture analysis methods such as gray-level co-occurrence matrix and local binary pattern can capture the texture characteristics of the liver surface and help detect early morphological abnormalities.
[0063] Finally, the extracted features are quantitatively evaluated. By calculating quantitative indicators such as liver volume and measuring the maximum oblique diameter of the right lobe, and comparing these indicators with normal reference values, it is possible to determine whether there are abnormal changes in liver morphology. For example, in cirrhosis, the liver may be smaller, irregular in shape, and have jagged or wavy edges; while liver tumors may cause localized abnormalities in the liver morphology, such as mass formation and raised bumps on the liver surface.
[0064] Optionally, in step 103, “determining the liver morphological change assessment result based on the extracted secondary ultrasound image features” specifically includes:
[0065] Step 1033 , based on the extracted liver morphological features, calculate a liver morphological quantitative index, wherein the liver morphological quantitative index includes liver volume and / or the maximum oblique diameter of the right lobe of the liver, and each liver morphological quantitative index corresponds to a preset normal reference value range.
[0066] Step 1034 : When any liver morphology quantitative index is outside the corresponding preset normal reference value range, the liver morphology change assessment result is determined to be abnormal.
[0067] Step 1035 : When all the quantitative indices of liver morphology are respectively within their corresponding preset normal reference value ranges, the liver morphology change assessment result is determined to be normal.
[0068] In the above embodiments of the present application, the formula for calculating the liver volume, for example, SLV=11.508×BW+334.0 (BW represents body weight), and the normal reference value range, for example: the liver volume of an adult can be set to between 1.5 and 2 liters. For the measurement and evaluation of the maximum oblique diameter of the right lobe of the liver, the measurement method can be: the patient takes a supine position, and the scan is performed under the right costal margin, with the sound beam directed to the right shoulder. The oblique section of the liver under the right costal margin where the right hepatic vein and the middle hepatic vein merge into the inferior vena cava is used as the standard measurement section, and the measurement points are placed on the liver capsule and diaphragm between the anterior and posterior edges of the right lobe of the liver, and the maximum vertical distance is measured. The normal value range of the maximum oblique diameter of the right lobe of the liver can be set to 12 to 14 cm.
[0069] When any liver morphology quantitative index is outside the corresponding preset normal reference value range, the liver morphology change assessment result is determined to be abnormal; when all liver morphology quantitative indexes are within their respective corresponding preset normal reference value ranges, the liver morphology change assessment result is determined to be normal.
[0070] Step 104 : When the determined liver lesion development rate is greater than or equal to a preset lesion development rate warning threshold, and / or the liver morphology change assessment result is abnormal, a lesion development warning signal is triggered.
[0071] Then, when the determined liver lesion development rate is greater than or equal to the preset lesion development rate warning threshold, and / or the liver morphological change assessment result is abnormal, a lesion development warning signal is triggered. By combining the two indicators, risks can be prevented in a timely manner.
[0072] Optionally, the lesion development warning signal includes a primary lesion development warning signal and an advanced lesion development warning signal. Regarding step 104, "triggering generation of a lesion development warning signal" specifically includes:
[0073] Step 1041 , obtaining liver fat content and liver-kidney ratio calculated based on ultrasound images of the liver and kidneys taken over multiple years for the same patient.
[0074] Step 1042: When the annual increase in liver fat content is greater than a preset increase ratio, a primary lesion development warning signal is generated.
[0075] Step 1043: When the liver-kidney ratio is greater than the preset warning ratio for two consecutive times, a high-level lesion development warning signal is generated.
[0076] In the above embodiment of the present application, graded warnings can be performed, such as a yellow warning (corresponding to a warning signal for the development of primary lesions): triggered when the annual increase in liver fat content is >15%, and a red warning (corresponding to a warning signal for the development of advanced lesions): triggered when the liver-kidney ratio is >1.8 and ALT>2×ULN for two consecutive examinations.
[0077] In particular, a critical medication duration warning threshold can also be set. For example, when the drug exposure duration (that is, the historical medication duration) exceeds a certain threshold (such as 5 years), the risk of liver damage increases significantly. At this time, the warning situation is that the warning is triggered when the predicted progression rate (lesion development rate) > threshold or the historical medication duration > critical value (such as 60 months).
