Cross-device liver and kidney index follow-up visit method and device

Abstract

The invention discloses a cross-device liver and kidney index follow-up visit method which comprises the following steps: detecting liver and kidney indexes of a patient on first ultrasonic equipment, and generating a detection report according to a detection result; detecting liver and kidney indexes of the patient on second ultrasonic equipment, and during detection, acquiring information in the liver and kidney indexes through the two-dimensional code in the first detection report and adjusting scanning parameters; and obtaining the same parameters and planes on the first ultrasonic equipment, placing the ROI, calculating a liver and kidney index HRI value, and storing the HRI value in a follow-up visit report. According to the follow-up detection method and the follow-up detection device, the detection information and the data of the previous detection are acquired, and the parameters and the images during follow-up detection are adjusted, so that the follow-up detection is easier to acquire the detection parameters and planes consistent with those during the previous detection, and the follow-up detection effect is improved.

Description

Technical Field

[0001] The present application relates to the field of ultrasound technology, and in particular to a method and apparatus for following up liver and kidney indices across multiple devices. Background Art

[0002] The Hepatorenal Index (HRI) is a simple, reliable, and cost-effective screening tool for identifying patients who should not undergo liver biopsy to assess for steatosis. When physicians perform follow-up HRI to diagnose the progression of a patient's fatty liver disease, it is difficult for them to easily scan the same planes and ROI (region of interest) placement as in previous scans for HRI measurements. Consequently, HRI measurements are not consistently consistent, preventing more precise diagnostic results. In particular, HRI values ​​can vary significantly if physicians use different ultrasound devices for follow-up HRI, which have different scanning parameters, ranges, and meanings of parameter values. This variability is not due to the progression of a patient's fatty liver disease, but rather to the use of different devices with different scanning parameters. Consequently, follow-up HRI measurements on different devices are challenging for physicians and prevent them from achieving more precise diagnostic results using traditional methods. Summary of the Invention

[0003] On the one hand, the present application provides a cross-device liver and kidney index follow-up method, which obtains the detection information and data of the previous detection and adjusts the parameters and images performed during the follow-up detection, making it easier for the follow-up detection to obtain detection parameters and planes consistent with the previous detection, thereby improving the effect of the follow-up detection.

[0004] To implement the above solution, this application adopts the following technical solution: a cross-device liver and kidney index follow-up method, comprising the following steps: S1: Detecting the patient's liver and kidney index on a first ultrasound device and generating a test report based on the test results; wherein the test report includes a QR code, and information stored in the QR code includes key scanning parameters of this test, a first image feature vector, a first liver contour and a renal cortex contour, liver and kidney index ROI information, and hyperechoic ROI information in the renal cortex; S2: Detecting the patient's liver and kidney indexes on a second ultrasound device. During the detection, information therein is obtained through the QR code in the first detection report. The same scanning parameters are set on the second ultrasound device according to the obtained scanning parameters and a scan is started. During the scanning process, each frame of the ultrasound image is processed to obtain a second image feature vector and a second liver contour and renal cortex contour. S3: comparing the first image feature vector with the second image feature vector, and comparing the first liver contour and renal cortex contour with the second liver contour and renal cortex contour; When the similarity between the first image feature vector and the second image feature vector, and the first liver contour and the renal cortex contour and the second liver contour and the renal cortex contour reaches a preset value, a prompt is issued to the ultrasound scanning doctor; Otherwise continue scanning; S4: placing a liver and kidney index ROI and a high echo ROI in the renal cortex in the ultrasound image obtained in step S3, and calculating a liver and kidney index HRI value; S5: Generate a follow-up report based on the detection results of the second ultrasound device.

[0005] Furthermore, in step S1, the first image feature vector is represented by no more than 256 elements, and the first liver contour and the renal cortex contour are represented by no more than 300 points.

[0006] Furthermore, in step S1, the liver-kidney index ROI information includes the coordinates, radius, mean and variance value of the liver-kidney index ROI; the hyperechoic ROI information in the renal cortex includes the coordinates, radius, mean and variance value of the hyperechoic ROI in the renal cortex.

