CN116363213A

CN116363213A - Method and device for detecting nasal root, storage medium and electronic equipment

Info

Publication number: CN116363213A
Application number: CN202310258047.7A
Authority: CN
Inventors: 马丽娟; 蔡巍; 张霞
Original assignee: Shenyang Neusoft Intelligent Medical Technology Research Institute Co Ltd
Current assignee: Shenyang Neusoft Intelligent Medical Technology Research Institute Co Ltd
Priority date: 2023-03-16
Filing date: 2023-03-16
Publication date: 2023-06-30

Abstract

The disclosure relates to a method and a device for detecting a nasion, a storage medium and electronic equipment. The method comprises the following steps: acquiring an NCCT image sequence of the head of an object to be detected; determining a skull region according to the NCCT image sequence; determining a nasal root candidate region from the skull region based on the positional characteristics of the nasal root; the location of the nasion is determined from the nasion candidate region according to the bone shape feature at the nasion. This way of determining the position of the root of the nose is more accurate and safer.

Description

Method and device for detecting nasal root, storage medium and electronic equipment

Technical Field

The disclosure relates to the technical field of image processing, in particular to a nose root detection method, a nose root detection device, a storage medium and electronic equipment.

Background

The Nasal Root (Nasal Root) is located at the upper end of the bridge of the nose between the two orbits. The point of the nose root (nose root point for short) is located at the upper part of the nose, the nose root point is the intersection point of the frontal nasal suture and the sagittal plane in the middle of the nose, and the nose root point is the point where the upper end of the nose bridge is connected with the frontal part. It follows that nasion is an important bony marker. Since the nasion is an important osseous marker of the face, and the plane in which it is located is the watershed of the face and brain, the nasion is an important reference for estimating the anatomical position of the head such as hairline, coronal suture, kocher's point, etc. in clinical brain surgery. For example, in clinical surgery, a tape is used to measure a distance in the overhead direction from the position of the root of the nose to obtain a coronal suture. In one embodiment, the intersection of the coronal and sagittal slots is determined by pushing the curved surface a distance of 13cm from the root of the nose to the top of the head.

In the related art, a medical staff determines the position of the nasal root by visual observation in combination with manual bone touch and marks the nasal root by using a marker pen according to own experience. However, when the position of the nasion is manually marked manually, the head of the patient needs to be moved, and moving the head of the patient is unsafe for the patient, secondary head injury is easy to cause, and the pulling force of moving the head of the patient can cause scalp displacement and further cause error in marking the position of the nasion.

Disclosure of Invention

In order to solve the problems in the related art, the present disclosure provides a method, an apparatus, a storage medium, and an electronic device for detecting a nasion.

To achieve the above object, a first aspect of embodiments of the present disclosure provides a method for detecting a nasion, the method including:

acquiring an NCCT image sequence of the head of an object to be detected;

determining a skull region according to the NCCT image sequence;

determining a nasal root candidate region from the skull region based on the positional characteristics of the nasal root;

the location of the nasion is determined from the nasion candidate region according to the bone shape feature at the nasion.

Optionally, the determining a nasion candidate region from the skull region based on the nasion-based location feature comprises:

Determining a median sagittal plane and a median coronal plane corresponding to the skull region;

determining an interested sagittal tomogram with the distance from the median sagittal plane smaller than a preset threshold value from the skull region to obtain an interested sagittal tomogram sequence;

and dividing the nose root candidate region from the bone region corresponding to the sagittal tomogram sequence of interest according to the median coronal plane.

Optionally, the determining the location of the nasion from the candidate region of the nasion according to the shape feature of the bone at the nasion comprises:

performing edge detection on the nasion candidate region to obtain a bone edge region, wherein the bone edge region represents a bone region in at least one bone edge sagittal tomogram;

constructing a filter according to the shape characteristics of the bone at the nasion;

for each bone edge sagittal tomogram, determining candidate nasion coordinates from the bone edge sagittal tomogram according to the filter;

and screening the positions of the nasal roots from all the candidate nasal root coordinates according to the shape characteristics of the bone at the nasal roots.

Optionally, the feature of the bone shape at the root of the nose includes a first feature representing that an included angle between a first line connecting the root of the nose and the forehead and a second line connecting the root of the nose and the back of the nose is an obtuse angle, and the constructing the filter according to the feature of the bone shape at the root of the nose includes:

Based on the first characteristic, constructing an isosceles obtuse triangle with an obtuse angle being a preset angle and a short side length being a preset length;

and performing digital conversion on the isosceles obtuse triangle to obtain the filter.

