CN119810167A - Method, device and storage medium for colorectal curved surface registration - Google Patents
Method, device and storage medium for colorectal curved surface registration Download PDFInfo
- Publication number
- CN119810167A CN119810167A CN202510295165.4A CN202510295165A CN119810167A CN 119810167 A CN119810167 A CN 119810167A CN 202510295165 A CN202510295165 A CN 202510295165A CN 119810167 A CN119810167 A CN 119810167A
- Authority
- CN
- China
- Prior art keywords
- curved surface
- dimensional curved
- intestinal
- registration
- prone
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Landscapes
- Image Processing (AREA)
- Apparatus For Radiation Diagnosis (AREA)
Abstract
本申请公开了一种用于结直肠曲面配准的方法、设备和存储介质。所述方法包括：采集结直肠的仰位三维曲面和俯位三维曲面，并分别将所述仰位三维曲面和所述俯位三维曲面参数化至对应的平面域；在对应的平面域内提取所述仰位三维曲面和所述俯位三维曲面各自的肠带特征点；基于所述各自的肠带特征点确定所述仰位三维曲面和所述俯位三维曲面对应的第一肠带线和第二肠带线；以及将所述第一肠带线和所述第二肠带线进行配准，以实现结直肠曲面配准。利用本申请的方案，可以提供精确、稳定的配准方式，对结直肠的仰位三维曲面和俯位三维曲面进行配准。
The present application discloses a method, device and storage medium for colorectal surface registration. The method includes: collecting the supine three-dimensional surface and prone three-dimensional surface of the colorectum, and parameterizing the supine three-dimensional surface and the prone three-dimensional surface to corresponding plane domains respectively; extracting the intestinal feature points of the supine three-dimensional surface and the prone three-dimensional surface in the corresponding plane domain; determining the first intestinal line and the second intestinal line corresponding to the supine three-dimensional surface and the prone three-dimensional surface based on the respective intestinal feature points; and registering the first intestinal line and the second intestinal line to achieve colorectal surface registration. Utilizing the scheme of the present application, an accurate and stable registration method can be provided to register the supine three-dimensional surface and the prone three-dimensional surface of the colorectum.
Description
Technical Field
The present application relates generally to the field of image processing technology. More particularly, the present application relates to a method, apparatus and computer storage medium for colorectal curved registration.
Background
Colorectal cancer is the second leading killer to the current involvement of cancer. The precursor of colorectal canceration is often polyp, and the polyp is discovered, intervened and excised as early as possible, so that the focus mortality rate can be greatly reduced. The traditional optical enteroscope can realize polyp observation and cutting, so that the optical enteroscope is widely applied. However, optical enteroscopes require general anesthesia, which is invasive and destructive. In addition, many patients often have undergone optimal dry-prediction when examined with an optical enteroscope, which is not suitable for large-scale screening.
Virtual enteroscopy solutions based on computer aided diagnosis (Computer Aided Diagnosis, "CAD") are currently the recommended method for large-scale screening, which is usually scanned with CT equipment and thus is not invasive. Through analysis and processing of three-dimensional colon curved surfaces acquired by CT equipment, colon polyps can be effectively found, and early warning is carried out in advance. For a virtual enteroscopy system, surface registration can be compared conveniently for unclear regions, or to confirm polyp location. However, in scanning with a CT apparatus, although the inner wall of the intestine is cleaned, there is still cleaning residue, and registration is generally required by scanning the curved surface of the colon at a plurality of angles (e.g., prone position, supine position, etc.) by the CT apparatus to confirm the polyp position. Thus, how to register the multi-angle curved surface of the colon is a technical problem to be solved.
In view of this, it is desirable to provide a solution for colorectal surface registration that simplifies the computational complexity of three-dimensional space by parameterizing the supine and prone three-dimensional surfaces into corresponding two-dimensional planar domains, and then performing colorectal surface registration in the two-dimensional planar domains by important and easily identifiable structural intestinal striplines to provide an accurate, stable registration approach for registering the supine and prone three-dimensional surfaces of the colorectal.
Disclosure of Invention
To address at least one or more of the technical problems mentioned above, the present application proposes, in various aspects, a solution for colorectal curved registration.