[0007] Furthermore, in step S4, the liver-kidney index ROI and the hyperechoic ROI in the renal cortex are placed respectively according to the liver-kidney index ROI information and the hyperechoic ROI information in the renal cortex.

[0008] Furthermore, in step S4, after placing the liver and kidney index ROI and the hyperechoic ROI in the renal cortex, other scanning parameters are adjusted until the image meets expectations.

[0009] Furthermore, in step S1, the key scanning parameters include the scanning field of view and the scanning depth; in step S4, the other scanning parameters adjusted are scanning parameters other than the scanning field of view and the scanning depth, including one or more of frequency, power output, gain, time gain compensation TGC, number of focal spots and focal position.

[0010] Furthermore, in step S4, a real-time scanning parameter search algorithm is used to adjust other scanning parameters. The real-time scanning parameter search algorithm is a grid search algorithm or a genetic algorithm. Image pixels in ROIs with different scanning parameters are input into the real-time scanning parameter search algorithm, and other scanning parameters are adjusted according to the input image.

[0011] Furthermore, the calculation method of the HRI value is as follows: Calculate the mean and variance of the echo intensity of the liver ROI; The mean and variance of the echo intensity of the renal cortex ROI were calculated; The HRI value is equal to the ratio of the mean or variance of the echo intensity of the liver ROI to the mean or variance of the echo intensity of the renal cortex ROI.

[0012] On the other hand, the present application provides a liver and kidney index detection device, comprising a first ultrasound device and a report generating unit, wherein the detection report generating unit receives the detection result of the liver and kidney index ultrasound device and generates a detection report; The test report includes a QR code, which stores key scanning parameters of this test, a first image feature vector, a first liver contour and a renal cortex contour, liver-kidney index ROI information, and high-echo ROI information in the renal cortex.

[0013] Through the above scheme, the liver and kidney index detection device provided by this application can generate a QR code with detection information such as scanning parameters, which can be obtained and used during follow-up.

[0014] On the other hand, the present application provides a liver and kidney index follow-up device, including a second ultrasound device, a QR code scanner and a follow-up report generation unit, wherein the QR code scanner is used to identify the QR code in the test report as described in claim 8 to obtain the information therein, and the second ultrasound device is used to perform liver and kidney index ultrasound detection based on the information obtained by the QR code scanner.

[0015] Through the above scheme, the liver and kidney index follow-up device provided by the present application obtains the detection information and data of the previous detection, adjusts the parameters and images performed during the follow-up detection, and makes it easier for the follow-up detection to obtain detection parameters and planes consistent with the previous detection, thereby improving the effect of the follow-up detection.

[0016] In summary, the beneficial effects of this application are: 1. The cross-device liver and kidney index follow-up method of this application obtains the test information and data of the previous test and adjusts the parameters and images used in the follow-up test, making it easier for the follow-up test to obtain the same test parameters and planes as the previous test, thereby improving the effectiveness of the follow-up test; 2. The cross-device liver and kidney index follow-up method of this application takes the test information and data of the previous test and stores them in a QR code. This allows for easy and accurate follow-up of HRI measurements across the same or different devices without network support. 3. This cross-device liver and kidney index follow-up method uses QR codes to transmit image feature vectors and liver and kidney cortex contours generated by a CNN model. This image information is then used to match the plane scanned on another ultrasound device. This method is highly efficient and less susceptible to network and other factors. 4. The cross-device HRI follow-up method of this application uses two ROIs in the renal cortex to automatically adjust scanning parameters to ensure a consistent HRI baseline. This can maintain consistency between pre- and post-examinations, making HRI measurements more stable and meaningful. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 This is a schematic diagram of the principle of a cross-device liver and kidney index follow-up method of the present application; Figure 2 This is a flowchart of a cross-device liver and kidney index follow-up method of the present application; Figure 3 This is a schematic diagram of the QR code information in a cross-device liver and kidney index follow-up method of the present application; Figure 4 This is a schematic diagram of a follow-up scan in a cross-device liver and kidney index follow-up method of the present application; Figure 5 This is a schematic diagram of scanning parameter adjustment in a cross-device liver and kidney index follow-up method of the present application; Figure 6 This is a schematic diagram of the scanning parameter adjustment principle in a cross-device liver and kidney index follow-up method of the present application; Figure 7 This is a schematic diagram of placing ROIs in a cross-device liver and kidney index follow-up method of the present application. DETAILED DESCRIPTION