[0078] In a specific embodiment, Figure 3As shown, the main steps include ultrasound image acquisition, input of drug metabolism parameters, radiomic feature extraction, and dynamic risk assessment. Specifically, ultrasound images from three consecutive examinations are obtained to extract ultrasound radiomic features, such as the liver-kidney ratio, liver echo attenuation coefficient, and liver fat content (calculated using a formula); serum markers: TC, TG, ALT, and AST levels; and drug exposure parameters (obtained from historical data): medication duration, drug dose, cumulative dose, drug type, and medication regimen (monotherapy / combination). Next, image processing software (such as NIH ImageJ) is used to extract features from the ultrasound images (such as liver echo intensity and liver morphology) and perform numerical conversion of the extracted features. Drug exposure parameters are then converted to numerical form, and time-related parameters are standardized to ensure consistent time units. The radiomic features and drug exposure parameters from three consecutive examinations are input into a time series Transformer network, which then outputs the liver injury progression rate (i.e., lesion development rate) (% / month) and the critical medication duration warning threshold. In particular, drug metabolism hotspot maps can be generated to label the S5 / S6 segments of the right hepatic lobe. Furthermore, an acoustic attenuation compensation matrix can be constructed for different probe pressures (used to correct sound velocity based on pressure sensor data), and a drug artifact simulation database can be developed to generate ultrasound images of clozapine metabolism characteristics based on ex vivo liver perfusion experiments.
[0079] By applying the technical solution of this embodiment, by extracting features based on ultrasound images and quantitatively evaluating the development rate of liver lesions and the results of morphological changes, early warnings can be triggered, which is conducive to the early detection and intervention of abnormal liver conditions.
[0080] Further, as Figure 1 The specific implementation of the method, the embodiment of the present application provides a lesion development early warning system based on ultrasound imaging, such as Figure 4 As shown, the system includes:
[0081] Ultrasonic image acquisition module 201, used to acquire ultrasonic images including liver and kidneys;
[0082] A primary image feature extraction module 202 is configured to extract primary ultrasound image features based on the ultrasound image, and determine the liver lesion development rate based on the extracted primary ultrasound image features, wherein the primary ultrasound image features include the liver echo attenuation coefficient and the liver-kidney ratio;
[0083] A secondary image feature extraction module 203 is configured to extract secondary ultrasound image features based on the ultrasound image, and determine a liver morphology change assessment result based on the extracted secondary ultrasound image features, wherein the secondary ultrasound image features include liver morphology, and the liver morphology change assessment result includes abnormality and normality;
[0084] The lesion development warning module 204 is used to trigger the generation of a lesion development warning signal when the determined liver lesion development rate is greater than or equal to a preset lesion development rate warning threshold, and / or the liver morphological change assessment result belongs to a preset warning change degree.
[0085] It should be noted that for other corresponding descriptions of the functional units involved in the ultrasound imaging-based lesion development warning system provided in the embodiment of the present application, please refer to Figures 1 to 2 The corresponding description in the method will not be repeated here.
[0086] Based on the above Figures 1 to 2 The method shown in FIG. 1 is a method for performing the above-mentioned operation. Accordingly, the embodiment of the present application further provides a medium on which a computer program is stored. When the computer program is executed by a processor, the above-mentioned operation is performed. Figures 1 to 2 The ultrasound imaging-based early warning method for lesion development is shown.
[0087] Based on this understanding, the technical solution of the present application can be embodied in the form of a software product, which can be stored in a non-volatile medium (which can be a CD-ROM, USB flash drive, mobile hard disk, etc.), including a number of instructions for enabling a device (which can be a personal computer, server, or network device, etc.) to execute the methods described in each implementation scenario of the present application.
[0088] Based on the above Figures 1 to 2 The method shown, and Figure 3 In order to achieve the above-mentioned purpose, the embodiment of the virtual system shown in the figure further provides a device, which can be a personal computer, a server, a network device, etc. The device includes a medium and a processor; the medium is used to store a computer program; the processor is used to execute the computer program to achieve the above-mentioned Figures 1 to 2 The ultrasound imaging-based early warning method for lesion development is shown.
[0089] Optionally, the device may further include a user interface, a network interface, a camera, a radio frequency (RF) circuit, a sensor, an audio circuit, a Wi-Fi module, etc. The user interface may include a display, an input unit such as a keyboard, etc., and the optional user interface may also include a USB interface, a card reader interface, etc. The network interface may optionally include a standard wired interface, a wireless interface (such as a Bluetooth interface, a Wi-Fi interface), etc.