Optionally, the determining candidate nasion coordinates from the bone edge sagittal tomographic image according to the filter includes:

performing matched filtering processing on the bone edge sagittal tomographic image according to the filter to obtain candidate sagittal plane coordinates, wherein the candidate sagittal plane coordinates comprise sagittal axis coordinates and vertical axis coordinates;

and determining the three-dimensional candidate nasion coordinates according to the candidate sagittal plane coordinates and the coronal axis coordinates corresponding to the bone edge sagittal tomogram.

Optionally, the feature of the shape of the bone at the root of the nose includes a second feature representing that a point of the root of the nose is a concave point, and the step of screening the location of the root of the nose from all the candidate coordinates of the root of the nose according to the feature of the shape of the bone at the root of the nose includes:

clustering is carried out according to all the candidate nasion coordinates, so that target nasion coordinates are obtained;

for each target root coordinate, determining the target root coordinate as the position of the root if the target root coordinate is determined to conform to the second feature according to the position feature between the target root coordinate and the neighboring bone coordinates of the target root coordinate.

Optionally, the clustering processing is performed according to all the candidate nasion coordinates to obtain target nasion coordinates, including:

clustering all the candidate nasion coordinates to obtain a plurality of clusters, wherein each cluster comprises at least one candidate nasion coordinate;

sequencing the clusters from more to less according to the number of the candidate nasion coordinates in each cluster to obtain a cluster sequence;

determining the first N clusters in the cluster sequence as target clusters;

and determining the cluster center coordinates of each target cluster as the target nasion coordinates.

A second aspect of embodiments of the present disclosure provides a nasion detection device, the device comprising:

the acquisition module is used for acquiring the NCCT image sequence of the head of the object to be detected;

the first determining module is used for determining a skull region according to the NCCT image sequence;

a second determination module for determining a nasion candidate region from the skull region based on the position features of the nasion;

and a third determining module for determining the position of the nasion from the candidate region of the nasion according to the shape characteristics of the bone at the nasion.

Optionally, the second determining module includes:

a first determining submodule, configured to determine a median sagittal plane and a median coronal plane corresponding to the skull region;

A second determining sub-module, configured to determine a sagittal tomogram of interest having a distance from the median sagittal plane smaller than a preset threshold from the skull region, and obtain a sagittal tomogram sequence of interest;

and the segmentation submodule is used for segmenting the nasion candidate region from the bone region corresponding to the sagittal tomogram sequence of interest according to the median coronal plane.

Optionally, the third determining module includes:

the detection sub-module is used for carrying out edge detection on the nasion candidate region to obtain a bone edge region, and the bone edge region represents a bone region in at least one bone edge sagittal tomogram;

a construction submodule for constructing a filter according to the shape characteristics of the bone at the nasion;

a third determining sub-module, configured to determine, for each bone edge sagittal tomogram, candidate nasion coordinates from the bone edge sagittal tomogram according to the filter;

and the screening sub-module is used for screening the positions of the nasion from all the candidate nasion coordinates according to the shape characteristics of the bone at the nasion.

Optionally, the bone shape feature at the nasion includes a first feature characterizing an obtuse angle between a first line connecting the nasion and the forehead and a second line connecting the nasion and the dorsum, and the constructing sub-module includes:

The first construction submodule is used for constructing an isosceles obtuse triangle with an obtuse angle being a preset angle and a short side length being a preset length based on the first characteristic;

and the conversion sub-module is used for carrying out digital conversion on the isosceles obtuse triangle to obtain the filter.

Optionally, the third determining sub-module includes:

the filtering sub-module is used for carrying out matched filtering processing on the bone edge sagittal tomographic image according to the filter to obtain candidate sagittal plane coordinates, wherein the candidate sagittal plane coordinates comprise sagittal axis coordinates and vertical axis coordinates;

and the first execution submodule is used for determining the three-dimensional candidate nasion coordinate according to the candidate sagittal plane coordinate and the coronal axis coordinate corresponding to the bone edge sagittal tomogram.

Optionally, the bone shape feature at the nasion includes a second feature characterizing a point of location at the nasion as a pit, and the screening sub-module includes:

the clustering sub-module is used for carrying out clustering processing according to all the candidate nasion coordinates to obtain target nasion coordinates;

and the second execution submodule is used for determining the target nasion coordinate as the position of the nasion under the condition that the target nasion coordinate accords with the second characteristic according to the position characteristic between the target nasion coordinate and the neighborhood bone coordinate of the target nasion coordinate.