[0018] The specific implementation of the present application is described in detail below with reference to the accompanying drawings.

[0019] Example 1: Reference Figure 1-7 , a method for following up liver and kidney indices across devices, comprising the following steps: S1: Detecting the patient's liver and kidney index on a first ultrasound device, and generating a test report based on the test results; wherein the test report includes a QR code, and information stored in the QR code includes key scanning parameters of this test, a first image feature vector, a first liver contour and a renal cortex contour, liver and kidney index ROI (region of interest, the same below) information, and high-echo ROI information in the renal cortex.

[0020] In this embodiment, in the first ultrasound device, the first image feature vector is obtained by a CNN model for feature extraction, and the first liver contour and the renal cortex contour are obtained by a U-Net model for segmentation.

[0021] The first image feature vector is represented by no more than 256 elements, and the first liver contour and the renal cortex contour are represented by no more than 300 points.

[0022] In step S1, the key scanning parameters include the scanning field of view FOV and the scanning depth; the liver and kidney index ROI information includes the coordinates, radius, mean and variance values ​​of the liver and kidney index ROI; the high echo ROI information in the renal cortex includes the coordinates, radius, mean and variance values ​​of the high echo ROI in the renal cortex.

[0023] Among the key scanning parameters, the scanning field of view FOV and scanning depth each occupy 4 bytes. The first image feature vector has a total of 256 elements, each element is 4 bytes. In the first liver contour and renal cortex contour, calculated based on 300 points, each x or y in each point (x, y) occupies 2 bytes. In the two ROIs placed by the doctor (i.e., the liver and kidney index ROI), the coordinate radius, mean, and variance values ​​of the two ROIs occupy 2*(2*2+2+4+4), totaling 28 bytes. In the hyperechoic ROI in the renal cortex, the coordinate radius, mean, and variance values ​​of the ROI occupy 2*2+2+4+4, totaling 14 bytes. In summary, the information that needs to be stored in the QR code totals 2274 bytes, that is, the uncompressed bytes are 2274, and the QR code version 40 can store these bytes. In some other embodiments of the present application, the above information can be encrypted and compressed and stored in the QR code.

[0024] S2: Detecting the patient's liver and kidney indexes on a second ultrasound device. During the detection, information therein is obtained through the QR code in the first detection report. The same scanning parameters are set on the second ultrasound device according to the obtained scanning parameters and a scan is started. During the scanning process, each frame of the ultrasound image is processed to obtain a second image feature vector and a second liver contour and renal cortex contour. In step S2, the same CNN model and U-Net model are used to process the ultrasound image.

[0025] During a follow-up liver and kidney examination, the first ultrasound device is the one used initially. The second ultrasound device is the one used for the follow-up examination. The first and second ultrasound devices can be the same or different types. The second ultrasound device is equipped with a QR code scanner to capture the information in the QR code.

[0026] S3: comparing the first image feature vector with the second image feature vector, specifically, comparing the similarity of the first image feature vector with the second image feature vector by calculating a preselected similarity of the two feature vectors, comparing the first liver contour and the renal cortex contour with the second liver contour and the renal cortex contour, and calculating the contours based on the degree of overlap; When the similarity between the first image feature vector and the second image feature vector, and the first liver contour and renal cortex contour and the second liver contour and renal cortex contour reaches a preset value, a prompt is issued to the ultrasound scanning doctor; specifically, a green progress bar is used on the ultrasound interface to issue a prompt to the ultrasound scanning doctor.

