Vessel diameter measurement method and apparatus, storage medium

Abstract

The present disclosure provides a blood vessel inner diameter measurement method and device, and a storage medium, and relates to the field of information processing. The blood vessel inner diameter measurement method comprises: performing image recognition on each ultrasound video frame in an ultrasound video to obtain an inner diameter measurement value of a specified blood vessel; generating an inner diameter curve of the inner diameter of the specified blood vessel changing with time by using all the inner diameter measurement values; calculating the saliency of each peak in the inner diameter curve; and taking the inner diameter value corresponding to the wave crest point of the peak with the maximum saliency as the inner diameter value of the specified blood vessel.

Description

Technical Field

[0001] This disclosure relates to the field of information processing, and in particular to a method and apparatus for measuring the inner diameter of blood vessels, and a storage medium. Background Technology

[0002] With the development of artificial intelligence technology, trained machine learning models can be used to identify ultrasound images, enabling the automatic measurement of blood vessel diameters such as the main pulmonary artery. This can effectively reduce the workload of doctors. Summary of the Invention

[0003] The inventors noticed that due to the influence of the cardiac cycle, the inner diameter of blood vessels also undergoes a process of changing from large to small and then from small to large. Therefore, the inner diameter of blood vessels measured at different times is not the same, and thus it is impossible to accurately determine the value of the inner diameter of blood vessels.

[0004] Accordingly, this disclosure provides a method for measuring the inner diameter of blood vessels, which generates an inner diameter curve that changes over time using all the obtained inner diameter measurements, and takes the inner diameter value corresponding to the peak point of the inner diameter curve with the greatest significance as the inner diameter value of the blood vessel, thereby effectively improving the accuracy of the inner diameter value of the blood vessel.

[0005] According to a first aspect of the present disclosure, a method for measuring the inner diameter of a blood vessel is provided, comprising: performing image recognition on each ultrasound video frame in an ultrasound video to obtain a measurement value of the inner diameter of a specified blood vessel; generating an inner diameter curve of the inner diameter of the specified blood vessel changing over time using all inner diameter measurement values; calculating the significance of each peak in the inner diameter curve; and taking the inner diameter value corresponding to the peak point of the peak with the maximum significance as the inner diameter value of the specified blood vessel.

[0006] In some embodiments, calculating the significance of each peak in the inner diameter curve includes: in the positive direction of the time axis, selecting the point with the smallest inner diameter value in a preset neighborhood to the right of the i-th peak in the inner diameter curve as a first candidate point, where 1≤i≤N, and N is the total number of peaks; in the negative direction of the time axis, selecting the point with the smallest inner diameter value in a preset neighborhood to the left of the i-th peak as a second candidate point; selecting the candidate point with the larger inner diameter value between the first and second candidate points as a reference point; generating a reference line parallel to the time axis through the reference point; and using the distance between the peak of the i-th peak and the reference line as the significance of the i-th peak.

[0007] In some embodiments, selecting the point with the smallest inner diameter value within a preset neighborhood to the right of the i-th peak in the inner diameter curve as a first candidate point includes: generating a line segment parallel to the time axis at the peak of the i-th peak; extending the line segment along the positive direction of the time axis, and determining whether the line segment intersects with other peaks in the inner diameter curve before the end time of the inner diameter curve; if the line segment intersects with the j-th peak in the inner diameter curve, then selecting the point with the smallest inner diameter value among all the troughs between the i-th peak and the j-th peak as the first candidate point, where i <j≤N。

[0008] In some embodiments, selecting the point with the smallest inner diameter value within a preset neighborhood to the right of the i-th peak in the inner diameter curve as the first candidate point further includes: if the line segment has not intersected with other peaks in the inner diameter curve by the end time of its extension to the inner diameter curve, determining whether there is a trough point between the i-th peak and the endpoint of the inner diameter curve; if there is a trough point between the i-th peak and the endpoint, then among all the trough points between the i-th peak and the endpoint and the endpoint, the point with the smallest inner diameter value is selected as the first candidate point.

[0009] In some embodiments, using the point with the smallest inner diameter value in a preset neighborhood to the right of the i-th peak in the inner diameter curve as the first candidate point further includes: if there is no trough point between the i-th peak and the endpoint, then the endpoint is used as the first candidate point.

[0010] In some embodiments, selecting the point with the smallest inner diameter value within a preset neighborhood to the left of the i-th peak as a second candidate point includes: generating a line segment parallel to the time axis at the peak of the i-th peak; extending the line segment in the opposite direction of the time axis, and determining whether the line segment intersects with other peaks in the inner diameter curve before the start time of the inner diameter curve; if the line segment intersects with the k-th peak in the inner diameter curve, then selecting the point with the smallest inner diameter value among all the troughs between the i-th peak and the k-th peak as a second candidate point, where 1≤k. <i。

[0011] In some embodiments, selecting the point with the smallest inner diameter value within a preset neighborhood to the left of the i-th peak as the second candidate point further includes: if the line segment has not intersected with other peaks in the inner diameter curve at the start time of its extension to the inner diameter curve, determining whether there is a trough point between the i-th peak and the starting point of the inner diameter curve; if there is a trough point between the i-th peak and the starting point, then selecting the point with the smallest inner diameter value among all the trough points between the i-th peak and the starting point and the starting point as the second candidate point.