Optionally, the clustering sub-module is configured to:

clustering all the candidate nasion coordinates to obtain a plurality of clusters, wherein each cluster comprises at least one candidate nasion coordinate; sequencing the clusters from more to less according to the number of the candidate nasion coordinates in each cluster to obtain a cluster sequence; determining the first N clusters in the cluster sequence as target clusters; and determining the cluster center coordinates of each target cluster as the target nasion coordinates.

A third aspect of the disclosed embodiments provides a non-transitory computer-readable storage medium having stored thereon a computer program which, when executed by a processor, implements the steps of the method of root detection of the first aspect.

A fourth aspect of the embodiments of the present disclosure provides an electronic device, including:

a memory having a computer program stored thereon;

a processor for executing the computer program in the memory to implement the steps of the method for detecting a nasal root according to the first aspect.

By adopting the technical scheme, at least the following beneficial technical effects can be achieved:

the method comprises the steps of obtaining NCCT image sequences of the head of an object to be detected, and determining the skull region according to the NCCT image sequences. A root candidate region is determined from the skull region based on the location characteristics of the root, and the location of the root is determined from the root candidate region based on the bone shape characteristics at the root. This way of reconstructing the skull region from the NCCT image sequence of the head and determining the position of the nasion based on the position features of the nasion and the shape features of the bone at the nasion is more accurate and safer than the way of moving the patient's head first and then looking and manually touching the bone to determine the position of the nasion in the related art, since there is no need to move the patient's head and no need to rely on manual experience.

Additional features and advantages of the present disclosure will be set forth in the detailed description which follows.

Drawings

The accompanying drawings are included to provide a further understanding of the disclosure, and are incorporated in and constitute a part of this specification, illustrate the disclosure and together with the description serve to explain, but do not limit the disclosure. In the drawings:

fig. 1 is a human body coordinate system shown according to an exemplary embodiment of the present disclosure.

Fig. 2 is a flow chart illustrating a method of nasal root detection according to an exemplary embodiment of the present disclosure.

Fig. 3 is another human body coordinate system shown according to an exemplary embodiment of the present disclosure.

Fig. 4 is a schematic diagram of a XOY-planar brain center seam, according to an exemplary embodiment of the present disclosure.

Fig. 5 is a three-dimensional image corresponding to one candidate region, according to an exemplary embodiment of the present disclosure.

Fig. 6 is a segmentation schematic diagram illustrating one example embodiment of the present disclosure.

Fig. 7 is a first feature diagram illustrating an exemplary embodiment of the present disclosure.

Fig. 8 is a schematic view of an isosceles obtuse triangle, according to an exemplary embodiment of the present disclosure.

Fig. 9 is a schematic diagram illustrating a process of performing a matched filtering process on a bone edge sagittal tomogram according to a filter according to an exemplary embodiment of the present disclosure.

Fig. 10 is a schematic view of a head contour region, according to an exemplary embodiment of the present disclosure.

Fig. 11 is a binary image schematic of a head contour region, according to an exemplary embodiment of the present disclosure.

Fig. 12 is a block diagram illustrating a nasion detection device according to an exemplary embodiment of the present disclosure.

Fig. 13 is a block diagram of an electronic device, according to an exemplary embodiment of the present disclosure.

Detailed Description

Specific embodiments of the present disclosure are described in detail below with reference to the accompanying drawings. It should be understood that the detailed description and specific examples, while indicating and illustrating the disclosure, are not intended to limit the disclosure.

It should be noted that, all actions for acquiring signals, information or data in the present disclosure are performed under the condition of conforming to the corresponding data protection rule policy of the country of the location and obtaining the authorization given by the owner of the corresponding device.

In order to facilitate easier understanding of the technical solutions of the present disclosure by those skilled in the art, the following description will first explain basic cut planes of the human body in human anatomy related to the embodiments of the present disclosure.

Sagittal plane: the section along the front and back of the body perpendicular to the ground is called the sagittal plane, which divides the body into left and right parts. The sagittal plane along the body midline is called the median or median sagittal plane. Such as the sagittal plane shown in fig. 1.

Coronal plane: the section along the body perpendicular to the ground from side to side is called the coronal section, also called the frontal section. The coronal plane is used for dividing the human body into front and rear parts. Such as the (median) coronal plane shown in fig. 1.

Horizontal plane: the section perpendicular to the longitudinal axis of the human body and parallel to the ground is called a horizontal plane, and the horizontal plane divides the human body into an upper part and a lower part. Such as the horizontal plane shown in fig. 1.

The basic axes in human anatomy involved in the embodiments of the present disclosure will be described below with reference to fig. 1.

Coronary axis: the axis passing vertically through the sagittal plane in the left-right direction is referred to as the coronal or frontal axis, which may be characterized as the x-axis, such as the x-axis shown in fig. 1.

