CN116777927A

CN116777927A - Pedicle detection method and device and computer equipment

Info

Publication number: CN116777927A
Application number: CN202210233149.9A
Authority: CN
Inventors: 付春萌; 杨云洪
Original assignee: Wuhan United Imaging Zhirong Medical Technology Co Ltd
Current assignee: Wuhan United Imaging Zhirong Medical Technology Co Ltd
Priority date: 2022-03-09
Filing date: 2022-03-09
Publication date: 2023-09-19

Abstract

The application relates to a pedicle detection method, a pedicle detection device, computer equipment and a readable storage medium. The method comprises the following steps: determining the center point coordinates of the center points of the pedicle areas in the pedicle segmented image according to the pedicle segmented image of the target spine, performing curve fitting treatment on the center point coordinates of the center points of the pedicle areas to obtain a pedicle center curve equation, acquiring the center point coordinates of the missed detection pedicle according to the pedicle center curve equation and the center point coordinates of the center points of the pedicle areas, and determining the pedicle detection result corresponding to the target spine according to the center point coordinates of the center points of the pedicle areas and the center point coordinates of the missed detection pedicle. The method can be realized by a set of computer program, and can target the pedicle detection result corresponding to the spine without manual participation, thereby reducing the detection error and improving the accuracy of the pedicle detection result.

Description

Pedicle detection method and device and computer equipment

Technical Field

The application relates to the technical field of medical image processing, in particular to a pedicle detection method, a pedicle detection device and computer equipment.

Background

As the spine rotates in cross-section, the pedicle center point of the spine shifts relative to the body center point. Thus, to assess the extent of vertebral body rotation, the pedicle center may be examined to assess the extent of vertebral body rotation by the amount of offset that occurs from the pedicle center relative to the vertebral body center. In the related art, a method based on a snake model or a method based on machine learning can be used for detecting the region where the pedicle is located by combining a manual interaction mode, and the pedicle center point is determined.

However, the detection method in the related art may cause inaccuracy in the detected pedicle center point.

Disclosure of Invention

Based on the above, it is necessary to provide a pedicle detection method, a pedicle detection device and a computer device.

A pedicle detection method, the method comprising:

according to the pedicle segmented image of the target spine, determining the center point coordinates of the center points of the pedicle areas in the pedicle segmented image;

performing curve fitting treatment on the central point coordinates of the central points of the pedicle areas to obtain a pedicle central curve equation;

acquiring the center point coordinates of the center point of the missed pedicle according to the pedicle center curve equation and the center point coordinates of the center point of each pedicle region;

And determining the pedicle detection result corresponding to the target spine according to the center point coordinates of the center points of the pedicle areas and the center point coordinates of the center points of the missed detection pedicle.

In one embodiment, determining center point coordinates of center points of respective pedicle regions in a pedicle segmented image includes:

acquiring the total number of pixel points in each pedicle region and the coordinates of each pixel point;

and obtaining the center point coordinates of the center points of the pedicle areas according to the total number of the pixel points in the pedicle areas and the coordinates of the pixel points.

In one embodiment, obtaining the center point coordinates of the missed pedicle center point according to the pedicle center curve equation and the center point coordinates of the center points of the pedicle regions includes:

according to a first preset direction, determining the central point coordinates of the central point of the first missed detection pedicle according to the pedicle central curve equation and the central point coordinates of the central points of the pedicle areas;

according to a second preset direction, determining the center point coordinates of the center point of the second missed detection pedicle according to the pedicle center curve equation, the center point coordinates of the center points of the pedicle areas and the center point coordinates of the center point of the first missed detection pedicle; any one of the first preset direction and the second preset direction is the direction from the head end to the tail end of the target spine, and the other preset direction is the direction from the tail end to the head end of the target spine;

And determining the center point coordinates of the center point of the missed-detection pedicle based on the center point coordinates of the center point of the first missed-detection pedicle and the center point coordinates of the center point of the second missed-detection pedicle.

In one embodiment, according to a first preset direction, according to a pedicle central curve equation and central point coordinates of central points of the pedicle regions, determining the central point coordinates of the central point of the first missed pedicle includes:

acquiring a plurality of first center point combinations according to the first preset direction and the position relation of each pedicle region; each first center point combination comprises three adjacent center points;

acquiring first distance evaluation values of each first center point combination according to the center point coordinates of three adjacent center points in each first center point combination; the first distance evaluation value represents the deviation degree of the distance between every two adjacent center points in each first center point combination;

if the first distance evaluation value is larger than a first preset threshold value, determining that a first missed detection pedicle central point exists in the current first central point combination, and determining the central point coordinate of the first missed detection pedicle central point according to the central point coordinate of the pedicle region central point and the pedicle central curve equation in the current first central point combination.

In one embodiment, according to a second preset direction, determining the center point coordinates of the center point of the second missed-detection pedicle according to the pedicle center curve equation, the center point coordinates of the center points of the respective pedicle regions, and the center point coordinates of the center point of the first missed-detection pedicle, includes:

acquiring a plurality of second center point combinations according to the second preset direction, the position relation of each pedicle region and the position relation of the center point of the first missed detection pedicle; each second center point combination comprises three adjacent center points in all pedicle region center points and all first missed detection pedicle center points;

acquiring second distance evaluation values of each second center point combination according to the center point coordinates of three adjacent center points in each second center point combination; the second distance evaluation value represents the deviation degree of the distance between every two adjacent center points in each second center point combination;

if the second distance evaluation value is larger than a second preset threshold value, determining that a second missed detection pedicle central point exists in the current second central point combination, and determining the central point coordinates of the second missed detection pedicle central point according to the central point coordinates of the pedicle region central point and the pedicle central curve equation in the current second central point combination.

In one embodiment, obtaining the first distance evaluation value of each first center point combination according to the center point coordinates of three adjacent center points in each first center point combination includes:

for each first center point combination, acquiring a first distance between the first two adjacent center points in the three adjacent center points and a second distance between the last two adjacent center points in the three adjacent center points according to coordinates of the three adjacent center points in the first center point combination;

a ratio between the first distance and the second distance is determined as a first distance evaluation value of the first center point combination.

In one embodiment, if the first distance evaluation value is greater than a first preset threshold, determining that there is a missing pedicle center point in the current first center point combination includes:

if the first distance evaluation value is larger than a first preset threshold value, determining that a missing pedicle center point exists between the first two adjacent center points in the three adjacent center points of the current first center point combination.

In one embodiment, determining the center point coordinates of the first missed approach pedicle center point according to the center point coordinates of the pedicle region center point and the pedicle center curve equation in the current first center point combination includes:

Determining an average value of first coordinates of the first two adjacent central points in the current first central point combination as the first coordinates of the first missed detection pedicle central point;

substituting the average value of the first coordinates of the first two adjacent central points in the current first central point combination into a pedicle central curve equation to obtain a second coordinate of the first missed-detection pedicle central point, and determining the central point coordinate of the first missed-detection pedicle central point through the first coordinate and the second coordinate of the first missed-detection pedicle central point;

one of the first coordinate and the second coordinate is an abscissa, and the other coordinate is an ordinate.

In one embodiment, the method further comprises:

acquiring a medical image of a target spine;

and segmenting the pedicles in the medical image of the target spine through the pedicle segmentation model to obtain pedicle segmentation images.

A pedicle detection device, the device comprising:

the first coordinate determining module is used for determining the center point coordinates of the center points of the pedicle areas in the pedicle segmented image according to the pedicle segmented image of the target spine;

the fitting processing module is used for performing curve fitting processing on the central point coordinates of the central points of the pedicle areas to obtain a pedicle central curve equation;

The second coordinate determining module is used for obtaining the center point coordinates of the center points of the missed detection pedicles according to the pedicle center curve equation and the center point coordinates of the center points of the pedicle areas;

the detection result determining module is used for determining the pedicle detection result corresponding to the target spine according to the center point coordinates of the center points of the pedicle areas and the center point coordinates of the center points of the missed pedicles.