[0012] In some embodiments, using the point with the smallest inner diameter value in the preset neighborhood to the left of the i-th peak as the second candidate point further includes: if there is no trough point between the i-th peak and the starting point, then the starting point is used as the second candidate point.

[0013] In some embodiments, image recognition of each ultrasound video frame in an ultrasound video includes: using a trained Keypoint-Region Convolutional Neural Network (RCNN) model to perform image recognition on each ultrasound video frame in the ultrasound video to obtain the internal diameter measurement value of the specified blood vessel.

[0014] In some embodiments, before calculating the significance of each peak in the inner diameter curve, the method further includes: performing at least one of filtering and linear interpolation on the inner diameter curve to update the inner diameter curve.

[0015] According to a second aspect of the present disclosure, a vascular diameter measuring device is provided, comprising: a first processing module configured to perform image recognition on each ultrasound video frame in an ultrasound video to obtain a measured value of the inner diameter of a specified vascular vessel; a second processing module configured to generate an inner diameter curve of the inner diameter of the specified vascular vessel changing over time using all inner diameter measured values; a third processing module configured to calculate the significance of each peak in the inner diameter curve; and a fourth processing module configured to use the inner diameter value corresponding to the peak point of the peak with the maximum significance as the inner diameter value of the specified vascular vessel.

[0016] In some embodiments, the third processing module is configured to, in the positive direction of the time axis, take the point with the smallest inner diameter value in a preset neighborhood to the right of the i-th peak in the inner diameter curve as the first candidate point, where 1≤i≤N, and N is the total number of peaks; in the negative direction of the time axis, take the point with the smallest inner diameter value in a preset neighborhood to the left of the i-th peak as the second candidate point; take the candidate point with the larger inner diameter value among the first and second candidate points as the reference point; generate a reference line parallel to the time axis through the reference point; and take the distance between the peak point of the i-th peak and the reference line as the significance of the i-th peak.

[0017] In some embodiments, the third processing module is configured to generate a line segment parallel to the time axis at the peak point of the i-th peak, extend the line segment along the positive direction of the time axis, and determine whether the line segment intersects with other peaks in the inner diameter curve before the end time of the inner diameter curve. If the line segment intersects with the j-th peak in the inner diameter curve, then among all the trough points between the i-th peak and the j-th peak, the point with the smallest inner diameter value is selected as the first candidate point, where i <j≤N。

[0018] In some embodiments, the third processing module is configured to determine whether there is a trough between the i-th peak and the endpoint of the inner diameter curve if the line segment does not intersect with other peaks in the inner diameter curve at the end time of its extension to the inner diameter curve. If there is a trough between the i-th peak and the endpoint, the point with the smallest inner diameter value is selected as the first candidate point among all the troughs between the i-th peak and the endpoint and the endpoint.

[0019] In some embodiments, the third processing module is configured to use the endpoint as a first candidate point if there is no trough point between the i-th peak and the endpoint.

[0020] In some embodiments, the third processing module is configured to generate a line segment parallel to the time axis at the peak point of the i-th peak, extend the line segment in the opposite direction of the time axis, and determine whether the line segment intersects with other peaks in the inner diameter curve before the start time of the inner diameter curve. If the line segment intersects with the k-th peak in the inner diameter curve, then among all the trough points between the i-th peak and the k-th peak, the point with the smallest inner diameter value is selected as the second candidate point, where 1≤k. <i。

[0021] In some embodiments, the third processing module is configured to determine whether there is a trough between the i-th peak and the starting point of the inner diameter curve if the line segment has not intersected with other peaks in the inner diameter curve at the start time of its extension to the inner diameter curve. If there is a trough between the i-th peak and the starting point, the point with the smallest inner diameter value is selected as the second candidate point among all the troughs between the i-th peak and the starting point and the starting point.

[0022] In some embodiments, the third processing module is configured to use the starting point as a second candidate point if there is no trough point between the i-th peak and the starting point.

[0023] In some embodiments, the first processing module is configured to perform image recognition on each ultrasound video frame in the ultrasound video using a trained Keypoint-Region Convolutional Neural Network (RCNN) model to obtain the internal diameter measurement value of the specified blood vessel.

[0024] In some embodiments, the third processing module is configured to perform at least one of filtering and linear interpolation processing on the inner diameter curve before calculating the significance of each peak in the inner diameter curve, so as to update the inner diameter curve.

[0025] According to a third aspect of the present disclosure, a blood vessel diameter measuring device is provided, comprising: a memory configured to store instructions; and a processor coupled to the memory, the processor being configured to execute, based on the instructions stored in the memory, the method described in any of the above embodiments.