Sagittal axis: the axis passing perpendicularly through the coronal plane in the anterior-posterior direction is referred to as the sagittal axis and may be characterized as the y-axis, such as the y-axis shown in fig. 1.

Vertical axis: the axis passing vertically through the horizontal plane in the up-down direction is referred to as the vertical axis, which may be characterized as the z-axis, such as the z-axis shown in fig. 1.

The following describes the embodiments of the present disclosure in detail.

Fig. 2 is a flow chart illustrating a method of nasal root detection according to an exemplary embodiment of the present disclosure. As shown in fig. 2, the nasion detection method includes the following steps.

S21, acquiring an NCCT image sequence of the head of the object to be detected.

CT is an acronym for Computer Tomography, chinese translated into electronic computed tomography. CT flat scan (NCCT) refers to an imaging examination method applied to the field of acute cerebral apoplexy, and is generally used for examining whether patients have cerebral lesions such as super acute cerebral infarction lesions, other cerebral hemorrhage lesions, or hemorrhagic transformation lesions caused by arterial reperfusion injury. The NCCT image sequence in the present disclosure refers to an image sequence obtained by CT scanning the head of an object to be detected (such as a patient and a tester) in a CT flat scanning manner, where the image sequence includes a plurality of faults, and each fault is a two-dimensional medical image.

That is, the NCCT image sequence of the head of the object to be detected includes a plurality of slices, each slice being a two-dimensional image. Since the multiple slices combine to characterize a three-dimensional image of the head, the NCCT image sequence characterizes a three-dimensional image of the head of the object to be examined.

It should be noted that, in the present disclosure, the NCCT image sequence of the head of the object to be detected may be an image sequence obtained by performing CT scanning on the head of the object to be detected in a CT flat scanning manner at any time and under any scene. That is, without specially performing CT panning on the object to be detected in order to detect the position of the nasion, any NCCT image sequence in the medical record of the object to be detected can be used. This has no effect on the accuracy of using the nasion detection method of the present disclosure to determine the location of the nasion.

S22, determining the skull region according to the NCCT image sequence.

Wherein the skull region is the region occupied by the skull in three-dimensional space.

Since the NCCT image sequence is obtained by CT flat scanning of the head of the object to be detected, the NCCT image sequence comprises head imaging information of the object to be detected. And the head includes the skull and scalp, the skull region of the subject to be detected can be determined/reconstructed from the NCCT image sequence.

S23, determining a nose root candidate region from the skull region based on the position characteristics of the nose root.

In human biology the nasion has biological location features of the face located near the brain center seam, so in order to reduce data throughput, a nasion candidate region including the brain center seam can be determined from the skull region to more quickly determine the location of the nasion from the nasion candidate region.

S24, determining the position of the nasion from the candidate region of the nasion according to the shape characteristics of the bone at the nasion.

The position point of the nose root in human biology also has concave point characteristics relative to the neighborhood bone points, and the specific position of the nose root can be determined from the nose root candidate region according to the bone shape characteristics of the nose root.

By adopting the method, the NCCT image sequence of the head of the object to be detected is obtained, and the skull region is determined according to the NCCT image sequence. A root candidate region is determined from the skull region based on the location characteristics of the root, and the location of the root is determined from the root candidate region based on the bone shape characteristics at the root. This way of reconstructing the skull region from the NCCT image sequence of the head and determining the position of the nasion based on the position features of the nasion and the shape features of the bone at the nasion is more accurate and safer than the way of moving the patient's head first and then looking and manually touching the bone to determine the position of the nasion in the related art, since there is no need to move the patient's head and no need to rely on manual experience.

determining a median sagittal plane and a median coronal plane corresponding to the skull region; determining an interested sagittal tomogram with the distance from the median sagittal plane smaller than a preset threshold value from the skull region to obtain an interested sagittal tomogram sequence; and dividing the nose root candidate region from the bone region corresponding to the sagittal tomogram sequence of interest according to the median coronal plane.

The preset threshold value can be an empirical value of 1.5 cm, 2 cm and the like. The subsequent embodiments of the present disclosure are illustrated by taking the preset threshold of 1.5 cm as an example.

The standard coordinate system for medical imaging is defined as shown in fig. 3, wherein the x-axis represents the coronal axis, the y-axis represents the sagittal axis, and the z-axis represents the vertical axis.

For example, since in human biology, the nasion has a biological position feature of a face located near the brain center seam, in determining the nasion candidate region, the candidate region including the brain center seam may be determined first, and then the nasion candidate region including the face may be determined from the candidate regions.