A computer device comprising a memory storing a computer program and a processor which when executing the computer program performs the steps of:

The pedicle detection method, the pedicle detection device and the computer equipment, wherein the pedicle detection method comprises the following steps: determining the center point coordinates of the center points of the pedicle areas in the pedicle segmented image according to the pedicle segmented image of the target spine, performing curve fitting treatment on the center point coordinates of the center points of the pedicle areas to obtain a pedicle center curve equation, acquiring the center point coordinates of the center points of the missed detection pedicle according to the pedicle center curve equation and the center point coordinates of the center points of the pedicle areas, and determining the pedicle detection result corresponding to the target spine according to the center point coordinates of the center points of the pedicle areas and the center point coordinates of the center points of the missed detection pedicle; the method can be realized by a set of computer program, and can target the pedicle detection result corresponding to the spine without manual participation, thereby reducing the detection error and improving the accuracy of the pedicle detection result.

Drawings

FIG. 1 is a diagram of an application environment for a pedicle detection method in one embodiment;

FIG. 2 is a flow chart of a pedicle detection method in one embodiment;

FIG. 3 is a flow chart of the step of determining center point coordinates of center points of pedicle regions in a pedicle segmented image, in one embodiment;

FIG. 4 is a flow chart of a method for acquiring coordinates of a center point of a missed approach pedicle in another embodiment;

FIG. 5 is a schematic view of a pedicle segmented image according to another embodiment;

FIG. 6 is a schematic view of a pedicle segmented image corresponding to the embodiment of FIG. 5;

FIG. 7 is a flowchart of another exemplary method for obtaining coordinates of a center point of a first missed approach pedicle;

FIG. 8 is a flowchart of another exemplary method for obtaining coordinates of a center point of a second missing pedicle in accordance with one embodiment;

FIG. 9 is a flowchart of a method for obtaining first distance estimation values of each first center point combination according to another embodiment;

FIG. 10 is a flow chart of another embodiment of a method for determining coordinates of a first missed pedicle central point;

FIG. 11 is a schematic view of a pedicle segmented image corresponding to the embodiment of FIG. 5;

FIG. 12 is a schematic view of the central point of the pedicle region and the central points of all missed pedicles corresponding to the embodiment of FIG. 11;

FIG. 13 is a flow chart of a method of acquiring pedicle segmented images in another embodiment;

fig. 14 is a schematic structural diagram of a segmentation model based on UNet in another embodiment;

FIG. 15 is a block diagram of a pedicle detection device in one embodiment;

fig. 16 is an internal structural view of a computer device in one embodiment.

Detailed Description

The present application will be described in further detail with reference to the drawings and examples, in order to make the objects, technical solutions and advantages of the present application more apparent. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the application.

The pedicle detection method provided by the application can be applied to the pedicle detection system shown in fig. 1, and can be applied to a scene for evaluating the degree of vertebral body rotation of an imaging object. The pedicle detection system comprises a medical scanning device and a computer device. The communication connection between the computer device and the medical scanning device may be Wi-Fi, a mobile network or a bluetooth connection, etc. The medical scanning device can be a computer X-ray photographic system or a direct digital X-ray photographic system, and the like, and can also be other medical photographic systems capable of acquiring X-ray images; the computer device may be, but is not limited to, various personal computers, notebook computers, smart phones, tablet computers, and portable wearable devices. The pedicle detection method can perform a series of analysis on the pedicle segmented image of the target spine of the imaging object, and determine the pedicle detection result corresponding to the target spine. Further, the offset of the pedicles in the spine relative to the center of the vertebral body can be analyzed according to the pedicle detection results corresponding to the target spine, so that the degree of rotation of the vertebral body can be obtained.

In one embodiment, as shown in fig. 2, there is provided a pedicle detection method, which is exemplified by the application of the method to the computer device in fig. 1, including the steps of:

and S100, determining the center point coordinates of the center points of the pedicle areas in the pedicle segmented image according to the pedicle segmented image of the target spine.

Specifically, the vertebra is a bone structure composed of a vertebral body, a pedicle, a lamina, a transverse process, a articular process and a spinous process; in general, the spine can be divided into five parts, cervical vertebrae, thoracic vertebrae, lumbar vertebrae, sacral vertebrae, and coccygeal vertebrae. The target spine may be at least a portion of five portions of the spine of the imaging subject. The pedicle segmented image of the target spine comprises an image of an area where a plurality of pedicles of the target spine are located.

It will be appreciated that the computer device may acquire a pedicle segmented image of the target spine of the imaging subject and perform arithmetic operations, analysis, comparison, and/or coordinate conversion on the pedicle segmented image to obtain center point coordinates of center points of the pedicle regions in the pedicle segmented image. Alternatively, the arithmetic operation processing may be addition, subtraction, division, multiplication, or the like; the analysis and treatment can be understood as a process of extracting the salient feature points of the pedicle region image in the pedicle segmented image; the above contrast process can be understood as a process of comparing the pedicle segmented image with a standard pedicle image of the imaging subject; the above coordinate conversion process may be understood as a process of adding an offset to the coordinates of each point on the edge of the pedicle region in the pedicle divided image to obtain the coordinates of the center point of the pedicle region, and the offset may be a two-dimensional coordinate, and the offset may be a coordinate (x ', y'). If the coordinates of one of the points on the edge of the pedicle region is (x, y), the coordinates of the central point of the pedicle region obtained by performing the coordinate conversion processing on the point are (x+x ', y+y').

The pedicle segmented image of the target spine can be an image generated by directly scanning data obtained by a medical scanning device by a computer device, or an image obtained by preprocessing an image generated by scanning data obtained by the medical scanning device by the computer device. The preprocessing may be denoising, data conversion, clipping, and/or analysis, etc. Alternatively, the pedicle region center points may be plural, and the total number of pedicle region center points may be determined based on the total number of pedicles in the target spine.

And S200, performing curve fitting processing on the central point coordinates of the central points of the pedicle regions to obtain a pedicle central curve equation.

Specifically, the computer device may perform curve fitting processing on the coordinates of the central points of all the central points of the pedicle regions of the target spine by using a curve fitting algorithm, so as to obtain a pedicle central curve equation. Alternatively, there may be one or two pedicle central curves. Alternatively, the curve fitting algorithm may be a method of approximating discrete data by using an analytical expression, a least square method, or the like, and of course, other curve fitting methods may be used, which is not limited to this embodiment.

In this embodiment, the computer device may perform curve fitting according to the following formula (1) to obtain a pedicle center curve equation Y, that is:

wherein N represents the order of the polynomial, K represents the polynomial parameter to be solved, and X and Y represent the abscissa and ordinate of the point on the fitting curve, respectively. N can be determined by the overall distribution of each pedicle in the pedicle segmentation image and is a tested value; preferably, in practical application, N may be 3 or 5, in which case the determined pedicle central curves may better fit the overall morphology of the central points of the pedicle regions.

S300, acquiring the center point coordinates of the center point of the missed pedicle according to the pedicle center curve equation and the center point coordinates of the center point of each pedicle region.

Specifically, the computer device may calculate the average value, that is, the average value coordinate, of the center point coordinates of the center points of the pedicle region through the center point coordinates of all or part of the center points of the pedicle region, then substitutes the abscissa or the ordinate of the average value coordinate into the pedicle center curve equation to obtain the ordinate or the abscissa, and then combines the ordinate or the abscissa with the abscissa or the ordinate in the average value coordinate to obtain the center point coordinate of the missed pedicle center point. Optionally, the computer device may also calculate a median value of the central point coordinates of the central points of the pedicle regions by using the central point coordinates of all or part of the central points of the pedicle regions, that is, the median coordinate, and then substituting the abscissa or the ordinate of the median coordinate into the pedicle central curve equation to obtain the ordinate or the abscissa, and then combining the ordinate or the abscissa with the abscissa or the ordinate in the median coordinate to obtain the central point coordinate of the central point of the pedicle to be missed. Or the computer equipment can also carry out coordinate conversion on the central point coordinates of all or part of the central points of the pedicle areas to obtain converted coordinates, then, calculating the mean value of the converted coordinates to obtain mean value coordinates, substituting the abscissa or the ordinate of the mean value coordinates into a pedicle central curve equation to obtain the ordinate or the abscissa, and then combining the ordinate or the abscissa with the abscissa or the ordinate in the mean value coordinates to obtain the central point coordinates of the central points of the missed pedicle; of course, the center point coordinates of the center point of the missed pedicle can also be obtained by other methods.