[0026] According to a fourth aspect of the present disclosure, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores computer instructions that, when executed by a processor, implement the method as described in any of the above embodiments.

[0027] Other features and advantages of this disclosure will become clear from the following detailed description of exemplary embodiments with reference to the accompanying drawings. Attached Figure Description

[0028] To more clearly illustrate the technical solutions in the embodiments of this disclosure or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this disclosure. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0029] Figure 1 This is a schematic flowchart of a method for measuring the inner diameter of blood vessels according to an embodiment of the present disclosure;

[0030] Figure 2 This is a schematic diagram of an echocardiogram according to an embodiment of the present disclosure;

[0031] Figure 3 This is a schematic diagram of the inner diameter curve of one embodiment of the present disclosure;

[0032] Figure 4 This is a schematic diagram of the inner diameter curve of another embodiment of the present disclosure;

[0033] Figure 5 This is a schematic diagram of the inner diameter curve of yet another embodiment of the present disclosure;

[0034] Figure 6 This is a schematic diagram of the structure of a blood vessel diameter measuring device according to an embodiment of the present disclosure;

[0035] Figure 7 This is a schematic diagram of the structure of a blood vessel diameter measuring device according to an embodiment of the present disclosure. Detailed Implementation

[0036] The technical solutions of the embodiments of this disclosure will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this disclosure, and not all embodiments. The following description of at least one exemplary embodiment is merely illustrative and is in no way intended to limit this disclosure or its application or use. All other embodiments obtained by those skilled in the art based on the embodiments of this disclosure without creative effort are within the scope of protection of this disclosure.

[0037] Unless otherwise specifically stated, the relative arrangement, numerical expressions, and values ​​of the components and steps set forth in these embodiments do not limit the scope of this disclosure.

[0038] At the same time, it should be understood that, for ease of description, the dimensions of the various parts shown in the accompanying drawings are not drawn according to actual scale.

[0039] Techniques, methods, and equipment known to those skilled in the art may not be discussed in detail, but where appropriate, such techniques, methods, and equipment should be considered part of the specification.

[0040] In all examples shown and discussed herein, any specific values ​​should be interpreted as merely exemplary and not as limitations. Therefore, other examples of exemplary embodiments may have different values.

[0041] It should be noted that similar labels and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be discussed further in subsequent figures.

[0042] Figure 1 This is a schematic flowchart illustrating a method for measuring the inner diameter of a blood vessel according to an embodiment of this disclosure. In some embodiments, the following method for measuring the inner diameter of a blood vessel is performed by a device for measuring the inner diameter of a blood vessel.

[0043] In step 101, image recognition is performed on each ultrasound video frame in the ultrasound video to obtain the internal diameter measurement value of the specified blood vessel.

[0044] For example, the designated blood vessel is the main pulmonary artery.

[0045] In some embodiments, a trained Keypoint-RCNN (Region Convolutional Neural Network) model is used to perform image recognition on each ultrasound video frame in the ultrasound video to obtain the internal diameter measurement value of a specified blood vessel.

[0046] For example, using Keypoint-RCNN to determine the locations of the left and right endpoints of the main pulmonary artery's internal diameter, and calculating the distance between these two locations, the internal diameter of the main pulmonary artery can be obtained. The internal diameter of the main pulmonary artery is shown below. Figure 2 Line 20 is shown in the diagram.

[0047] In step 102, a curve showing the change of the inner diameter of a specified blood vessel over time is generated using all inner diameter measurements.

[0048] For example, the obtained inner diameter curve is as follows Figure 3 As shown.

[0049] In some embodiments, the inner diameter curve is updated by at least one of filtering and linear interpolation before calculating the significance of each peak in the inner diameter curve.

[0050] It should be noted that during the measurement of blood vessel diameter, unclear images or external interference may lead to measurement failures or errors. In such cases, filtering and linear interpolation of the diameter curve can effectively remove outliers, thus obtaining accurate blood vessel diameter measurement results.

[0051] For example, by means of Figure 3 The inner diameter curve shown is filtered and linearly interpolated to obtain the inner diameter curve as follows: Figure 4 As shown.

[0052] In step 103, the significance of each peak in the inner diameter curve is calculated.

[0053] In some embodiments, calculating the significance of each peak in the inner diameter curve includes the following steps:

[0054] 1) In the positive direction of the time axis, the point with the smallest inner diameter value in the preset neighborhood to the right of the i-th peak in the inner diameter curve is taken as the first candidate point, where 1≤i≤N and N is the total number of peaks.