In determining the candidate region including the brain center seam, it is necessary to first determine the median sagittal plane in which the brain center seam is located, for example, the YOZ sagittal plane in which the brain center seam is located in the XOY horizontal plane view shown in fig. 4. A sagittal tomogram of interest is then determined from the region of the skull at a distance of less than 1.5 cm from the median sagittal plane, such as the YOZ sagittal plane in the two white line range shown in fig. 4. All sagittal tomograms of interest constitute the candidate region described above, e.g. the candidate region corresponds to the three-dimensional image shown in fig. 5.

It should be noted here that the present disclosure is not limited to the number of slices of the sagittal tomogram of interest. By way of example, assuming that the spatial resolution of the coronal axis X is normalized to 1mm, the number of slices of the sagittal tomogram of interest in the 3cm range may be 30 slices.

After the candidate region including the brain center seam is determined, a nasion candidate region including the face needs to be further determined from among the candidate regions. For example, since the median coronal plane may divide the human body into front and rear parts, i.e., the median coronal plane may divide the skull region into front and rear parts, each sagittal tomographic image of interest may be segmented according to the median coronal plane, resulting in a nasion candidate region including facial bones. For example, the sagittal tomogram of interest shown in fig. 6 is segmented, and the nasion candidate region includes a bone region in the sagittal tomogram of interest on the left side in fig. 6.

performing edge detection on the nasion candidate region to obtain a bone edge region, wherein the bone edge region represents a bone region in at least one bone edge sagittal tomogram; constructing a filter according to the shape characteristics of the bone at the nasion; for each bone edge sagittal tomogram, determining candidate nasion coordinates from the bone edge sagittal tomogram according to the filter; and screening the positions of the nasal roots from all the candidate nasal root coordinates according to the shape characteristics of the bone at the nasal roots.

Since the root candidate region is segmented from the bone region corresponding to the sagittal tomogram sequence of interest according to the median coronal plane, edge detection is performed on the root candidate region, and the resulting bone edge region can be characterized as a bone region in the bone edge sagittal tomogram sequence corresponding to the number of images in the sagittal tomogram sequence of interest.

Illustratively, the filter is constructed from the bone shape features at the nasion. And traversing each bone edge sagittal tomographic image, and filtering candidate nasion coordinates from the bone edge sagittal tomographic images according to the filter, wherein the candidate nasion coordinates all accord with the bone shape characteristics at the nasion.

Since it is not discriminated which side is the inside side and which side is the outside side of the two sides of the bone edge in filtering the candidate nasion coordinates from the bone edge sagittal tomographic image according to the filter, the filtered candidate nasion coordinates may be pits or bumps on the bone edge. There is therefore a need to further screen out more accurate positions of the nasion from all candidate nasion coordinates based on the bone shape characteristics at the nasion.

Based on the first characteristic, constructing an isosceles obtuse triangle with an obtuse angle being a preset angle and a short side length being a preset length; and performing digital conversion on the isosceles obtuse triangle to obtain the filter.

For example, referring to fig. 7, according to the first feature that the included angle between the first line of the nose root and the forehead and the second line of the nose root and the nose back is an obtuse angle, and the statistical feature that the obtuse angle is about 120 ° to 130 °, an isosceles obtuse triangle with an obtuse angle being a preset angle such as 120 and a short side being a preset length such as 2 cm can be constructed, as shown in fig. 8 (p ₁ 、p ₂ 、p _m ). Further, the lumbar triangles are digitally converted, e.g., into a 12 x 12, or 9*9, or 9 x 12 digital matrix, to obtain a filter.

performing matched filtering processing on the bone edge sagittal tomographic image according to the filter to obtain candidate sagittal plane coordinates, wherein the candidate sagittal plane coordinates comprise sagittal axis coordinates and vertical axis coordinates; and determining the three-dimensional candidate nasion coordinates according to the candidate sagittal plane coordinates and the coronal axis coordinates corresponding to the bone edge sagittal tomogram.

For example, the bone edge sagittal tomographic image is subjected to matched filtering processing according to a filter to obtain candidate sagittal plane coordinates, wherein the candidate sagittal plane coordinates include sagittal axis coordinates y and vertical axis coordinates z. And determining three-dimensional candidate nasion coordinates (x, y, z) according to the candidate sagittal plane coordinates and the coronal axis coordinates x corresponding to the bone edge sagittal tomogram.