In a first aspect, the application provides a method for colorectal curved surface registration, comprising the steps of collecting an elevation three-dimensional curved surface and a depression three-dimensional curved surface of a colorectal, respectively parameterizing the elevation three-dimensional curved surface and the depression three-dimensional curved surface to corresponding plane domains, extracting respective intestinal band characteristic points of the elevation three-dimensional curved surface and the depression three-dimensional curved surface in the corresponding plane domains, determining a first intestinal band line and a second intestinal band line corresponding to the elevation three-dimensional curved surface and the depression three-dimensional curved surface based on the respective intestinal band characteristic points, and registering the first intestinal band line and the second intestinal band line to realize colorectal curved surface registration.
In one embodiment, extracting respective intestinal characteristic points of the elevation three-dimensional curved surface and the depression three-dimensional curved surface in a corresponding plane domain comprises constructing respective height maps of the elevation three-dimensional curved surface and the depression three-dimensional curved surface in the corresponding plane domain, wherein the height maps represent distances from intestinal wall points on the elevation three-dimensional curved surface or the depression three-dimensional curved surface to central axes of curved surfaces, extracting respective characteristic lines from the height maps of the elevation three-dimensional curved surface and the depression three-dimensional curved surface, and extracting respective intestinal characteristic points of the elevation three-dimensional curved surface and the depression three-dimensional curved surface based on the respective characteristic lines.
In another embodiment, extracting respective feature lines from the height maps of the elevation three-dimensional curved surface and the depression three-dimensional curved surface respectively comprises accumulating pixel values of each line in the height maps of the elevation three-dimensional curved surface and the depression three-dimensional curved surface respectively, calculating a local maximum value in the accumulated result in the height maps of the elevation three-dimensional curved surface and the depression three-dimensional curved surface respectively, and extracting respective feature lines of the elevation three-dimensional curved surface and the depression three-dimensional curved surface based on points of the local maximum value.
In yet another embodiment, wherein extracting the respective intestinal characteristic points of the superior three-dimensional curved surface and the prone three-dimensional curved surface based on the characteristic line includes extracting a target point from the characteristic line having a local minimum value less than a preset threshold value, and determining a point in the target point at which a rotation angle with respect to a central axis of the curved surface satisfies a target central axis rotation amount as the superior three-dimensional curved surface and the prone three-dimensional curved surface to extract the respective intestinal characteristic points.
In yet another embodiment, wherein determining the first and second intestinal lines corresponding to the superior and prone three-dimensional curved surfaces based on the respective intestinal band feature points includes connecting a curved shortest distance between the respective intestinal band feature points and smoothing to determine the first and second intestinal lines corresponding to the superior and prone three-dimensional curved surfaces.
In yet another embodiment, wherein the first and second intestinal lines each comprise three and registering the first and second intestinal lines to achieve colorectal surface registration comprises calculating a target feature value between any one of the first and second intestinal lines and selecting a pair of intestinal lines corresponding to a minimum target feature value to achieve colorectal surface registration.
In yet another embodiment, wherein the target feature value comprises at least a curvature feature value.
In yet another embodiment, the method further comprises constructing a differential homoembryo function and optimizing a registration result of colorectal curved registration based on the differential homoembryo function.
In a second aspect, the present application provides an apparatus for colorectal curved registration, comprising a processor, and a memory having stored thereon computer instructions for colorectal curved registration, which when executed by the processor, cause the implementation of the plurality of embodiments of the aforementioned first aspect.
In a third aspect, the application provides a computer-readable storage medium having stored thereon computer program instructions for colorectal curved registration, which, when executed by one or more processors, cause implementation of the various embodiments of the aforementioned first aspect.
With the scheme for colorectal surface registration provided above, the embodiment of the application performs colorectal surface registration by parameterizing the elevation three-dimensional surface and the depression three-dimensional surface to corresponding planar domains, extracting intestinal characteristic points in the planar domains, and determining corresponding intestinal strip lines in the elevation three-dimensional surface and the depression three-dimensional surface. Among other things, the intestinal band line identifies one of the important structures of the colon. Therefore, the embodiment of the application enables the curved surface registration to have simplicity and stability by mapping the three-dimensional curved surface to the two-dimensional plane and performing colorectal curved surface registration through the important identification structure (intestinal band line).