S400, determining the pedicle detection result corresponding to the target spine according to the center point coordinates of the center points of the pedicle areas and the center point coordinates of the center points of the missed detection pedicle.

Specifically, the computer device may combine the center point coordinates of the center points of the pedicle regions and the center point coordinates of the missing pedicle center points to obtain the pedicle detection results corresponding to the target spine. That is, the pedicle detection results may be the center point coordinates of all pedicle region center points and the missed pedicle center points.

In addition, in the pedicle detection process, if the center point coordinates of the center points of the missed detection pedicles are not detected, the center point coordinates of the center points of the pedicle areas can be directly used as pedicle detection results corresponding to the target spine.

In the pedicle detection method, the computer equipment can determine the center point coordinates of the center points of the pedicle areas in the pedicle segmentation image according to the pedicle segmentation image of the target spine, perform curve fitting treatment on the center point coordinates of the center points of the pedicle areas to obtain a pedicle center curve equation, obtain the center point coordinates of the missed pedicle according to the pedicle center curve equation and the center point coordinates of the center points of the pedicle areas, and determine the pedicle detection result corresponding to the target spine according to the center point coordinates of the center points of the pedicle areas and the center point coordinates of the center points of the missed pedicle; the method can be realized by a set of computer program, and can target the pedicle detection result corresponding to the spine without manual participation, so that the detection error can be reduced, the accuracy of the pedicle detection result is improved, human resources can be saved, the detection efficiency is improved, further, medical staff can accurately evaluate the rotation degree of the vertebral body according to the obtained pedicle detection result, and the problem that the medical staff cannot timely adopt an optimal diagnosis and treatment scheme to treat the scoliosis of an imaging object is avoided; meanwhile, the method can also acquire the coordinate of the center point of the missed detection pedicle of vertebral arch so as to avoid the condition that the center point of the vertebral arch pedicle area is missed, thereby further improving the accuracy of the obtained vertebral arch pedicle detection result corresponding to the target vertebral column.

As one embodiment, as shown in fig. 3, the step of determining the coordinates of the center point of each pedicle region in the pedicle divided image in S100 may be implemented by the following steps:

s110, acquiring the total number of the pixel points in each pedicle region and the coordinates of each pixel point.

Specifically, the computer device may count the total number of the pixel points included in each pedicle region in the pedicle segmented image of the target spine, and may also establish a two-dimensional rectangular coordinate system xoy in the pedicle segmented image, and determine coordinates of each pixel point included in each pedicle region. Optionally, the x-axis and the y-axis in the two-dimensional rectangular coordinate system xoy may also be any two vertical coordinate axes established by taking any one pixel point in the pedicle divided image as the position of the origin o. For example, a pixel point corresponding to any corner point in the pedicle divided image is used as the position of the origin o, and two vertical edge sections adjacent to the origin o in the pedicle divided image are respectively used as the x axis and the y axis in the two-dimensional rectangular coordinate system xoy. The two-dimensional rectangular coordinate system xoy established is different, and the coordinates of each pixel point in each pedicle region are also different.

In the pedicle segmented image of the target spine of the imaging object, the total number of the pixels included in the different pedicle regions may be equal or unequal. Optionally, the coordinates of different pixels contained in each pedicle region in the pedicle segmented image are different. The coordinates of the pixel points can be understood as the specific positions of the pixel points corresponding to the pedicle areas in the pedicle segmented image; in this embodiment, the coordinates of the pixel point may be two-dimensional coordinates.

And S120, obtaining the center point coordinates of the center points of the pedicle areas according to the total number of the pixel points in the pedicle areas and the coordinates of the pixel points.

Specifically, the computer device may perform arithmetic operation and/or coordinate conversion and the like on the total number of the pixel points in each pedicle region and coordinates of each pixel point, to obtain center point coordinates of center points of each pedicle region in the pedicle segmented image of the target spine of the imaging object. Alternatively, the arithmetic operations may be addition operations, subtraction operations, division operations, multiplication operations, logarithmic operations, and/or exponential operations, and the like.

In this embodiment, the computer device may calculate the center point coordinates (x, y) of the center point of each pedicle region according to the following formula (2), such as:

Wherein x is ^k Representing the abscissa, y, of the pixel points in the pedicle region ^k Representing the ordinate of the pixels in the pedicle region, n representing the total number of all pixels in each pedicle region, k representing the index of the position of the pixels in the pedicle region, x representing the abscissa of the center point of the pedicle region, and y representing the ordinate of the center point of the pedicle region.

The pedicle detection method can obtain the total number of the pixel points in each pedicle region and the coordinates of each pixel point, and obtain the center point coordinates of the center point of each pedicle region according to the total number of the pixel points in each pedicle region and the coordinates of each pixel point; the method can be realized by a set of computer program, and the center point coordinates of the center points of the pedicle areas can be determined without manual participation, so that the detection error can be reduced, the accuracy of the detection result can be improved, the manpower resources can be saved, and the detection efficiency can be improved.

As an embodiment, as shown in fig. 4, the step of obtaining the center point coordinates of the missing pedicle center point according to the pedicle center curve equation and the center point coordinates of the center points of the pedicle regions in S300 may be implemented by the following steps:

S310, according to a first preset direction, determining the central point coordinates of the central point of the first missed detection pedicle according to the pedicle central curve equation and the central point coordinates of the central points of the pedicle areas.

The central point coordinates of the central points of the pedicle areas comprise the central point coordinates of the central points of the pedicle areas on the left side of the target spine and the central point coordinates of the central points of the pedicle areas on the right side of the target spine.

Specifically, the pedicles in the spine are generally divided into left pedicles and right pedicles, and thus, each pedicle region in the pedicle segmentation image of the target spine includes a left pedicle region and a right pedicle region, and correspondingly, the center point coordinates of the center points of each pedicle region may include the center point coordinates of the center point of the pedicle region on the left side of the target spine and the center point coordinates of the center point of the pedicle region on the right side of the target spine. In this embodiment, the pedicle segmented image of the target spine includes images of at least two pedicle regions including at least any one of the left pedicle region and the corresponding right pedicle region.

It should be noted that, the first preset direction may be determined according to the distribution direction of each pedicle region in the pedicle segmented image. Alternatively, the direction of distribution of the pedicle regions in the pedicle segmented image may be understood as the direction of the target spine from beginning to end. Fig. 5 shows a pedicle segmented image of a target spine of an imaging subject, in which the region resembling an "S" shape and having a width represents a target spine region, which is displayed on the left and right sides with a plurality of irregularly black small regions, respectively, representing different pedicle region center points in the target spine, the target spine in fig. 5 being all the spines of the imaging subject. Since the distribution direction of each pedicle region in fig. 5 is a top-to-bottom or bottom-to-top direction, and is specifically determined according to the head and tail of the target spine, in this case, the first preset direction may be a top-to-bottom or bottom-to-top direction. With continued reference to the example of fig. 5, fig. 6 shows a pedicle segmented image, and fig. 6 is the same as the pedicle segmented image of fig. 5 except that the left pedicle center curve is marked in fig. 6 as a left pedicle region center point fit, and the right pedicle center curve is marked as a right pedicle region center point fit.

In addition, if the distribution direction of each pedicle region in the pedicle segmented image is a left-to-right or right-to-left direction, specifically determined according to the head and tail of the target spine, the first preset direction may be determined as a left-to-right or right-to-left direction. In addition, the distribution direction and the first preset direction may be other directions, which is not limited in this embodiment. Alternatively, the distribution direction of each pedicle region in the pedicle segmented image may be determined in azimuth in addition to the up-down-left-right direction. Such as east, south, west, north, southeast, northeast, southwest, northwest, etc. orientations.

In this embodiment, the first preset direction may be a direction from the head end to the tail end of the target spine or a direction from the tail end to the head end of the target spine.

It can be appreciated that the computer device may perform, according to the first preset direction, processing such as arithmetic operation, analysis, comparison, and/or coordinate conversion on the center point coordinates of the center points of the pedicle regions, to obtain processed center point coordinates, and determine, according to the processed center point coordinates and the pedicle center curve equation, the center point coordinates of the center point of the first missed approach pedicle.