The following takes fig. 9 as an example to visually illustrate the process of performing matched filtering processing on a bone edge sagittal tomographic image according to a filter. Control p of isosceles obtuse triangle (i.e. visual representation of filter) in sliding filtering mode ₁ And p ₂ The points slide on the bone edge voxel (white area) locations in the bone edge sagittal tomogram shown in fig. 9. During sliding, p ₁ And p ₂ The points are always located on bone edge voxels. During the moving process, if the isosceles triangle is p _m And if the point is also positioned on the bone edge voxel, determining that the isosceles obtuse triangle at the moment is matched with the bone edge, and determining the coordinate of the Pm point in the bone edge sagittal tomogram at the moment as a candidate sagittal plane coordinate.

Clustering is carried out according to all the candidate nasion coordinates, so that target nasion coordinates are obtained; for each target root coordinate, determining the target root coordinate as the position of the root if the target root coordinate is determined to conform to the second feature according to the position feature between the target root coordinate and the neighboring bone coordinates of the target root coordinate.

The method comprises the following steps of clustering according to all candidate nasion coordinates to obtain target nasion coordinates: clustering all the candidate nasion coordinates to obtain a plurality of clusters, wherein each cluster comprises at least one candidate nasion coordinate; sequencing a plurality of clusters from more to less according to the number of candidate nasion coordinates in each cluster to obtain a cluster sequence; determining the first N clusters in the cluster sequence as target clusters; and determining the cluster center coordinates of each target cluster as target nasion coordinates.

The candidate nasion coordinate P obtained by sliding filtering is due to various error reasons such as acquisition precision and imaging errors of NCCT images, segmentation errors of bone edge sagittal tomograms, errors introduced by other factors and the like _x1,y1,z1 ,P _x2,y2,z2 ……P _xm,ym,zm M represents the number of candidate nasion coordinates. May fall into multiple categories, it is not possible to determine that the candidate nasion coordinate entities fall into several categories prior to the clustering process, and it is not determined where the candidate nasion coordinates are distributed. Therefore, in the embodiment of the disclosure, a DBSCAN clustering algorithm is adopted to perform clustering processing on candidate nasion coordinates, so as to obtain a plurality of classes (clusters). In the process of clustering by using the DBSCAN clustering algorithm, the input value of the DBSCAN clustering algorithm is clustered into a radius

m

Set to 5cm and the number of each category set to 2.

Further, according to the number of candidate nasion coordinates in each cluster, the clusters are ordered from more to less, and a cluster sequence is obtained. The first N clusters in the cluster sequence are determined as target clusters. And determining the cluster center coordinates of each target cluster as target nasion coordinates. It should be noted that, in addition to the above-described ranking manner based on the number of the target clusters from the plurality of clusters, other manners of selecting the top N target clusters including the largest number of candidate nasion coordinates from the plurality of clusters may be adopted, such as a ranking manner based on the number of the target clusters from the small to the large, such as a manner based on a bubbling algorithm, and the like. The present disclosure is not particularly limited thereto. The value of N can be 1, 2 or 3.

Further, for each target root coordinate, in the case where it is determined that the target root coordinate meets the second feature according to the feature of the position between the target root coordinate and the neighborhood bone coordinates of the target root coordinate, that is, in the case where the target root coordinate is the pit coordinate, the target root coordinate is determined as the position of the root.

Optionally, the determining the skull region according to the NCCT image sequence includes:

Preprocessing the NCCT image sequence to obtain a preprocessed NCCT image sequence; determining a head contour region according to the pre-processed NCCT image sequence; the skull region is determined from within the head contour region based on the CT value interval of the bony structure.

Wherein, the preprocessing of the NCCT image sequence comprises denoising and/or image correction processing of the NCCT image sequence.

Under the condition, in the CT (computed tomography) flat scanning process of a patient, an auxiliary device is often used for fixing the skull of the patient so as to avoid the problems of residual shadows and the like in the image caused by shaking of the head of the patient in the CT flat scanning process. Therefore, in addition to the head imaging information of the patient, the NCCT image sequence obtained by scanning may further include imaging information of an auxiliary device, where the imaging information of the auxiliary device is interference information, which may cause interference to subsequent image processing. Accordingly, in the embodiments of the present disclosure, preprocessing, such as denoising, is performed on the NCCT image sequence to remove noise interference, so as to obtain a preprocessed NCCT image sequence from which noise is removed. For example, for each image in the NCCT image sequence, imaging information of an auxiliary device such as a couch plate on the image is removed, resulting in a first NCCT image sequence after removing the imaging information of the couch plate.