[0055] In some embodiments, a line segment parallel to the time axis is generated at the peak point of the i-th peak. This line segment is extended along the positive direction of the time axis, and it is determined whether it intersects with other peaks in the inner diameter curve before reaching the end time of the inner diameter curve. If the line segment intersects with the j-th peak in the inner diameter curve, then among all the trough points between the i-th and j-th peaks, the point with the smallest inner diameter value is selected as the first candidate point, where i... <j≤N。

[0056] If the line segment does not intersect with any other peak in the inner diameter curve by the time it extends to the end of the inner diameter curve, determine whether there is a trough between the i-th peak and the endpoint of the inner diameter curve. If there is a trough between the i-th peak and the endpoint, then among all the troughs and endpoints between the i-th peak and the endpoint, the point with the smallest inner diameter value is selected as the first candidate point.

[0057] If there is no trough between the i-th peak and the endpoint, then the endpoint is taken as the first candidate point.

[0058] 2) In the opposite direction of the time axis, the point with the smallest inner diameter value in the preset neighborhood to the left of the i-th peak is taken as the second candidate point.

[0059] In some embodiments, a line segment parallel to the time axis is generated at the peak of the i-th peak. This line segment is extended in the opposite direction of the time axis, and it is determined whether the line segment intersects with other peaks in the inner diameter curve before the start time of the inner diameter curve. If the line segment intersects with the k-th peak in the inner diameter curve, then among all the troughs between the i-th and k-th peaks, the point with the smallest inner diameter value is selected as the second candidate point, where 1 ≤ k. <i。

[0060] If the line segment has not intersected with any other peak in the inner diameter curve by the time it extends to the beginning of the inner diameter curve, determine whether there is a trough between the i-th peak and the starting point of the inner diameter curve. If there is a trough between the i-th peak and the starting point, then among all the troughs and starting points between the i-th peak and the starting point, the point with the smallest inner diameter value is selected as the second candidate point.

[0061] If there is no trough between the i-th peak and the starting point, then the starting point is taken as the second candidate point.

[0062] 3) Use the candidate point with the larger inner diameter value among the first and second candidate points as the reference point.

[0063] 4) Generate a reference line parallel to the time axis through the reference point.

[0064] 5) The distance between the peak of the i-th peak and the reference line is taken as the significance of the i-th peak.

[0065] In step 104, the inner diameter value corresponding to the peak point of the peak with the greatest significance is taken as the inner diameter value of the specified blood vessel.

[0066] The embodiments of this disclosure will be described below with specific examples.

[0067] like Figure 5 As shown, the inner diameter curve includes 8 peaks and 7 troughs. The starting point of the inner diameter curve is S, and the ending point is E.

[0068] 1) Regarding peak 1

[0069] A line segment parallel to the time axis is generated at the peak of the first peak. The line segment is extended along the positive direction of the time axis. Since the line segment has not intersected with other peaks in the inner diameter curve by the end time of the extension to the inner diameter curve, and there are troughs a, b, c, d, e, f, and g between the first peak and the endpoint E, the point with the smallest inner diameter value among troughs a, b, c, d, e, f, g and the endpoint E (trough d) is selected as the first candidate point.

[0070] In addition, a line segment parallel to the time axis is generated at the peak of the first peak. The line segment is extended in the opposite direction of the time axis. Since the line segment has not intersected with other peaks in the inner diameter curve at the beginning time of its extension to the inner diameter curve, and there is no trough between the first peak and the starting point S, the starting point S is selected as the second candidate point.

[0071] from Figure 5 As can be seen, since the inner diameter of the second candidate point (starting point S) is greater than that of the first candidate point (trough d), a reference line L1 parallel to the time axis is generated through the starting point S, and the distance between the peak of the first peak and the reference line L1 is used as the significance of the first peak.

[0072] 2) Regarding peak 2

[0073] A line segment parallel to the time axis is generated at the peak of the second peak. The line segment is extended along the positive direction of the time axis. Since the line segment intersects the third peak in the inner diameter curve, and there is only one trough b between the second and third peaks, trough b is selected as the first candidate point.

[0074] In addition, a line segment parallel to the time axis is generated at the peak point of the second peak. The line segment is extended in the opposite direction of the time axis. Since the line segment intersects the first peak in the inner diameter curve, and there is only one trough a between the second peak and the first peak, trough a is taken as the second candidate point.

[0075] from Figure 5 As can be seen, since the inner diameter of the first candidate point (valley b) is greater than that of the second candidate point (valley a), a reference line L2 parallel to the time axis is generated through valley b, and the distance between the peak of the second peak and the reference line L2 is used as the significance of the second peak.

[0076] 3) Regarding peak 3

[0077] A line segment parallel to the time axis is generated at the peak of the third peak. The line segment is extended along the positive direction of the time axis. Since the line segment intersects the sixth peak in the inner diameter curve, and the third and sixth peaks include troughs c, d, and e, the trough d with the smallest inner diameter value among troughs c, d, and e is selected as the first candidate point.

[0078] In addition, a line segment parallel to the time axis is generated at the peak of the third peak. The line segment is extended in the opposite direction of the time axis. Since the line segment intersects the first peak in the inner diameter curve, and there are troughs a and b between the third peak and the first peak, the trough a with the smallest inner diameter value among troughs a and b is selected as the second candidate point.