S320, according to a second preset direction, determining the center point coordinates of the center point of the second missed detection pedicle according to the pedicle center curve equation, the center point coordinates of the center points of the pedicle regions and the center point coordinates of the center point of the first missed detection pedicle. Wherein, any one of the first preset direction and the second preset direction is the direction from the head end to the tail end of the target spine, and the other preset direction is the direction from the tail end to the head end of the target spine.

Specifically, the second preset direction may be opposite to the first preset direction. The second preset direction may also be a direction from the left side to the right side of the target spine or a direction from the right side to the left side of the target spine. However, in this embodiment, if the first preset direction is the direction from the cephalad end to the caudal end of the target spine, the second preset direction is the direction from the caudal end to the cephalad end of the target spine; if the first predetermined direction is a direction from the caudal end to the cephalad end of the target spinal column, the second predetermined direction is a direction from the cephalad end to the caudal end of the target spinal column.

It should be noted that, the central point coordinates of the central points of the pedicle areas and the determined central point coordinates of the central points of the first missed detection pedicle are combined, and further, the computer device may perform arithmetic operation, analysis, comparison, coordinate conversion and other processes on the central point coordinates of the central points of the pedicle areas and the central point coordinates of the central points of the first missed detection pedicle according to the second preset direction, so as to obtain the processed central point coordinates, and determine the central point coordinates of the central points of the second missed detection pedicle according to the processed central point coordinates and the pedicle central curve equation.

S330, determining the center point coordinates of the center point of the missed detection pedicle based on the center point coordinates of the center point of the first missed detection pedicle and the center point coordinates of the center point of the second missed detection pedicle.

In this embodiment, the determined center point coordinates of the first missed-detection pedicle of vertebral arch center point and the determined center point coordinates of the second missed-detection pedicle of vertebral arch center point have overlapping coordinates, so that the computer device may only reserve one of the overlapping coordinates of the center point coordinates of the first missed-detection pedicle of vertebral arch center point and the center point coordinates of the second missed-detection pedicle of vertebral arch center point, delete one of the overlapping coordinates, and determine all remaining center point coordinates as the center point coordinates of the missed-detection pedicle of vertebral arch center point.

For example, the coordinates of the center point of the first missed approach pedicle determined according to the first preset direction are: coordinates 1 (1, 2), coordinates 2 (2, 3), coordinates 3 (1, 2), and the center point coordinates of the second missed approach pedicle center point determined according to the second preset direction are: coordinate 4 (2, 3), coordinate 5 (3,2.4), coordinate 6 (2,3.5), wherein coordinate 1 and coordinate 3 overlap, and coordinate 2 and coordinate 4 overlap, therefore, only coordinate 1 or coordinate 3, and coordinate 2 or coordinate 4 can be retained, and the center point coordinates of all the missed pedicle center points finally obtained are: coordinate 1 or coordinate 3, coordinate 2 or coordinate 4, coordinate 5, coordinate 6.

According to the pedicle detection method, the coordinate of the central point of the missed detection pedicle can be obtained according to the pedicle central curve equation and the coordinate of the central point of each pedicle region, so that the condition that the central point of the pedicle region is missed in the pedicle detection process can be avoided, and the accuracy of the pedicle detection result corresponding to the obtained target spine is further improved.

The steps of S310 and S320 described above will be described in detail.

In an embodiment, as shown in fig. 7, the step of determining, in the S310, the coordinates of the center point of the first missed approach pedicle according to the first preset direction and according to the pedicle center curve equation and the coordinates of the center point of the region of each pedicle may include:

s311, acquiring a plurality of first center point combinations according to the first preset direction and the position relation of each pedicle region; each first center point combination includes three adjacent center points.

In particular, the positional relationship of the pedicle regions can be understood as a positional arrangement sequence of the pedicle regions determined according to the distribution direction of the pedicle regions. In this embodiment, the above pedicle segmented image is a segmented image with a label, so that the computer device may determine each left pedicle region and each right pedicle region in the pedicle segmented image, then further obtain the center point coordinates of the center point of each left pedicle region and the center point coordinates of the center point of each right pedicle region in the pedicle segmented image, and detect the left pedicle region and the right pedicle region in the pedicle segmented image according to the center point coordinates of the center point of each left pedicle region and the center point coordinates of the center point of each right pedicle region, and finally obtain the left pedicle detection result and the right pedicle detection result.

Illustratively, the target spine region in the pedicle segmented image comprises a left pedicle region 1, a left pedicle region 2, a left pedicle region 3, a left pedicle region 4 and a left pedicle region 5, if the distribution direction of each left pedicle region in the pedicle segmented image is from top to bottom, the left pedicle region 1 is the uppermost pedicle region, the left pedicle region 3 is adjacent to the left pedicle region 1 according to the order from top to bottom, the left pedicle region 4 is adjacent to the left pedicle region 3, the left pedicle region 2 is adjacent to the left pedicle region 4, and the left pedicle region 5 is adjacent to the left pedicle region 2, i.e. the position arrangement sequence of each left pedicle region in the pedicle segmented image is the left pedicle region 1, the left pedicle region 3, the left pedicle region 4, the left pedicle region 2 and the left pedicle region 5 in sequence. In this embodiment, if the left pedicle regions in the pedicle segmented image correspond to other distribution directions, the positional relationship of the right pedicle regions is also similar, and will not be described again.

It can also be appreciated that the computer device can obtain a plurality of first center point combinations according to the first preset direction and the positional relationship of the pedicle regions; each first center point combination includes the center points of three adjacent pedicle regions.

With continued reference to the example of fig. 5, according to the distribution direction from top to bottom, the position relationship of each right pedicle region is a left pedicle region 1, a left pedicle region 3, a left pedicle region 4, a left pedicle region 2 and a left pedicle region 5, and the computer device may obtain, according to the position relationship, all three adjacent combinations of left pedicle regions, which are respectively region combination 1: left pedicle region 1, left pedicle region 3, and left pedicle region 4, region combination 2: left pedicle region 3, left pedicle region 4, and left pedicle region 2, region combination 3: the left pedicle region 4, the left pedicle region 2 and the left pedicle region 5, and further, a corresponding first center point combination is obtained from the region combinations. Alternatively, the first center point combination may be a combination of center points corresponding to each region in each region combination. For example, the first center point combination 1 corresponding to the region combination 1: left pedicle region center point 1, left pedicle region center point 3, and left pedicle region center point 4, first center point combination 2 corresponding to region combination 2: left pedicle region center point 3, left pedicle region center point 4 and left pedicle region center point 2, first center point combination 3 corresponding to region combination 3: a left pedicle region center point 4, a left pedicle region center point 2, and a left pedicle region center point 5.

S312, acquiring first distance evaluation values of each first center point combination according to the center point coordinates of three adjacent center points in each first center point combination; the first distance evaluation value represents the deviation degree of the distances between every two adjacent pedicle region central points in each first central point combination.

Specifically, the computer device may perform processes such as screening, coordinate conversion, comparison, and/or arithmetic operation on the coordinates of the central points of the pedicle regions in each first central point combination, to obtain the distance evaluation value of each first central point combination.

Or, the computer device may further perform processes such as screening, coordinate conversion, comparison, and/or arithmetic operation on the coordinates of the central points of the adjacent pedicle regions in each first central point combination, and then perform arithmetic operation on the processing results corresponding to the central points of the adjacent pedicle regions to obtain the first distance evaluation value of each first central point combination. The above screening process can be understood as a process of screening the center point coordinates of the center points of the partial pedicle regions from the center point coordinates of the center points of the multiple pedicle regions. The above comparison process may be understood as a process of comparing the center point coordinates of the center points of the pedicle region with a threshold value to determine a first distance evaluation value corresponding to the first center point combination according to the comparison result.

S313, if the first distance evaluation value is larger than a first preset threshold value, determining that a first missed detection pedicle central point exists in the current first central point combination, and determining the central point coordinates of the first missed detection pedicle central point according to the central point coordinates of the pedicle region central point and the pedicle central curve equation in the current first central point combination.

It will be appreciated that the computer device may determine whether the first distance evaluation value of the current first center point combination is greater than a first preset threshold, and if it is determined that the first distance evaluation value of the current first center point combination is greater than the first preset threshold, determine that a first missed pedicle center point exists in the current first center point combination, that is, that a first missed pedicle region exists among three adjacent pedicle regions included in the current first center point combination. Optionally, the center point coordinates of the center point of the first missed-detection pedicle may include center point coordinates of the center point of the left first missed-detection pedicle and center point coordinates of the center point of the right first missed-detection pedicle corresponding to the pedicle region.