[0090] Those skilled in the art will understand that the device structure provided in this embodiment does not constitute a limitation on the device, and may include more or fewer components, or a combination of certain components, or different component arrangements.
[0091] The medium may also include an operating system and a network communication module. The operating system is a program that manages and stores the device's hardware and software resources, supporting the execution of information processing programs and other software and / or programs. The network communication module facilitates communication between components within the medium, as well as with other hardware and software within the physical device.
[0092] Through the description of the above implementation methods, those skilled in the art can clearly understand that the present application can be implemented by means of software plus the necessary general hardware platform, or it can be implemented by hardware based on ultrasound images containing the liver and kidneys, extracting first-level ultrasound image features, and determining the liver lesion development rate. The first-level ultrasound image features include the liver echo attenuation coefficient and the liver-kidney ratio; then, extracting second-level ultrasound image features to determine the liver morphological change assessment results. The second-level ultrasound image features include liver morphology, and the liver morphological change assessment results include abnormal and normal; when the determined liver lesion development rate is greater than or equal to the preset lesion development rate warning threshold, and / or the liver morphological change assessment result is abnormal, a lesion development warning signal is triggered. By extracting features based on ultrasound images and quantitatively evaluating the liver lesion development rate and morphological change results, an early warning can be triggered, which is conducive to the early detection and intervention of abnormal liver conditions.
[0093] Those skilled in the art will appreciate that the accompanying drawings are merely schematic diagrams of a preferred implementation scenario, and that the modules or processes in the accompanying drawings are not necessarily required for the implementation of this application. Those skilled in the art will appreciate that the modules in the system in the implementation scenario can be distributed in the system of the implementation scenario according to the implementation scenario description, or can be modified accordingly and located in one or more systems different from the implementation scenario. The modules in the above-mentioned implementation scenario can be combined into one module, or can be further split into multiple submodules.
[0094] The serial numbers of the above application are for descriptive purposes only and do not represent the advantages or disadvantages of the implementation scenarios. The above disclosures are only a few specific implementation scenarios of the present application, but the present application is not limited thereto, and any changes that can be made by those skilled in the art should fall within the scope of protection of the present application.
Claims
1. A method for early warning of lesion development based on ultrasound imaging, characterized in that: The method comprises: Obtain ultrasound images that include the liver and kidneys; Extracting primary ultrasound image features based on the ultrasound image, and determining a liver lesion development rate based on the extracted primary ultrasound image features, wherein the primary ultrasound image features include a liver echo attenuation coefficient and a liver-kidney ratio; Extracting secondary ultrasound image features based on the ultrasound image, and determining a liver morphology change assessment result based on the extracted secondary ultrasound image features, wherein the secondary ultrasound image features include liver morphology, and the liver morphology change assessment result includes abnormal and normal; When the determined liver lesion development rate is greater than or equal to a preset lesion development rate warning threshold, and / or the liver morphological change assessment result is abnormal, a lesion development warning signal is triggered.
2. The method according to claim 1, characterized in that The determining of the liver lesion development rate based on the extracted primary ultrasound image features includes: Obtain the historical medication duration of the patient captured by ultrasound imaging; Liver fat content was determined based on the liver fat content calculation formula, liver echo attenuation coefficient, and liver-kidney ratio; The determined liver fat content is used as the starting liver fat content in the liver index decay process. Combined with the historical medication duration and the liver index decay model, the change in liver fat content over time is determined, and the determined change is used as the liver lesion development rate. The liver index decay model is obtained by correlating the influence of the historical medication duration and the starting liver fat content on the change in liver fat content over time. The liver fat content calculation formula is: Liver fat content = 62.592 × liver-kidney ratio + 168.076 × liver echo attenuation coefficient - 27.863; The liver exponential decay model is: Q(t)=Q0+α(1-e -βt ), Where Q(t) is the liver fat content of the liver that changes with time t, which is used to characterize the liver lesion development rate, Q0 is the initial liver fat content, α is the preset long-term steady-state value, β is the preset metabolic rate constant, e is the Euler number, and e -βt Used to characterize the attenuation degree of liver fat content over time t.