[0079] from Figure 5 As can be seen, since the inner diameter of the second candidate point (valley a) is greater than that of the first candidate point (valley d), a reference line L3 parallel to the time axis is generated through valley a, and the distance between the peak of the third peak and the reference line L3 is used as the significance of the third peak.

[0080] 4) Regarding peak 4

[0081] A line segment parallel to the time axis is generated at the peak of the 4th peak. The line segment is extended along the positive direction of the time axis. Since the line segment intersects the 5th peak in the inner diameter curve, and there is only one trough d between the 4th and 5th peaks, trough d is selected as the first candidate point.

[0082] In addition, a line segment parallel to the time axis is generated at the peak of the fourth peak. The line segment is extended in the opposite direction of the time axis. Since the line segment intersects the third peak in the inner diameter curve, and there is only one trough c between the fourth peak and the third peak, trough c is selected as the second candidate point.

[0083] from Figure 5 As can be seen, since the inner diameter of the second candidate point (trough c) is greater than that of the first candidate point (trough d), a reference line L4 parallel to the time axis is generated through trough c, and the distance between the peak of the fourth peak and the reference line L4 is used as the significance of the fourth peak.

[0084] 5) Regarding peak 5

[0085] A line segment parallel to the time axis is generated at the peak of the 5th peak. Extending the line segment along the positive direction of the time axis, since the line segment intersects the 6th peak in the inner diameter curve, and there is only one trough e between the 5th and 6th peaks, trough e is selected as the first candidate point.

[0086] In addition, a line segment parallel to the time axis is generated at the peak of the 5th peak. The line segment is extended in the opposite direction of the time axis. Since the line segment intersects the 3rd peak in the inner diameter curve, and the 5th peak and the 3rd peak include troughs c and d, the trough d with the smallest inner diameter value among troughs c and d is taken as the second candidate point.

[0087] from Figure 5As can be seen, since the inner diameter of the first candidate point (valley e) is greater than that of the second candidate point (valley d), a reference line L5 parallel to the time axis is generated through valley e, and the distance between the peak of the fifth peak and the reference line L5 is used as the significance of the fifth peak.

[0088] 6) Regarding peak 6

[0089] A line segment parallel to the time axis is generated at the peak of the 6th peak. The line segment is extended along the positive direction of the time axis. Since the line segment has not intersected with other peaks in the inner diameter curve by the end time of the extension, and there are troughs f and g between the 6th peak and the endpoint E, the trough f with the smallest inner diameter value among troughs f and g is selected as the first candidate point.

[0090] Furthermore, a line segment parallel to the time axis is generated at the peak of the 6th peak. Extending the line segment in the opposite direction of the time axis, since the line segment intersects the 1st peak in the inner diameter curve, and the 6th peak and the 1st peak include troughs a, b, c, d and e, the trough d with the smallest inner diameter value among troughs a, b, c, d and e is selected as the second candidate point.

[0091] from Figure 5 As can be seen, since the inner diameter of the first candidate point (valley f) is greater than that of the second candidate point (valley d), a reference line L6 parallel to the time axis is generated through valley f, and the distance between the peak of the 6th peak and the reference line L6 is used as the significance of the 6th peak.

[0092] 7) Regarding peak 7

[0093] A line segment parallel to the time axis is generated at the peak of the 7th peak. The line segment is extended along the positive direction of the time axis. Since the line segment has not intersected with other peaks in the inner diameter curve at the end time of the extension, and there is a trough g between the 7th peak and the endpoint E, the point with the smallest inner diameter value (end point E) between the trough g and the endpoint E is selected as the first candidate point.

[0094] In addition, a line segment parallel to the time axis is generated at the peak of the 7th peak. The line segment is extended in the opposite direction of the time axis. Since the line segment intersects the 6th peak in the inner diameter curve, and the 7th peak and the 6th peak include a trough f, the trough f is used as the second candidate point.

[0095] from Figure 5 As can be seen, since the inner diameter of the first candidate point (end point E) is greater than that of the second candidate point (trough f), a reference line L7 parallel to the time axis is generated through the end point E, and the distance between the peak of the 7th peak and the reference line L7 is used as the significance of the 7th peak.

[0096] 8) Regarding peak 8

[0097] A line segment parallel to the time axis is generated at the peak of the 8th peak. The line segment is extended along the positive direction of the time axis. Since the line segment has not intersected with other peaks in the inner diameter curve at the end time of the extension, and there is no trough between the 8th peak and the endpoint E, the endpoint E is selected as the first candidate point.

[0098] In addition, a line segment parallel to the time axis is generated at the peak of the 8th peak. The line segment is extended in the opposite direction of the time axis. Since the line segment intersects the 7th peak in the inner diameter curve, and the 8th peak and the 7th peak include a trough g, the trough g is selected as the second candidate point.

[0099] from Figure 5 As can be seen, since the inner diameter of the second candidate point (trough g) is greater than that of the first candidate point (end point E), a reference line L8 parallel to the time axis is generated through the trough g, and the distance between the peak of the 8th peak and the reference line L8 is used as the significance of the 8th peak.