Optionally, the first missed detection pedicle center point may be all the missed detection pedicle center points corresponding to the pedicle region, or may be part of the missed detection pedicle center points corresponding to the pedicle region. If the first missed detection pedicle central point is the center point of all the missed detection pedicle corresponding to the pedicle region, the second missed detection pedicle central point does not exist in practice; if the first missed detection pedicle central point is a partial missed detection pedicle central point corresponding to the pedicle region, the second missed detection pedicle central point exists, and all the first missed detection pedicle central points and all the second missed detection pedicle central points are determined to be all the missed detection pedicle central points corresponding to the pedicle region.

In this embodiment, the pedicle central curve equation may be a left pedicle central curve equation and a right pedicle central curve equation. The left pedicle central curve equation is determined by the central point coordinates of the central point of the left pedicle region in the pedicle segmented image, and the right pedicle central curve equation is determined by the central point coordinates of the central point of the right pedicle region in the pedicle segmented image. The computer device may obtain the left first center point combination and the right first center point combination, and then determine the center point coordinates of the left first missed approach pedicle center point and the center point coordinates of the right first missed approach pedicle center point by the left pedicle center curve equation, the right pedicle center curve equation, the left first center point combination, and the right first center point combination.

According to the pedicle detection method, the first missed detection pedicle center point coordinate can be obtained according to the pedicle center curve equation and the center point coordinates of the center points of the pedicle areas, so that the condition that the center points of the pedicle areas are missed in the pedicle detection process can be avoided, and the accuracy of the obtained pedicle detection result corresponding to the target spine is further improved.

In another embodiment, as shown in fig. 8, the step of determining the center point coordinates of the center point of the second missed approach pedicle according to the second preset direction in S320 according to the pedicle center curve equation, the center point coordinates of the center points of the respective pedicle regions, and the center point coordinates of the center point of the first missed approach pedicle may include:

s321, acquiring a plurality of second center point combinations according to the second preset direction, the position relation of each pedicle region and the position relation of the first missed pedicle region. Each second center point combination comprises all the center points of the pedicle areas and three adjacent center points in the center points of the first missed approach pedicle.

Specifically, the positional relationship of the first missed pedicle region may be understood as a positional arrangement sequence of the first missed pedicle regions determined according to the distribution direction of the first missed pedicle regions. The computer equipment can comprehensively arrange the positions of all the central points of the pedicle areas and all the central points of the first missed detection pedicle according to the position relation of the pedicle areas and the position relation of the first missed detection pedicle areas, and further divide all the central points of the pedicle areas and all the central points of the first missed detection pedicle according to the second preset direction and the comprehensive arrangement sequence result to obtain a plurality of second central point combinations.

It should be noted that, each second center point combination may include a pedicle region center point and a first missed pedicle center point, or may include only three pedicle region center points.

The specific dividing process in step S321 is similar to that in step S311, and will not be described again.

S322, acquiring second distance evaluation values of each second center point combination according to the center point coordinates of three adjacent center points in each second center point combination; the second distance evaluation value represents the deviation degree of the distances between every two adjacent center points in each second center point combination.

Specifically, the computer device may perform processes such as screening, coordinate conversion, comparison, and/or arithmetic operation on the coordinates of the central points of the pedicle regions in each second central point combination, to obtain two-distance evaluation values of each second central point combination.

Or, the computer device may further perform processes such as screening, coordinate conversion, comparison, and/or arithmetic operation on the coordinates of the central points of the adjacent pedicle regions in each second central point combination, and then perform arithmetic operation on the processing results corresponding to the central points of the adjacent pedicle regions to obtain second distance evaluation values of each second central point combination.

S323, if the second distance evaluation value is larger than a second preset threshold value, determining that a second missed detection pedicle center point exists in the current second center point combination, and determining the center point coordinates of the second missed detection pedicle center point according to the center point coordinates of the pedicle region center point and the pedicle center curve equation in the current second center point combination.

It will be appreciated that the computer device may determine whether the second distance evaluation value of the current second center point combination is greater than a second preset threshold, and if it is determined that the second distance evaluation value of the current second center point combination is greater than the second preset threshold, determine that a second missed pedicle center point exists in the current second center point combination, that is, that a second missed pedicle region exists among three adjacent pedicle regions included in the current second center point combination. Optionally, the center point coordinates of the center point of the second missed-detection pedicle may include center point coordinates of the center point of the left second missed-detection pedicle and center point coordinates of the center point of the right second missed-detection pedicle in the pedicle region and the first missed-detection pedicle region. Optionally, the second preset threshold may be equal to or different from the first preset threshold, which is not limited in this embodiment.

In this embodiment, the computer device may obtain the left second center point combination and the right second center point combination, respectively, and then determine the center point coordinates of the left second missed approach pedicle center point and the center point coordinates of the right second missed approach pedicle center point by using the left pedicle center curve equation, the right pedicle center curve equation, the left second center point combination, and the right second center point combination.

According to the pedicle detection method, the second missed detection pedicle center point coordinate can be obtained according to the pedicle center curve equation, the center point coordinates of the center points of the pedicle areas and the center point coordinates of the first missed detection pedicle center point, so that the condition that the center points of the pedicle areas are missed in the pedicle detection process can be avoided, and the accuracy of the obtained pedicle detection result corresponding to the target spine is further improved.

As one embodiment, as shown in fig. 9, the step of obtaining the first distance evaluation value of each first center point combination according to the center point coordinates of three adjacent center points in each first center point combination in S312 may be implemented by the following steps:

s3121, for each first center point combination, obtaining a first distance between the first two adjacent center points in the three adjacent center points and a second distance between the last two adjacent center points in the three adjacent center points according to coordinates of the three adjacent center points in the first center point combination.

Specifically, for each first center point combination, the computer device may calculate, according to the coordinates of the center points of three adjacent center points in each first center point combination, a distance between two adjacent pedicle region center points in each first center point combination by using a euclidean distance method or a trigonometric function method, to obtain a first distance L1 between the first two adjacent center points in the three adjacent center points in each first center point combination, and a second distance L2 between the last two adjacent center points in the three adjacent center points. Alternatively, the euclidean distance method may be a manhattan distance calculation method, a euclidean distance calculation method, or a cosine similarity calculation method, or the like.

S3122, determining a ratio between the first distance and the second distance as a first distance evaluation value of the first center point combination.

Specifically, the computer device may perform an arithmetic operation on the first distance L1 and the second distance L2 corresponding to each first center point combination, to obtain a first distance evaluation value of each first center point combination. Alternatively, the arithmetic operations may be addition operations, subtraction operations, division operations, multiplication operations, exponent operations, and/or logarithmic operations, among others. However, in this embodiment, the computer device may obtain the first distance evaluation value of each first center point combination by using the ratio (i.e., L1/L2) between the first distance L1 and the second distance L2 corresponding to each first center point combination.

Further, in S313, if the first distance evaluation value is greater than the first preset threshold, the step of determining that the missing pedicle center point exists in the current first center point combination may include: if the first distance evaluation value is larger than a first preset threshold value, determining that a missing pedicle center point exists between the first two adjacent center points in the three adjacent center points of the current first center point combination.

In this embodiment, if the computer device determines that the first distance evaluation value of the current first center point combination is greater than the first preset threshold, it may determine that there is a missing pedicle center point between the first two adjacent center points of the three adjacent center points of the current first center point combination, that is, there is no missing pedicle center point between the last two adjacent center points of the three adjacent center points of the current first center point combination.

Meanwhile, the specific implementation process of step S322 is similar to that of step S312, and the description of this embodiment is omitted.

According to the pedicle detection method, the first distance evaluation value of each first center point combination can be obtained, and the center point coordinates corresponding to the missed pedicle center points in each first center point combination are further determined according to the first distance evaluation value of each first center point combination, so that the condition that the center points of the pedicle areas are missed in the pedicle detection process can be avoided, and the accuracy of the pedicle detection result corresponding to the obtained target spine is further improved.