[0027] Otherwise, continue scanning to find the matching plane; S4: placing a liver and kidney index ROI and a high echo ROI in the renal cortex in the ultrasound image obtained in step S3, and calculating a liver and kidney index HRI value; In step S4 , the liver-kidney index ROI and the hyperechoic ROI in the renal cortex are placed respectively according to the liver-kidney index ROI information and the hyperechoic ROI information in the renal cortex.

[0028] After placing the liver-kidney index ROI and the hyperechoic ROI in the renal cortex, other scanning parameters were adjusted until the image met expectations. The adjusted ROI image was compared with the previous liver-kidney index ROI and the hyperechoic ROI in the renal cortex using mean and variance. The image was considered to meet expectations when the mean and variance reached or fell below the preset values.

[0029] Other scan parameters to be adjusted are scan parameters other than the scan field of view and scan depth, and include one or more of frequency, power output, gain, time gain compensation (TGC), number of focal spots, and focal position. Other scan parameters are adjusted using a real-time scan parameter search algorithm, such as a grid search algorithm or a genetic algorithm. Image pixels from ROIs with different scan parameters are input into the real-time scan parameter search algorithm, which then adjusts the other scan parameters based on the input image.

[0030] For HRI, the overall echo intensity of the renal cortex is almost stable and can be used as a benchmark for assessing the progression of fatty liver disease. Therefore, we can select two ROIs at the same location in the renal cortex, one for HRI and the other as a hyperechoic ROI as a reference to adjust scanning parameters such as frequency, power output, gain, TGC, number of focal spots, and focal position.

[0031] S5: Generate a follow-up report based on the detection results of the second ultrasound device.

[0032] The calculation method of the hepatorenal index HRI value is: Calculate the mean and variance of the echo intensity of the liver ROI; The mean and variance of the echo intensity of the renal cortex ROI were calculated; The HRI value is equal to the ratio of the mean or variance of the echo intensity of the liver ROI to the mean or variance of the echo intensity of the renal cortex ROI.

[0033] For other liver diagnostic and measurement tools, such as acoustic attenuation, backscatter distribution, shear wave dispersion, liver texture index, etc., similar solutions can be used to ensure consistent measurement conditions and obtain reliable and accurate follow-up values.

[0034] Embodiment 2: A liver and kidney index detection device includes a first ultrasound device and a report generating unit, wherein the detection report generating unit receives the detection result of the liver and kidney index ultrasound device and generates a detection report; The test report includes a QR code, which stores key scanning parameters of this test, a first image feature vector, a first liver contour and a renal cortex contour, liver-kidney index ROI information, and high-echo ROI information in the renal cortex.

[0035] The liver and kidney index detection device provided in this application can generate a QR code with detection information such as scanning parameters, which can be obtained and used during follow-up.

[0036] Example 3: A liver and kidney index follow-up device, comprising a second ultrasound device, a QR code scanner and a follow-up report generation unit, wherein the QR code scanner is used to identify the QR code in the test report as described in claim 8 to obtain the information therein, and the second ultrasound device is used to perform liver and kidney index ultrasound detection based on the information obtained by the QR code scanner.

[0037] The liver and kidney index follow-up device provided in this application obtains the detection information and data of the previous detection, adjusts the parameters and images performed during the follow-up detection, and makes it easier for the follow-up detection to obtain detection parameters and planes consistent with the previous detection, thereby improving the effect of the follow-up detection.

[0038] The above is only a preferred embodiment of the present application. It should be pointed out that for ordinary technicians in this field, several variations and improvements can be made without departing from the creative concept of the present application, and these all fall within the scope of protection of the present application.