[0100] from Figure 6 As can be seen from the data, among the 1st to the 8th peaks, the distance H between the peak of the 6th peak and the reference line L6 is the largest. In this case, the inner diameter value corresponding to the peak of the 6th peak is taken as the inner diameter value of the specified blood vessel.

[0101] It should be noted that although the inner diameter value corresponding to the peak of the first peak is greater than that corresponding to the peak of the sixth peak, the inner diameter value corresponding to the peak of the first peak is not used as the inner diameter value of the specified blood vessel because the significance of the first peak is relatively small. This effectively improves the accuracy of blood vessel inner diameter measurement.

[0102] Figure 6 This is a schematic diagram of the structure of a blood vessel diameter measuring device according to an embodiment of this disclosure. Figure 7 As shown, the blood vessel diameter measuring device includes a first processing module 61, a second processing module 62, a third processing module 63, and a fourth processing module 64.

[0103] The first processing module 61 is configured to perform image recognition on each ultrasound video frame in the ultrasound video to obtain the internal diameter measurement value of the specified blood vessel.

[0104] For example, the designated blood vessel is the main pulmonary artery.

[0105] In some embodiments, the first processing module 61 uses a trained Keypoint-RCNN model to perform image recognition on each ultrasound video frame in the ultrasound video to obtain the inner diameter measurement value of the specified blood vessel.

[0106] The second processing module 62 is configured to generate a curve showing the change of the inner diameter of a specified blood vessel over time using all inner diameter measurements.

[0107] The third processing module 63 is configured to calculate the significance of each peak in the inner diameter curve.

[0108] In some embodiments, the third processing module 63 updates the inner diameter curve by at least one of filtering and linear interpolation before calculating the significance of each peak in the inner diameter curve.

[0109] In some embodiments, the third processing module 63, in the positive direction of the time axis, selects the point with the smallest inner diameter value in a preset neighborhood to the right of the i-th peak in the inner diameter curve as the first candidate point, where 1≤i≤N and N is the total number of peaks; in the negative direction of the time axis, selects the point with the smallest inner diameter value in a preset neighborhood to the left of the i-th peak as the second candidate point; selects the candidate point with the larger inner diameter value among the first and second candidate points as the reference point; generates a reference line parallel to the time axis through the reference point; and uses the distance between the peak of the i-th peak and the reference line as the significance of the i-th peak.

[0110] In some embodiments, the third processing module 63 generates a line segment parallel to the time axis at the peak point of the i-th peak, extends the line segment along the positive direction of the time axis, and determines whether the line segment intersects with other peaks in the inner diameter curve before the end time of the inner diameter curve. If the line segment intersects with the j-th peak in the inner diameter curve, then among all the trough points between the i-th peak and the j-th peak, the point with the smallest inner diameter value is selected as the first candidate point, where i <j≤N。

[0111] In some embodiments, if the line segment has not intersected with other peaks in the inner diameter curve by the end time of its extension to the inner diameter curve, the third processing module 63 determines whether there is a trough between the i-th peak and the end point of the inner diameter curve. If there is a trough between the i-th peak and the end point, the point with the smallest inner diameter value is selected as the first candidate point among all the troughs and the end point between the i-th peak and the end point.

[0112] In some embodiments, if there is no trough between the i-th peak and the endpoint, the third processing module 63 will use the endpoint as the first candidate point.

[0113] In some embodiments, the third processing module 63 generates a line segment parallel to the time axis at the peak point of the i-th peak, extends the line segment in the opposite direction of the time axis, and determines whether the line segment intersects with other peaks in the inner diameter curve before the start time of the inner diameter curve. If the line segment intersects with the k-th peak in the inner diameter curve, then among all the trough points between the i-th peak and the k-th peak, the point with the smallest inner diameter value is selected as the second candidate point, where 1≤k. <i。

[0114] In some embodiments, if the line segment has not intersected with other peaks in the inner diameter curve at the start time of its extension to the inner diameter curve, the third processing module 63 determines whether there is a trough between the i-th peak and the starting point of the inner diameter curve. If there is a trough between the i-th peak and the starting point, the point with the smallest inner diameter value is selected as the second candidate point among all the troughs and starting points between the i-th peak and the starting point.

[0115] In some embodiments, if there is no trough point between the i-th peak and the starting point, the third processing module 63 will use the starting point as the second candidate point.

[0116] The fourth processing module 64 is configured to use the inner diameter value corresponding to the peak point of the peak with the greatest significance as the inner diameter value of the specified blood vessel.

[0117] Figure 7 This is a schematic diagram of the structure of a blood vessel diameter measuring device according to another embodiment of this disclosure. Figure 1 As shown, the blood vessel diameter measuring device includes a memory 71 and a processor 72.

[0118] Memory 71 is used to store instructions, and processor 72 is coupled to memory 71. Processor 72 is configured to execute instructions based on memory storage, as shown in the example below. Figure 7 The method involved in any of the embodiments.