Drawings
The above, as well as additional purposes, features, and advantages of exemplary embodiments of the present application will become readily apparent from the following detailed description when read in conjunction with the accompanying drawings. In the drawings, embodiments of the application are illustrated by way of example and not by way of limitation, and like reference numerals refer to similar or corresponding parts and in which:
FIG. 1 is an exemplary flow diagram illustrating a method for colorectal curved registration according to an embodiment of the application;
FIG. 2 is an exemplary schematic diagram illustrating an elevation three-dimensional curved surface and a depression three-dimensional curved surface of a colorectal according to an embodiment of the present application;
FIG. 3 is an exemplary schematic diagram illustrating planar fields corresponding to an elevation three-dimensional curved surface and a depression three-dimensional curved surface in accordance with an embodiment of the present application;
FIG. 4 is an exemplary coordinate diagram illustrating the registration of the respective central axes of an elevation three-dimensional curved surface and a depression three-dimensional curved surface before and after registration according to an embodiment of the present application;
FIG. 5 is a schematic image showing the three-dimensional surface in the up position and the three-dimensional surface in the down position after registration of the respective central axes;
FIG. 6 is an exemplary schematic diagram illustrating a filtered central axis according to an embodiment of the present application;
FIG. 7 is an exemplary schematic diagram illustrating a height map according to an embodiment of the present application;
FIG. 8 is an exemplary diagram illustrating extracted feature lines according to an embodiment of the present application;
FIG. 9 is an exemplary schematic diagram illustrating intestinal band feature points of a supine three-dimensional curved surface and corresponding three intestinal band lines in accordance with an embodiment of the present application;
FIG. 10 is a graph showing the final intestinal band corresponding to an elevation three-dimensional curved surface in accordance with an embodiment of the present application;
FIG. 11 is an exemplary schematic diagram illustrating a supine three-dimensional curved surface and a prone three-dimensional curved surface after intestinal band registration in accordance with an embodiment of the present application;
FIG. 12 is an exemplary flow diagram illustrating an ensemble for colorectal curved registration according to an embodiment of the present application;
Fig. 13 is an exemplary block diagram illustrating an apparatus for colorectal curved registration according to an embodiment of the present application.
Detailed Description
The following description of the embodiments of the present application will be made clearly and fully with reference to the accompanying drawings, in which it is evident that the embodiments described are some, but not all embodiments of the application. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to be within the scope of the application.
It should be understood that the terms "comprises" and "comprising," when used in this specification and in the claims, specify the presence of stated features, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and/or groups thereof.
It is also to be understood that the terminology used in the description of the application herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the application. As used in the specification and claims, the singular forms "a", "an" and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It should be further understood that the term "and/or" as used in the present specification and claims refers to any and all possible combinations of one or more of the associated listed items, and includes such combinations.
As used in this specification and the claims, the term "if" may be interpreted as "when..once" or "in response to a determination" or "in response to detection" depending on the context. Similarly, the phrase "if a determination" or "if a [ described condition or event ] is detected" may be interpreted in the context of meaning "upon determination" or "in response to determination" or "upon detection of a [ described condition or event ]" or "in response to detection of a [ described condition or event ]".
Specific embodiments of the present application are described in detail below with reference to the accompanying drawings.
Fig. 1 is an exemplary flow diagram illustrating a method 100 for colorectal curved registration according to an embodiment of the application. As shown in fig. 1, at step S101, an elevation three-dimensional curved surface and a depression three-dimensional curved surface of the colorectal are acquired, and the elevation three-dimensional curved surface and the depression three-dimensional curved surface are parameterized to corresponding planar domains, respectively. In one implementation scenario, the elevation three-dimensional curved surface and the depression three-dimensional curved surface of the colorectal may be acquired by using, for example, a CT scanning apparatus, where the elevation three-dimensional curved surface is a colorectal three-dimensional curved surface scanned in an elevation posture of the subject, and the depression three-dimensional curved surface is a colorectal three-dimensional curved surface scanned in a prone posture of the subject. In some embodiments, the supine three-dimensional curved surface and the prone three-dimensional curved surface of the colorectal can be obtained by performing three-dimensional reconstruction on the supine and prone CT data acquired by the CT scanning device. The three-dimensional reconstruction algorithm may include a multi-view geometry-based method, a depth learning-based method, and the like.