In some scenarios, as shown in fig. 10, in order to reduce the computation load, in an embodiment, the step of determining the first missed pedicle central point coordinate in S313 according to the central point coordinate of the central point of the pedicle region in the current first central point combination and the pedicle central curve equation may specifically include:

s3131, determining an average value of first coordinates of the first two adjacent central points in the current first central point combination as the first coordinates of the first missed-detection pedicle central point.

Specifically, the computer device may calculate an average value of the first coordinates of the central points of the first two adjacent pedicle regions in each first central point combination, and use the calculated average value of the first coordinates as the first coordinates of the central point of the first missed approach pedicle. Alternatively, the first coordinate may be an abscissa or an ordinate.

In this embodiment, the first two adjacent center points and the second two adjacent center points in each first center point combination may be determined according to the positional relationship of each pedicle region. With continued reference to the example of fig. 5, the obtained corresponding first center point combination 1 is a left pedicle region center point 1, a left pedicle region center point 3 and a left pedicle region center point 4, and correspondingly, the first two adjacent center points in the first center point combination 1 are the left pedicle region center point 1 and the left pedicle region center point 3, and the last two adjacent center points in the first center point combination 1 are the left pedicle region center point 3 and the left pedicle region center point 4; the method of determining the first two adjacent central points and the second two adjacent central points from the three adjacent pedicle region central points in the other first central point combinations is also similar, and will not be described again.

S3132, substituting an average value of first coordinates of the first two adjacent central points in the current first central point combination into a pedicle central curve equation to obtain second coordinates of the first missed-detection pedicle central points, and determining the central point coordinates of the first missed-detection pedicle central points through the first coordinates and the second coordinates of the first missed-detection pedicle central points.

It will be appreciated that if the first coordinate is an abscissa, the second coordinate may be an ordinate; if the first coordinate is the ordinate, the second coordinate may be the abscissa. Alternatively, the computer device may calculate an average value of the second coordinates of the central points of the first two adjacent pedicle regions in each of the first central point combinations, and use the calculated average value of the second coordinates as the second coordinates of the central points of the first missed approach pedicle. Further, combining the first coordinate and the second coordinate of the center point of the first missed-detection pedicle of vertebral arch to obtain the center point coordinate of the center point of the first missed-detection pedicle of vertebral arch.

With continued reference to the example of fig. 5, fig. 11 shows a pedicle segmented image, and fig. 11 is identical to the pedicle segmented image of fig. 5 except that left, right, and right pedicle regions are marked in fig. 11, all of which are represented by irregularly black small regions in the pedicle segmented image. Based on fig. 11, in fig. 12, all the pedicle region center points and all the missed pedicle center points (i.e., the first missed pedicle center point and the second missed pedicle center point) in the pedicle segmented image are correspondingly shown, and the small black circles on the target spine in fig. 12 represent the pedicle region center points or the missed pedicle center points.

In order to ensure the singleness of the obtained coordinates of the center point of the missed detection pedicle, in practical application, if the distribution direction of the pedicle region in the pedicle segmented image is parallel to the length direction of the pedicle segmented image, the first coordinate may be the ordinate, and the second coordinate may be the abscissa; if the distribution direction of the pedicle region in the pedicle segmented image is parallel to the width direction of the pedicle segmented image, the first coordinate may be an abscissa and the second coordinate may be an ordinate.

The pedicle detection method can acquire the center point coordinates of the center point of the second missed detection pedicle, so that the condition that the center point of the pedicle region is missed in the pedicle detection process can be avoided, and the accuracy of the pedicle detection result corresponding to the acquired target spine is further improved.

As one embodiment, before performing the above S100, as shown in fig. 13, the above pedicle detection method may further include:

s500, acquiring a medical image of the target spine.

In this embodiment, the medical scanning apparatus may perform real-time scanning on a target spine of an imaging object to obtain target spine scanning data. In this embodiment, the medical scanning apparatus may set the pixel interval resampling of the medical image to a target value (i.e., the interval of data scanned in both x-axis and y-axis directions is normalized to the target value, for example, 0.7 mm). If the target spine scanning data obtained by the scanning of the medical scanning equipment does not meet the pixel interval resampling parameters, the computer equipment needs to perform interval normalization processing on the target spine scanning data, and then rebuilds the interval normalization processing result to obtain at least one frame of medical image of the target spine. Alternatively, the interval normalization process may be to calculate the average value of the gray values of all the pixels in the medical image to obtain a gray average value, calculate the standard deviation of the gray values of the medical image to obtain a gray standard deviation, and then subtract the gray average value from the gray values of all the pixels in the medical image and divide the gray standard deviation to obtain a normalization result.

Alternatively, the computer device may obtain, from the cloud or locally, a medical image of at least one frame of the target spine obtained in advance. If the computer equipment acquires the medical images of the multi-frame target spinal column, a frame of medical image of the target spinal column with highest resolution can be selected from the medical images of the multi-frame target spinal column.

It should be noted that the medical image may be an electronic computed tomography image, a magnetic resonance image, or the like, but in the present embodiment, the medical image is an X-ray image and is an orthotopic X-ray image of the target spine.

S600, segmenting the pedicle in the medical image of the target spine through the pedicle segmentation model to obtain a pedicle segmentation image.

Specifically, the pedicle segmentation model may be a pre-trained segmentation network model. The computer device can input the medical image of the target spine into a pedicle segmentation model, the pedicle segmentation model segments the pedicle in the medical image of the target spine to obtain a pedicle mask image, and then the pedicle mask image and the medical image are mapped to obtain a pedicle segmentation image. Alternatively, the pedicle mask image is the same size as the medical image, and can be understood as a labeled binary image. The pixel value of the pedicle region in the pedicle mask image is 1, and the pixel values of other regions are all 0.

The pedicle segmented image may include at least two pedicle regions (i.e., at least any one left pedicle region and a corresponding right pedicle region) and a background region image, that is, an image of the pedicle segmented image excluding the vertebral body, lamina, transverse processes, articular processes, and spinous process regions of the target spine, in which case the pedicle segmented image of the target spine may be understood as a noiseless segmented image.

However, in this embodiment, the pedicle segmented image of the target spine may include an image of other bone structure regions around the pedicle region in the target spine in addition to the image of at least two pedicle regions and the background region image, in which case the pedicle segmented image of the target spine may be understood as a noisy segmented image. The above background area image may be understood as an image of the area correspondence between different bone structures. The pedicle segmented image in fig. 5 is a noisy segmented image.

It is understood that the pedicle segmentation model may be composed of at least one of a convolutional neural network model, a recurrent neural network model, and an antagonistic neural network model; the computer equipment can perform network model training on the initial pedicle segmentation model through an image set formed by medical images of different spines of different imaging objects to obtain a pedicle segmentation model (namely a pre-trained segmentation network model).

Specifically, the computer device may input medical images of different spines in the image set into the initial pedicle segmentation model to obtain a pedicle segmentation prediction result, calculate a prediction error value between the pedicle segmentation prediction result and a standard pedicle segmentation result through a loss function, update initial network parameters in the initial pedicle segmentation model according to the prediction error value, and iterate the training steps until the prediction error value meets a preset error threshold or the iteration number reaches a preset iteration number threshold, so as to obtain the pedicle segmentation model. The standard pedicle segmentation result may be an idealized pedicle segmentation result. In the training process, medical images with larger sizes are required to be input into an initial pedicle segmentation model for training so as to obtain a larger receptive field.

Since the medical image is two-dimensionally distributed information, such as information distributed on the x-axis and the y-axis, the medical scanning apparatus can perform downsampling multiple times in both directions of the x-axis and the y-axis when scanning. In this embodiment, the pedicle segmentation model is formed by a multi-layer deep convolutional neural network structure, for example, a multi-layer deep convolutional neural network structure including a UNet network structure is taken as an example, that is, the pedicle segmentation model is a UNet segmentation model. The UNet segmentation model includes the structure of fig. 14 and the convolutional layers after the structure of fig. 14.