[0119] like Figure 1 As shown, the vascular diameter measuring device also includes a communication interface 73 for exchanging information with other devices. Additionally, the device includes a bus 74, through which the processor 72, communication interface 73, and memory 71 communicate with each other.

[0120] The memory 71 may include high-speed RAM, and may also include non-volatile memory, such as at least one disk drive. The memory 71 may also be a memory array. The memory 71 may also be divided into blocks, and these blocks may be combined into virtual volumes according to certain rules.

[0121] Furthermore, processor 72 may be a central processing unit (CPU), an application-specific integrated circuit (ASIC), or one or more integrated circuits configured to implement embodiments of the present disclosure.

[0122] This disclosure also relates to a computer-readable storage medium storing computer instructions that, when executed by a processor, implement... ​ The method involved in any of the embodiments.

[0123] In some embodiments, the functional units described above may be implemented as general-purpose processors, programmable logic controllers (PLCs), digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, or any suitable combination thereof for performing the functions described herein.

[0124] Those skilled in the art will understand that all or part of the steps of the above embodiments can be implemented by hardware or by a program instructing related hardware. The program can be stored in a computer-readable storage medium, such as a read-only memory, a disk, or an optical disk.

[0125] The description in this disclosure is provided for illustrative and descriptive purposes only and is not intended to be exhaustive or to limit the disclosure to its forms. Many modifications and variations will be apparent to those skilled in the art. The embodiments were chosen and described in order to better illustrate the principles and practical application of this disclosure and to enable those skilled in the art to understand this disclosure and to design various embodiments with various modifications suitable for a particular purpose.

Claims

1. A method for measuring the inner diameter of a blood vessel, comprising: Image recognition is performed on each ultrasound video frame to obtain the internal diameter measurement value of the specified blood vessel. Generate an inner diameter curve of the specified blood vessel over time using all inner diameter measurements; Calculate the significance of each peak in the inner diameter curve; The inner diameter value corresponding to the peak point of the peak with the greatest significance is taken as the inner diameter value of the specified blood vessel. The calculation of the significance of each peak in the inner diameter curve includes: In the positive direction of the time axis, the point with the smallest inner diameter value within a preset neighborhood to the right of the i-th peak in the inner diameter curve is selected as the first candidate point, where... N is the total number of peaks; In the opposite direction of the time axis, the point with the smallest inner diameter value in the preset neighborhood to the left of the i-th peak is taken as the second candidate point; The candidate point with the larger inner diameter value among the first and second candidate points is used as the reference point. A reference line parallel to the time axis is generated through the reference point; The distance between the peak of the i-th peak and the reference line is taken as the significance of the i-th peak.

2. The method according to claim 1, wherein, The point with the smallest inner diameter value within a preset neighborhood to the right of the i-th peak in the inner diameter curve is selected as the first candidate point, including: A line segment parallel to the time axis is generated at the peak point of the i-th peak; Extend the line segment along the positive direction of the time axis and determine whether the line segment intersects with other peaks in the inner diameter curve before the end time of the inner diameter curve. If the line segment intersects with the j-th peak in the inner diameter curve, then among all the troughs between the i-th and j-th peaks, the point with the smallest inner diameter value is selected as the first candidate point. .

3. The method according to claim 2, wherein, The point with the smallest inner diameter value within a preset neighborhood to the right of the i-th peak in the inner diameter curve is also considered as the first candidate point. If the line segment does not intersect with other peaks in the inner diameter curve by the time it extends to the end of the inner diameter curve, determine whether there is a trough between the i-th peak and the end point of the inner diameter curve. If there is a trough between the i-th peak and the endpoint, then among all the troughs between the i-th peak and the endpoint and the endpoint, the point with the smallest inner diameter value will be selected as the first candidate point.

4. The method according to claim 3, wherein, The point with the smallest inner diameter value within a preset neighborhood to the right of the i-th peak in the inner diameter curve is also considered as the first candidate point. If there is no trough between the i-th peak and the endpoint, then the endpoint is taken as the first candidate point.

5. The method according to claim 1, wherein, The points with the smallest inner diameter value within a preset neighborhood to the left of the i-th peak are included as the second candidate points: A line segment parallel to the time axis is generated at the peak point of the i-th peak. Extend the line segment in the opposite direction of the time axis and determine whether the line segment intersects with other peaks in the inner diameter curve before the start time of the inner diameter curve. If the line segment intersects the k-th peak in the inner diameter curve, then among all the troughs between the i-th and k-th peaks, the point with the smallest inner diameter value is selected as the second candidate point. .

6. The method according to claim 5, wherein, The point with the smallest inner diameter value within the preset neighborhood to the left of the i-th peak is also considered as a second candidate point: If the line segment does not intersect with other peaks in the inner diameter curve at the beginning time of its extension to the inner diameter curve, determine whether there is a trough point between the i-th peak and the starting point of the inner diameter curve. If there is a trough between the i-th peak and the starting point, then among all the troughs between the i-th peak and the starting point and the starting point, the point with the smallest inner diameter value will be selected as the second candidate point.