Claims

1. A method for following up liver and kidney index across devices, characterized in that: The following steps are involved: S1: Detecting the patient's liver and kidney index on a first ultrasound device and generating a test report based on the test results; wherein the test report includes a QR code, and information stored in the QR code includes key scanning parameters of this test, a first image feature vector, a first liver contour and a renal cortex contour, liver and kidney index ROI information, and hyperechoic ROI information in the renal cortex; S2: Detecting the patient's liver and kidney indexes on a second ultrasound device. During the detection, information therein is obtained through the QR code in the first detection report. The same scanning parameters are set on the second ultrasound device according to the obtained scanning parameters and a scan is started. During the scanning process, each frame of the ultrasound image is processed to obtain a second image feature vector and a second liver contour and renal cortex contour. S3: comparing the first image feature vector with the second image feature vector, and comparing the first liver contour and renal cortex contour with the second liver contour and renal cortex contour; When the similarity between the first image feature vector and the second image feature vector, and the first liver contour and the renal cortex contour and the second liver contour and the renal cortex contour reaches a preset value, a prompt is issued to the ultrasound scanning doctor; Otherwise continue scanning; S4: placing a liver and kidney index ROI and a high echo ROI in the renal cortex in the ultrasound image obtained in step S3, and calculating a liver and kidney index HRI value; S5: Generate a follow-up report based on the detection results of the second ultrasound device.

2. The cross-device liver and kidney index follow-up method according to claim 1, characterized in that: In step S1 , the first image feature vector is represented by no more than 256 elements, and the first liver contour and the renal cortex contour are represented by no more than 300 points.

3. The cross-device liver and kidney index follow-up method according to claim 1, characterized in that: In step S1 , the liver-kidney index ROI information includes the coordinates, radius, mean and variance of the liver-kidney index ROI; the renal cortex hyperechoic ROI information includes the coordinates, radius, mean and variance of the renal cortex hyperechoic ROI.

4. The cross-device liver and kidney index follow-up method according to claim 3, characterized in that: In step S4 , the liver-kidney index ROI and the hyperechoic ROI in the renal cortex are placed respectively according to the liver-kidney index ROI information and the hyperechoic ROI information in the renal cortex.

5. The cross-device liver and kidney index follow-up method according to claim 4, characterized in that: In step S4 , after placing the liver and kidney index ROI and the hyperechoic ROI in the renal cortex, other scanning parameters are adjusted until the image meets expectations.

6. The cross-device liver and kidney index follow-up method according to claim 5, characterized in that: In step S1, the key scanning parameters include the scanning field of view and the scanning depth; in step S4, the other scanning parameters adjusted are scanning parameters other than the scanning field of view and the scanning depth, including one or more of frequency, power output, gain, time gain compensation TGC, number of focal spots and focal position.

7. The cross-device liver and kidney index follow-up method according to claim 6, characterized in that: In step S4, a real-time scanning parameter search algorithm is used to adjust other scanning parameters. The real-time scanning parameter search algorithm is a grid search algorithm or a genetic algorithm. Image pixels in ROIs with different scanning parameters are input into the real-time scanning parameter search algorithm, and other scanning parameters are adjusted according to the input image.

8. The cross-device liver and kidney index follow-up method according to claim 1, characterized in that: The calculation method of the hepatorenal index HRI value is: Calculate the mean and variance of the echo intensity of the liver ROI; The mean and variance of the echo intensity of the renal cortex ROI were calculated; The HRI value is equal to the ratio of the mean or variance of the echo intensity of the liver ROI to the mean or variance of the echo intensity of the renal cortex ROI.

9. A liver and kidney index detection device, characterized in that: It includes a first ultrasound device and a report generating unit, wherein the detection report generating unit receives the detection result of the liver and kidney index ultrasound device and generates a detection report; The test report includes a QR code, which stores key scanning parameters of this test, a first image feature vector, a first liver contour and a renal cortex contour, liver-kidney index ROI information, and high-echo ROI information in the renal cortex.

10. A liver and kidney index follow-up device, characterized in that: It includes a second ultrasound device, a QR code scanner and a follow-up report generation unit. The QR code scanner is used to identify the QR code in the test report as described in claim 8 to obtain the information therein, and the second ultrasound device is used to perform liver and kidney index ultrasound detection based on the information obtained by the QR code scanner.

Images