In another case, during CT panning of a patient, the head imaging image of the patient in the NCCT image sequence obtained by scanning may be tilted, which may interfere with subsequent image processing if the head imaging image is tilted, because the patient's skull may be tilted without using an auxiliary device to fix the patient's skull or while using an auxiliary device to fix the patient's skull. Accordingly, in embodiments of the present disclosure, a preprocessing, such as image correction, is performed on the NCCT image sequence to correct the tilted head imaging image, resulting in a corrected preprocessed NCCT image sequence.

In addition, there may be noise interference caused by the auxiliary device and a problem that the head imaging image is tilted in the NCCT image sequence, so in some embodiments, the preprocessing the NCCT image sequence to obtain a preprocessed NCCT image sequence includes:

removing imaging information of an auxiliary device on each image in the NCCT image sequence to obtain a first NCCT image sequence; and carrying out image rigid registration processing on the first NCCT image sequence according to a standard NCCT image sequence to obtain the pre-processed NCCT image sequence after registration.

For example, for each image in the NCCT image sequence, imaging information of the auxiliary device on the image is removed, resulting in a first NCCT image sequence. Each image in the first NCCT image sequence is a clean image from which imaging information of the auxiliary device is removed. Further, since in the medical field, the two images of the image rigid registration may be medical images from different subjects. Therefore, the NCCT image sequence with no focus on the head, no obvious abnormality on the bone characteristics of the head and no angular offset on the head of the patient during CT flat scanning can be selected as the standard NCCT image sequence. And performing image rigid registration processing on the first NCCT image sequence according to the standard NCCT image sequence to obtain a registered preprocessing NCCT image sequence.

The implementation manner of obtaining the pre-processed NCCT image sequence after registration may be that a ITK (Insight Toolkit) three-dimensional rigid registration tool in the related technology is adopted to perform image rigid registration processing on the first NCCT image sequence according to the standard NCCT image sequence, so as to obtain the pre-processed NCCT image sequence after registration.

And the auxiliary devices may be bed boards, head bolsters, head holders, etc. An embodiment of removing the imaging information of the auxiliary device on the image may be to remove the imaging information of the auxiliary device based on morphological characteristics of the auxiliary device. For example, the HU value characterizing the voxel of the auxiliary device is set to 0. The embodiment of removing the imaging information of the auxiliary device on the image may be to manually remove the imaging information of the auxiliary device.

Optionally, the determining the head contour region according to the preprocessing NCCT image sequence includes:

for each pre-processed image in the pre-processed NCCT image sequence, determining a first target voxel in the pre-processed image, the CT value of which is greater than a first preset threshold; and determining the region formed by all the first target voxels in the pre-processed NCCT image sequence as the head contour region.

In one embodiment, if an ITK three-dimensional rigid registration tool in the related art is used to perform image rigid registration processing on the first NCCT image sequence according to the standard NCCT image sequence, a preprocessed NCCT image sequence after registration is obtained. Then, since the own algorithm principle of the ITK three-dimensional rigid registration tool normalizes the CT value (HU value) of the region voxels near the head to 0, whereas the HU value of air is about-1000, and the HU value of the head is greater than 0, the first preset threshold may be set to 0. And determining a first target voxel with a CT value greater than 0 in the preprocessed image for each preprocessed image in the sequence of preprocessed NCCT images. The region composed of all the first target voxels in the pre-processed NCCT image sequence is determined as the head contour region.

To facilitate a more visual understanding of how the present disclosure determines the head contour region, a person of ordinary skill in the art will now take the example of fig. 10 and 11 as an illustration. For each pre-processed image in the pre-processed NCCT image sequence shown in fig. 10, determining a first target voxel with a CT value greater than 0 in the pre-processed image, setting the HU value of the first target voxel to 1, and setting the remaining voxels to 0, to obtain a binary image as shown in fig. 11, wherein 1 represents white and 0 represents black in the image field. The white areas in each image in fig. 11 represent the head area. The three-dimensional image obtained by combining the head areas in all the images is the head outline area.

Optionally, the determining the skull region from within the head contour region based on the CT value interval of the bony structure includes:

determining a second target voxel with a CT value in the CT value interval from the head contour area; and determining the area formed by all the second target voxels as the skull area.

Since the CT value (HU value) of the bone structure is 150 to 1000, the CT value interval of the bone structure may be [150,1000 ].

For example, since the skull is located within the head contour region, all voxels characterizing the skull may be determined from within the head contour region. A voxel whose CT value is in the interval 150,1000 in the region of the head contour is determined as a second target voxel characterizing the skull. Since the second target voxels are voxels representing the skull, the three-dimensional region formed by all the second target voxels is the skull region.

The skull region may be represented as a white region in the three-dimensional binary image corresponding to the obtained NCCT binary image sequence after setting the second target voxel in the NCCT image sequence to 1 and the remaining voxels to 0 for the NCCT image sequence.