In the training process, the size of the medical image in the image set of the UNet segmentation model is 640×512, and then the first layer in the UNet segmentation model comprises two convolution layers, and the input medical image of 640×512 is subjected to convolution processing twice to obtain 16 feature images of 640×512, wherein 16 represents 16 convolution kernels in the process of convolution processing twice. Further, the 16 feature images of 640×512 are subjected to downsampling through a second layer and then subjected to convolution processing once again to obtain 32 feature images of 320×256, 32 convolution kernels are sequentially processed in the convolution processing process of the layer until 512 feature images of 10×8 are obtained, and then the 512 feature images of 10×8 are subjected to upsampling and convolution processing and then are subjected to convolution processing with the result of downsampling of the last time, and the 512 feature images of 20×16 are spliced to obtain 1024 feature images of 20×16. Further, up-sampling and convolution processing are carried out on 1024 20×16 feature images, then the up-sampling and convolution processing is carried out on the result 256 40×32 feature images after the last down-sampling, 512 40×32 feature images are obtained, other layers are similar in the same way, and finally 32 640×512 feature images are obtained. And, the final layer of convolution layer in the UNet segmentation model outputs pedicle segmentation mask images after 32 feature images of 640 x 512.

Alternatively, the dashed line in fig. 14 indicates a process of splicing the result output from the front side with the result after the current layer downsampling and convolution processing, the right triangle indicates the convolution layer, and the downward arrow indicates the downsampling processing. The convolution kernel of the convolution processing is 3×3, the step size of the convolution layer is 1×1, the step size of the downsampling is 2×2, and a normalization processing is needed before each convolution processing. The convolution kernel of the upsampling is 2 x 2.

In addition, the above-mentioned segmentation of the pedicles in the medical image of the target spine by the pedicle segmentation model may also be understood as a process of three-classification segmentation of the medical image. The pedicle segmentation model may include three channel outputs for outputting three channel images, which are respectively images of a left pedicle region image, a right pedicle region image, a background region, and other bone structure region combinations in the pedicle segmentation image, and the three regions may be labeled images. The pedicle segmented image can be a labeled image formed by combining three parts of an image of a left pedicle region, an image of a right pedicle region, a background region and other bone structure regions.

According to the pedicle detection method, the medical image of the target spine can be obtained, the pedicle in the medical image of the target spine is segmented through the pedicle segmentation model to obtain the pedicle segmentation image, so that the image detection range can be reduced, only the pedicle region in the pedicle segmentation image is detected, the pedicle detection result is obtained, the calculation amount of an algorithm is reduced, the pedicle detection time is shortened, and the detection efficiency is improved; meanwhile, the central point of the pedicle region can be detected in the effective region, and the detection accuracy can be improved.

It should be understood that, although the steps in the flowcharts of fig. 2-4, 7-10, and 13 are shown in order as indicated by the arrows, these steps are not necessarily performed in order as indicated by the arrows. The steps are not strictly limited to the order of execution unless explicitly recited herein, and the steps may be executed in other orders. Moreover, at least some of the steps of FIGS. 2-4, 7-10, and 13 may include multiple steps or stages that are not necessarily performed at the same time, but may be performed at different times, nor does the order in which the steps or stages are performed necessarily occur sequentially, but may be performed alternately or alternately with other steps or at least a portion of the steps or stages in other steps.

In one embodiment, as shown in fig. 15, there is provided a pedicle detection device comprising: a first coordinate determining module 11, a fitting processing module 12, a second coordinate determining module 13, and a detection result determining module 14, wherein:

the first coordinate determining module 11 is used for determining the center point coordinates of the center points of the pedicle areas in the pedicle segmented image according to the pedicle segmented image of the target spine;

the fitting processing module 12 is used for performing curve fitting processing on the coordinates of the central points of the pedicle areas to obtain a pedicle central curve equation;

the second coordinate determining module 13 is configured to obtain coordinates of center points of the missed detection pedicles according to a pediculus arcus vertebrae center curve equation and coordinates of center points of the pediculus arcus vertebrae areas;

the detection result determining module 14 is configured to determine a pedicle detection result corresponding to the target spine according to the center point coordinates of the center points of the pedicle regions and the center point coordinates of the center points of the missed pedicles.

The pedicle detection device provided in this embodiment may implement the above method embodiment, and its implementation principle and technical effects are similar, and will not be described in detail herein.

In one embodiment, the first coordinate determination module 11 includes: an information acquisition unit and a first coordinate acquisition unit, wherein:

The information acquisition unit is used for acquiring the total number of the pixel points in each pedicle region and the coordinates of each pixel point;

the first coordinate acquisition unit is used for obtaining the center point coordinates of the center points of the pedicle areas according to the total number of the pixel points in the pedicle areas and the coordinates of the pixel points.

In one embodiment, the second coordinate determination module 13 includes: the device comprises a first central point coordinate acquisition unit, a second central point coordinate acquisition unit and a third central point coordinate acquisition unit, wherein:

the first central point coordinate acquisition unit is used for determining the central point coordinate of the central point of the first missed-detection pedicle according to the central curve equation of the pedicle and the central point coordinate of the central point of each pedicle region according to a first preset direction;

the second central point coordinate acquisition unit is used for determining the central point coordinate of the central point of the second missed-detection pedicle according to a second preset direction, the central point coordinate of the central point of each pedicle region and the central point coordinate of the central point of the first missed-detection pedicle according to a pedicle central curve equation; any one of the first preset direction and the second preset direction is the direction from the head end to the tail end of the target spine, and the other preset direction is the direction from the tail end to the head end of the target spine;

The third central point coordinate acquisition unit is used for determining the central point coordinate of the central point of the missed-detection pedicle of vertebral arch according to the central point coordinate of the central point of the first missed-detection pedicle of vertebral arch and the central point coordinate of the central point of the second missed-detection pedicle of vertebral arch.

In one embodiment, the first center point coordinate acquisition unit includes: a first acquisition subunit, a second acquisition subunit, and a third acquisition subunit, wherein:

the first acquisition subunit is used for acquiring a plurality of first center point combinations according to the first preset direction and the position relation of each pedicle region; each first center point combination comprises three adjacent center points;

the second acquisition subunit is used for acquiring first distance evaluation values of each first center point combination according to the center point coordinates of three adjacent center points in each first center point combination; the first distance evaluation value represents the deviation degree of the distance between every two adjacent center points in each first center point combination;

and the third acquisition subunit is used for determining that a first missed detection pedicle central point exists in the current first central point combination when the first distance evaluation value is larger than a first preset threshold value, and determining the central point coordinates of the first missed detection pedicle central point according to the central point coordinates of the pedicle region central point in the current first central point combination and the pedicle central curve equation.

In one embodiment, the second center point coordinate acquisition unit includes: a fourth acquisition subunit, a fifth acquisition subunit, and a sixth acquisition subunit, wherein:

a fourth obtaining subunit, configured to obtain a plurality of second center point combinations according to a second preset direction, a position relationship of each pedicle region, and a position relationship of the first missed pedicle region; each second center point combination comprises three adjacent center points in all pedicle region center points and all first missed detection pedicle center points;

a fifth obtaining subunit, configured to obtain second distance evaluation values of each second center point combination according to center point coordinates of three adjacent center points in each second center point combination; the second distance evaluation value represents the deviation degree of the distance between every two adjacent center points in each second center point combination;

and the sixth acquisition subunit is used for determining that a second missed detection pedicle central point exists in the current second central point combination when the second distance evaluation value is larger than a second preset threshold value, and determining the central point coordinates of the second missed detection pedicle central point according to the central point coordinates of the pedicle region central point in the current second central point combination and the pedicle central curve equation.

In one embodiment, the second obtaining subunit is specifically configured to obtain, for each first center point combination, a first distance between a first two adjacent center points of the three adjacent center points and a second distance between a second two adjacent center points of the three adjacent center points according to coordinates of three adjacent center points in the first center point combination, and determine a ratio between the first distance and the second distance as a first distance evaluation value of the first center point combination.

In one embodiment, the third acquisition subunit comprises: a missed detection point determination subunit, wherein:

and the missed detection point determining subunit is used for determining that a missed detection pedicle central point exists between the first two adjacent central points in the three adjacent central points of the current first central point combination when the first distance evaluation value is larger than a first preset threshold value.