3. The method according to claim 1, characterized in that The extracting of first-level ultrasound image features based on ultrasound images includes: Determining a liver region in an ultrasound image and measurement parameters of a liver echo attenuation coefficient, wherein the measurement parameters include a starting depth and an ending depth, and the starting depth and the ending depth are used to ensure that the ultrasound signal used for detection can cover the liver region where the echo attenuation needs to be measured; Automatically collect echo signals based on the liver region and measurement parameters, and calculate the liver echo attenuation coefficient based on the intensity changes of the echo signals at different depths in the liver region, wherein the echo attenuation coefficient represents the attenuation degree of the ultrasound signal when propagating in the liver tissue; Based on the ultrasound images, the cortical areas of the liver and kidney are determined respectively, and the average gray intensity of the cortical area of the liver is calculated according to the gray value of the pixels in the cortical area of the liver, and the average gray intensity of the cortical area of the kidney is calculated according to the gray value of the pixels in the cortical area of the kidney; The liver-kidney ratio was obtained based on the ratio of the average gray intensity of the liver cortex area to the average gray intensity of the kidney cortex area.
4. The method according to claim 1, wherein The ultrasound image is acquired by an ultrasound apparatus, and corresponds to a fixed standard section selected during acquisition. The fixed standard section includes at least one of a longitudinal section, a transverse section, a subcostal oblique section, and a right intercostal section for the liver and kidneys. Extracting secondary ultrasound image features based on the ultrasound image includes: Identify liver regions on ultrasound images; In the liver region, liver morphological features are extracted, wherein the liver morphological features include at least one of the size, shape, and capsule integrity of the liver.
5. The method according to claim 4, characterized in that The step of determining the liver morphological change assessment result based on the extracted secondary ultrasound image features includes: Calculating liver morphology quantitative indices based on the extracted liver morphological features, wherein the liver morphology quantitative indices include liver volume and / or maximum oblique diameter of the right lobe of the liver, and each liver morphology quantitative indices corresponds to a preset normal reference value range; When any liver morphology quantitative index is outside the corresponding preset normal reference value range, the liver morphology change assessment result is determined to be abnormal; When all the quantitative indices of liver morphology are respectively within their corresponding preset normal reference value ranges, the assessment result of the liver morphology change is determined to be normal.
6. The method according to claim 3, characterized in that The lesion development warning signal includes a primary lesion development warning signal and an advanced lesion development warning signal. The triggering and generating of the lesion development warning signal includes: Obtain liver fat content and liver-kidney ratio calculated based on ultrasound images of the liver and kidneys taken over multiple years for the same patient; When the annual increase in liver fat content exceeds the preset rate, a warning signal for the development of primary lesions is generated; When the liver-kidney ratio is greater than the preset warning ratio for two consecutive times, a warning signal for the development of advanced lesions is generated.
7. The method according to any one of claims 1 to 6, characterized in that The ultrasonic image is acquired through an ultrasonic instrument, which corresponds to a probe pressure sensor and an ultrasonic probe. The probe pressure sensor is integrated into the ultrasonic probe. When the ultrasonic probe is used to acquire the ultrasonic image, the scanning pressure of the ultrasonic probe is monitored based on the probe pressure sensor, and the ultrasonic image initially acquired by the ultrasonic probe is compensated for image deformation based on the monitored scanning pressure, and then the final ultrasonic image is output.
8. An ultrasound imaging-based early warning system for lesion development, characterized in that: The system comprises: An ultrasound image acquisition module, used for acquiring ultrasound images including the liver and kidneys; a primary image feature extraction module, configured to extract primary ultrasound image features based on ultrasound images, and determine the liver lesion development rate based on the extracted primary ultrasound image features, wherein the primary ultrasound image features include the liver echo attenuation coefficient and the liver-kidney ratio; a secondary image feature extraction module, configured to extract secondary ultrasound image features based on the ultrasound image, and determine a liver morphology change assessment result based on the extracted secondary ultrasound image features, wherein the secondary ultrasound image features include liver morphology, and the liver morphology change assessment result includes abnormal and normal; The lesion development warning module is used to trigger the generation of a lesion development warning signal when the determined liver lesion development rate is greater than or equal to a preset lesion development rate warning threshold, and / or the liver morphological change assessment result belongs to a preset warning change degree.
9. A medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method for early warning of lesion development based on ultrasound imaging as described in any one of claims 1 to 7 is implemented.
10. A device comprising a medium, a processor, and a computer program stored on the medium and executable on the processor, wherein: When the processor executes the computer program, the method for early warning of lesion development based on ultrasound imaging as described in any one of claims 1 to 7 is implemented.