As shown in fig. 12, the nasion detection device 1200 includes:

an acquisition module 1201, configured to acquire an NCCT image sequence of a head of an object to be detected;

a first determining module 1202 for determining a skull region according to the NCCT image sequence;

a second determining module 1203 for determining a nasal root candidate region from the skull region based on the positional characteristics of the nasal root;

a third determination module 1204 is configured to determine a location of the root from the root candidate region based on the bone shape feature at the root.

With the apparatus 1200 described above, the skull region is determined from the NCCT image sequence by acquiring the NCCT image sequence of the head of the subject to be detected. A root candidate region is determined from the skull region based on the location characteristics of the root, and the location of the root is determined from the root candidate region based on the bone shape characteristics at the root. This way of reconstructing the skull region from the NCCT image sequence of the head and determining the position of the nasion based on the position features of the nasion and the shape features of the bone at the nasion is more accurate and safer than the way of moving the patient's head first and then looking and manually touching the bone to determine the position of the nasion in the related art, since there is no need to move the patient's head and no need to rely on manual experience.

Optionally, the second determining module 1203 includes:

Optionally, the third determining module 1204 includes:

Optionally, the third determining sub-module includes:

Optionally, the clustering sub-module is configured to:

The specific manner in which the various modules perform the operations in the apparatus of the above embodiments have been described in detail in connection with the embodiments of the method, and will not be described in detail herein.

Fig. 13 is a block diagram of an electronic device 700, according to an exemplary embodiment of the present disclosure. As shown in fig. 13, the electronic device 700 may include: a processor 701, a memory 702.

The processor 701 is configured to control the overall operation of the electronic device 700 to perform all or part of the steps in the method for detecting a nasion described above. The memory 702 is used to store various types of data to support operation at the electronic device 700, which may include, for example, instructions for any application or method operating on the electronic device 700, as well as application-related data, such as picture data.

In an exemplary embodiment, a computer readable storage medium is also provided, comprising program instructions which, when executed by a processor, implement the steps of the method of nasal root detection described above. For example, the computer readable storage medium may be the memory 702 including program instructions described above, which are executable by the processor 701 of the electronic device 700 to perform the method of root detection described above.

In another exemplary embodiment, a computer program product is also provided, comprising a computer program executable by a programmable apparatus, the computer program having code portions for performing the above-described root detection method when executed by the programmable apparatus.

The preferred embodiments of the present disclosure have been described in detail above with reference to the accompanying drawings, but the present disclosure is not limited to the specific details of the above embodiments, and various simple modifications may be made to the technical solutions of the present disclosure within the scope of the technical concept of the present disclosure, and all the simple modifications belong to the protection scope of the present disclosure.

In addition, the specific features described in the above embodiments may be combined in any suitable manner without contradiction. The various possible combinations are not described further in this disclosure in order to avoid unnecessary repetition.

Moreover, any combination between the various embodiments of the present disclosure is possible as long as it does not depart from the spirit of the present disclosure, which should also be construed as the disclosure of the present disclosure.

Claims

1. A method of nasion detection, the method comprising:

acquiring an NCCT image sequence of the head of an object to be detected;

determining a skull region according to the NCCT image sequence;

2. The method of claim 1, wherein the determining a nasion candidate region from the skull region based on the nasion-based location feature comprises:

3. The method of claim 1 or 2, wherein determining the location of the nasion from the nasion candidate region based on the bone shape feature at the nasion comprises:

4. A method according to claim 3, wherein the root canal bone shape feature comprises a first feature characterizing an obtuse angle between a first line of the root canal and the forehead and a second line of the root canal and the dorsum of the nose, the constructing a filter from the root canal bone shape feature comprising:

5. A method according to claim 3, wherein said determining candidate nasion coordinates from the bone edge sagittal tomogram based on said filter comprises:

6. A method according to claim 3, wherein the root-canal bone shape feature comprises a second feature that characterizes a point of a canal as a pit, and wherein the screening the root location from all the candidate root coordinates based on the root-canal bone shape feature comprises:

7. The method of claim 6, wherein the clustering according to all the candidate nasion coordinates to obtain target nasion coordinates comprises:

determining the first N clusters in the cluster sequence as target clusters;

8. A nasion detection device, said device comprising:

9. A non-transitory computer readable storage medium having stored thereon a computer program, characterized in that the program when executed by a processor realizes the steps of the method according to any of claims 1-7.

10. An electronic device, comprising:

a memory having a computer program stored thereon;

a processor for executing the computer program in the memory to implement the steps of the method of any one of claims 1-7.