In one embodiment, the third acquisition subunit further comprises: an ordinate acquisition subunit and an abscissa acquisition subunit, wherein:

the ordinate acquisition subunit is used for determining an average value of first coordinates of the first two adjacent central points in the current first central point combination as the first coordinate of the first missed detection pedicle central point;

the abscissa obtaining subunit is used for substituting the average value of the first coordinates of the first two adjacent central points in the current first central point combination into a pedicle central curve equation to obtain the second coordinates of the first missed-detection pedicle central points, and determining the central point coordinates of the first missed-detection pedicle central points through the first coordinates and the second coordinates of the first missed-detection pedicle central points;

In one embodiment, the pedicle detection device further comprises: the medical image acquisition module and the segmentation module, wherein:

acquiring a medical image of a target spine;

the segmentation module is used for segmenting the pedicles in the medical image of the target spine through the pedicle segmentation model to obtain pedicle segmentation images.

For specific limitations of the pedicle detection device, reference is made to the above limitations of the pedicle detection method, and no further description is given here. The various modules in the pedicle detection device described above may be implemented in whole or in part by software, hardware, and combinations thereof. The above modules may be embedded in hardware or may be independent of a processor in the computer device, or may be stored in software in a memory in the computer device, so that the processor may call and execute operations corresponding to the above modules.

In one embodiment, a computer device is provided, which may be a server, and the internal structure of which may be as shown in fig. 16. The computer device includes a processor, a memory, and a network interface connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The database of the computer device is used to store medical images and pedicle segmented images. The network interface of the computer device is for communicating with an external endpoint via a network connection. The computer program is executed by a processor to implement a pedicle detection method.

It will be appreciated by those skilled in the art that the structure shown in FIG. 16 is merely a block diagram of some of the structures associated with the present inventive arrangements and is not limiting of the computer device to which the present inventive arrangements may be applied, and that a particular computer device may include more or fewer components than shown, or may combine some of the components, or have a different arrangement of components.

In one embodiment, a computer device is provided comprising a memory and a processor, the memory having stored therein a computer program, the processor when executing the computer program performing the steps of:

In one embodiment, a readable storage medium is provided having a computer program stored thereon, which when executed by a processor, performs the steps of:

In one embodiment, a computer program product is provided comprising a computer program which, when executed by a processor, performs the steps of:

Those skilled in the art will appreciate that implementing all or part of the above described methods may be accomplished by way of a computer program stored on a non-transitory computer readable storage medium, which when executed, may comprise the steps of the embodiments of the methods described above. Any reference to memory, storage, database, or other medium used in embodiments provided herein may include at least one of non-volatile and volatile memory. The nonvolatile Memory may include Read-Only Memory (ROM), magnetic tape, floppy disk, flash Memory, optical Memory, or the like. Volatile memory can include random access memory (Random Access Memory, RAM) or external cache memory. By way of illustration, and not limitation, RAM can be in the form of a variety of forms, such as static random access memory (Static Random Access Memory, SRAM) or dynamic random access memory (Dynamic Random Access Memory, DRAM), and the like.

The technical features of the above embodiments may be arbitrarily combined, and all possible combinations of the technical features in the above embodiments are not described for brevity of description, however, as long as there is no contradiction between the combinations of the technical features, they should be considered as the scope of the description.

The above examples illustrate only a few embodiments of the application, which are described in detail and are not to be construed as limiting the scope of the application. It should be noted that it will be apparent to those skilled in the art that several variations and modifications can be made without departing from the spirit of the application, which are all within the scope of the application. Accordingly, the scope of protection of the present application is to be determined by the appended claims.

Claims

1. A pedicle detection method, the method comprising:

acquiring the center point coordinates of the center point of the missed-detection pedicle according to the pedicle center curve equation and the center point coordinates of the center point of each pedicle region;

2. The method of claim 1, wherein the obtaining the center point coordinates of the missed pedicle center point based on the pedicle center curve equation and the center point coordinates of each of the pedicle region center points comprises:

according to a first preset direction, determining the central point coordinates of a first missed-detection pedicle central point according to the pedicle central curve equation and the central point coordinates of the central points of the pedicle areas;

according to a second preset direction, determining the center point coordinates of the center points of the second missed detection pedicles according to the pedicle center curve equation, the center point coordinates of the center points of the pedicles area and the center point coordinates of the center points of the first missed detection pedicles; any one of the first preset direction and the second preset direction is a direction from the head end to the tail end of the target spine, and the other preset direction is a direction from the tail end to the head end of the target spine;

3. The method of claim 2, wherein said determining, in accordance with a first predetermined direction, the center point coordinates of the first missed pedicle center point based on the pedicle center curve equation and the center point coordinates of the respective pedicle region center points, comprises:

acquiring first distance evaluation values of the first center point combinations according to center point coordinates of three adjacent center points in the first center point combinations; the first distance evaluation value represents the deviation degree of the distance between every two adjacent center points in each first center point combination;

if the first distance evaluation value is larger than a first preset threshold value, determining that a first missed detection pedicle central point exists in the current first central point combination, and determining the central point coordinate of the first missed detection pedicle central point according to the central point coordinate of the pedicle region central point in the current first central point combination and the pedicle central curve equation.

4. The method of claim 3, wherein said determining, in accordance with a second predetermined direction, the center point coordinates of the second missed pedicle center point based on the pedicle center curve equation, the center point coordinates of each of the pedicle region center points, and the center point coordinates of the first missed pedicle center point, comprises:

Acquiring a plurality of second center point combinations according to a second preset direction, the position relation of each pedicle region and the position relation of the center point of the first missed approach pedicle; each second center point combination comprises three adjacent center points in all pedicle region center points and all first missed detection pedicle center points;

acquiring second distance evaluation values of the second center point combinations according to the center point coordinates of three adjacent center points in the second center point combinations; the second distance evaluation value represents the deviation degree of the distance between every two adjacent center points in each second center point combination;

and if the second distance evaluation value is larger than a second preset threshold value, determining a second missed detection pedicle central point in the current second central point combination, and determining the central point coordinate of the second missed detection pedicle central point according to the central point coordinate of the pedicle region central point in the current second central point combination and the pedicle central curve equation.

5. The method of claim 4, wherein the obtaining the first distance evaluation value of each of the first center point combinations according to the center point coordinates of three adjacent center points in each of the first center point combinations comprises:

For each first center point combination, acquiring a first distance between the first two adjacent center points in the three adjacent center points and a second distance between the last two adjacent center points in the three adjacent center points according to coordinates of three adjacent center points in the first center point combination;

and determining a ratio between the first distance and the second distance as a first distance evaluation value of the first center point combination.

6. The method of claim 5, wherein determining that there is a missing pedicle center point in the current first center point combination if the first distance estimate is greater than a first preset threshold comprises:

if the first distance evaluation value is larger than the first preset threshold value, determining that a missing pedicle center point exists between the first two adjacent center points in the three adjacent center points of the current first center point combination.

7. The method of claim 6, wherein determining the center point coordinates of the first missed pedicle center point based on the center point coordinates of the pedicle region center points in the current first center point combination and the pedicle center curve equation comprises:

substituting an average value of first coordinates of the first two adjacent central points in the current first central point combination into the pedicle central curve equation to obtain second coordinates of the first missed-detection pedicle central points, and determining the central point coordinates of the first missed-detection pedicle central points through the first coordinates and the second coordinates of the first missed-detection pedicle central points;

one of the first coordinate and the second coordinate is an abscissa, the other coordinate is an ordinate, and the average value of the ordinate of the first two adjacent center points is determined as the ordinate of the first missed-detection pedicle center point;

substituting the average value of the ordinate of the first two adjacent central points into the pedicle central curve equation to obtain the abscissa of the first missed-detection pedicle central point, and determining the central point coordinate of the first missed-detection pedicle central point through the ordinate and the abscissa of the first missed-detection pedicle central point.

8. The method according to claim 1, wherein the method further comprises:

Acquiring a medical image of the target spine;

and segmenting the pedicle in the medical image of the target spine through a pedicle segmentation model to obtain the pedicle segmentation image.

9. A pedicle detection device, the device comprising:

the second coordinate determining module is used for obtaining the center point coordinates of the center point of the missed detection pedicle according to the pedicle center curve equation and the center point coordinates of the center point of each pedicle region;

and the detection result determining module is used for determining the pedicle detection result corresponding to the target spine according to the center point coordinates of the center points of the pedicle areas and the center point coordinates of the center points of the missed detection pedicles.

10. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor implements the steps of the method of any of claims 1-8 when the computer program is executed.