Based on the acquired supine three-dimensional curved surface and prone three-dimensional curved surface of the colorectal, the three-dimensional curved surfaces are firstly parameterized to corresponding plane domains respectively. It is understood that parameterization is the process of expanding a three-dimensional surface into a two-dimensional planar domain, which includes, but is not limited to, rectangular domains, disk domains, and the like. Preferably, the embodiment of the application parameterizes the elevation three-dimensional curved surface and the depression three-dimensional curved surface to rectangular areas respectively. In some embodiments, the parameterization may be performed using methods such as conformal mapping or conformal mapping. For example, taking conformal mapping as an example, by making a hole in the prone or supine curved surface, respectively, in the entire intestine at the anus and cecum, then calculating a cut (or closed curve), then cutting the intestinal wall along the cut (or closed curve), and spreading out to a rectangular area by conformal mapping. Based on this, by parameterizing the three-dimensional surface to the planar domain, the subsequent calculations can be greatly simplified.
Next, at step S102, intestinal characteristic points of each of the upward three-dimensional curved surface and the downward three-dimensional curved surface are extracted in the corresponding planar domain. In one embodiment, first, a height map corresponding to each of the elevation three-dimensional curved surface and the depression three-dimensional curved surface is constructed in a corresponding planar domain. Wherein the height chart represents the distance of the intestinal wall point on the elevation three-dimensional curved surface or the depression three-dimensional curved surface from the central axis of the curved surface. Then, respective feature lines are extracted from the respective height maps of the upward three-dimensional curved surface and the downward three-dimensional curved surface to extract respective intestinal characteristic points based on the respective feature lines. That is, feature lines are extracted from the height maps corresponding to the upward three-dimensional curved surface and the downward three-dimensional curved surface, and feature points of the intestinal tract are extracted.
In one implementation scenario, when constructing the height map, the respective central axes of the elevation three-dimensional curved surface and the depression three-dimensional curved surface are determined first. Specifically, multiple groups of sampling point sets can be collected along the longitudinal direction of the rectangular domain corresponding to the upward three-dimensional curved surface and the downward three-dimensional curved surface, and each group of sampling point set comprises multiple sampling points. And calculating the coordinate mean value of each group of multiple sampling points, wherein the multiple groups of sampling point sets correspond to the multiple coordinate mean value points, and restoring the multiple coordinate mean value points to the upward three-dimensional curved surface or the downward three-dimensional curved surface for connection, so that the respective central axes of the upward three-dimensional curved surface and the downward three-dimensional curved surface can be obtained. As an example, it is assumed that 1000 sets of sampling points are acquired in the longitudinal direction, i.e. 1000 points are included in the corresponding lateral direction. Each set of sampling points contains 100 sampling points, i.e. each set of points contains 100 points in the longitudinal direction. The central axes of the upward three-dimensional curved surface and the downward three-dimensional curved surface can be obtained by averaging 100 longitudinal points of each group of point sets, 1000 groups of point sets correspond to 1000 coordinate mean points, and then restoring the 1000 coordinate mean points to the upward three-dimensional curved surface or the downward three-dimensional curved surface for connection.
In some embodiments, in order to improve the subsequent calculation accuracy, the embodiment of the present application further performs central axis registration on respective central axes of the elevation three-dimensional curved surface and the depression three-dimensional curved surface. The aforementioned registration of the central axes may be characterized by their length, for example, using dynamic time warping (DYNAMIC TIME WARPING, "DTW").