7. The method according to claim 6, wherein, The point with the smallest inner diameter value within the preset neighborhood to the left of the i-th peak is also considered as a second candidate point: If there is no trough between the i-th peak and the starting point, then the starting point is taken as the second candidate point.

8. The method according to claim 1, wherein, Image recognition for each ultrasound video frame includes: Using a trained Keypoint-Region Convolutional Neural Network (RCNN) model, image recognition is performed on each ultrasound video frame to obtain the internal diameter measurement value of the specified blood vessel.

9. The method according to any one of claims 1-8, wherein, Before calculating the significance of each peak in the inner diameter curve, the following steps are also included: The inner diameter curve is updated by performing at least one of filtering and linear interpolation.

10. A device for measuring the inner diameter of a blood vessel, comprising: The first processing module is configured to perform image recognition on each ultrasound video frame in the ultrasound video to obtain the internal diameter measurement value of the specified blood vessel. The second processing module is configured to generate an inner diameter curve of the specified blood vessel inner diameter over time using all inner diameter measurements. The third processing module is configured to calculate the significance of each peak in the inner diameter curve, and in the positive direction of the time axis, select the point with the smallest inner diameter value within a preset neighborhood to the right of the i-th peak in the inner diameter curve as the first candidate point. N is the total number of peaks. In the opposite direction of the time axis, the point with the smallest inner diameter value in the preset neighborhood to the left of the i-th peak is taken as the second candidate point. The candidate point with the larger inner diameter value among the first and second candidate points is taken as the reference point. A reference line parallel to the time axis is generated through the reference point. The distance between the peak point of the i-th peak and the reference line is taken as the significance of the i-th peak. The fourth processing module is configured to use the inner diameter value corresponding to the peak point of the peak with the greatest significance as the inner diameter value of the specified blood vessel.

11. The apparatus according to claim 10, wherein, The third processing module is configured to generate a line segment parallel to the time axis at the peak point of the i-th peak, extend the line segment along the positive direction of the time axis, and determine whether the line segment intersects with other peaks in the inner diameter curve before the end time of the inner diameter curve. If the line segment intersects with the j-th peak in the inner diameter curve, then among all the trough points between the i-th and j-th peaks, the point with the smallest inner diameter value is selected as the first candidate point. .

12. The apparatus according to claim 11, wherein, The third processing module is configured to determine whether there is a trough between the i-th peak and the endpoint of the inner diameter curve if the line segment does not intersect with other peaks in the inner diameter curve at the end time of its extension to the inner diameter curve. If there is a trough between the i-th peak and the endpoint, the point with the smallest inner diameter value is selected as the first candidate point among all the troughs between the i-th peak and the endpoint and the endpoint.

13. The apparatus according to claim 12, wherein, The third processing module is configured to use the endpoint as the first candidate point if there is no trough point between the i-th peak and the endpoint.

14. The apparatus according to claim 10, wherein, The third processing module is configured to generate a line segment parallel to the time axis at the peak point of the i-th peak, extend the line segment in the opposite direction of the time axis, and determine whether the line segment intersects with other peaks in the inner diameter curve before the start time of the inner diameter curve. If the line segment intersects with the k-th peak in the inner diameter curve, then among all the trough points between the i-th peak and the k-th peak, the point with the smallest inner diameter value is selected as the second candidate point. .

15. The apparatus according to claim 14, wherein, The third processing module is configured to determine whether there is a trough between the i-th peak and the starting point of the inner diameter curve if the line segment has not intersected with other peaks in the inner diameter curve at the start time of its extension to the inner diameter curve. If there is a trough between the i-th peak and the starting point, the point with the smallest inner diameter value is selected as the second candidate point among all the troughs between the i-th peak and the starting point and the starting point.

16. The apparatus according to claim 15, wherein, The third processing module is configured to use the starting point as the second candidate point if there is no trough point between the i-th peak and the starting point.

17. The apparatus according to claim 10, wherein, The first processing module is configured to use a trained Keypoint-Region Convolutional Neural Network (RCNN) model to perform image recognition on each ultrasound video frame in the ultrasound video to obtain the internal diameter measurement value of the specified blood vessel.

18. The apparatus according to any one of claims 10-17, wherein, The third processing module is configured to perform at least one of filtering and linear interpolation processing on the inner diameter curve before calculating the significance of each peak in the inner diameter curve, so as to update the inner diameter curve.

19. A device for measuring the inner diameter of a blood vessel, comprising: The memory is configured to store instructions; A processor, coupled to a memory, configured to implement the method as described in any one of claims 1-9 based on memory-stored instruction execution.

20. A non-transient computer-readable storage medium, wherein, A computer-readable storage medium stores computer instructions that, when executed by a processor, implement the method as described in any one of claims 1-9.

Images