After the registration of the central axes is completed, the primary registration of the intestines is completed longitudinally (namely along the direction of the central axes of the intestines), but the two intestines are rotated. Therefore, the embodiment of the application performs registration on the intestinal band line to finish local rotation by extracting the intestinal band line of the colorectal so as to improve the registration accuracy. In some embodiments, the present application implements a filtering operation on the registered central axes as well. In one implementation scenario, the registered central axes are filtered by optimizing the rotation of the central axes (i.e., the rotation of the coordinate system about the tangent vector) until it meets the target rotation. Specifically, there is a movable frame coordinate system (i.e., a coordinate system with a point on the central axis as an origin) on the central axis, and the rotation amount of the coordinate system can be obtained by integration. As an example, a discrete method may be employed to calculate a coordinate system for a point on the central axis, and calculate the rotation angles of the adjacent two coordinate systems to determine the rotation amount of the central axis. Since the initial central axis is relatively curved, its initial rotation must be greater than the target rotation (e.g., 25, i.e., 4 rotations). In this case, the filtering is repeated on the central axis until the rotation amount is lower than the target rotation amount 25.
It should be understood that the above-mentioned extraction of the central axis, the registration of the central axis and the filtering of the central axis are all operations performed on the upward three-dimensional curved surface and the downward three-dimensional curved surface respectively, so as to obtain the respective filtered central axes of the upward three-dimensional curved surface and the downward three-dimensional curved surface. For the central axis after filtering, a height map corresponding to each of the upward three-dimensional curved surface and the downward three-dimensional curved surface can be constructed in a corresponding plane. In one implementation scenario, the distance from the intestinal wall point to the central axis is calculated by finding the point corresponding to the closest central axis to the point on all intestinal walls on the supine three-dimensional curved surface and the prone three-dimensional curved surface, respectively. In this scenario, a distance, i.e. a function of the distance of the point from the nearest central axis, can be constructed in the parameter space. Further, the distance function is subjected to, for example, laplace filtering, and the difference in horizontal and vertical directions in the distance function is emphasized by, for example, a gabor filter, so as to obtain respective height maps of the elevation three-dimensional curved surface and the depression three-dimensional curved surface, denoted as h (x, y), where x represents coordinates of pixels of the horizontal sequence and y represents vertical pixel values (i.e., height values). In some embodiments, the filter weight coefficients of the laplace filter may be calculated based on a generalized cross-validation method.
Next, respective feature lines are extracted from the height maps corresponding to the upward three-dimensional curved surface and the downward three-dimensional curved surface, respectively. In one implementation scenario, the local maximum value in the accumulation result in the height map corresponding to each of the three-dimensional curved surface and the three-dimensional curved surface is calculated by accumulating the pixel values of each row in the height map corresponding to each of the three-dimensional curved surface and the three-dimensional curved surface, so as to extract the characteristic row of each of the three-dimensional curved surface and the three-dimensional curved surface based on the point of the local maximum value. The influence of noise can be removed by accumulating pixel values of each row, and the accuracy of subsequent calculation is ensured. After accumulation, one-dimensional data can be obtained, which is recorded as. Since there are not enteric feature points in each row of pixels, embodiments of the present application extract more significant rows as feature rows by extracting local maxima in the accumulated results (i.e., considering the maxima points for each row and sequence of pixels).
After the feature lines are obtained, extracting the upward three-dimensional curved surface and the downward three-dimensional curved surface based on the respective feature lines, and extracting respective intestinal characteristic points. In one implementation scenario, a target point with a local minimum value smaller than a preset threshold value is extracted from a feature line, and respective intestinal band feature points are extracted by determining points in the target point, which satisfy the rotation amount of the target axis relative to the rotation angle of the central axis of the curved surface, as a upward three-dimensional curved surface and a downward three-dimensional curved surface. In some embodiments, the aforementioned preset threshold may be, for example, 20% of the value of x. Specifically, the luminance h of the feature line is set as a y value, the sequence pixels of the feature line are set as x values, the relative minimum value x (i.e., local minimum value x) of h (x) is obtained from the information of h (x), and the minimum value is ensured to be 20% smaller than the x value of the surrounding points, so as to extract the target point.
It will be appreciated that the colonic bands are formed by thickening of the longitudinal muscles of the intestinal wall, and comprise three bands, the omentum band, the tie-film band and the independent band, whereby after dividing the colon into a plurality of turns, three target points (i.e. three minutiae points) are included on each turn (i.e. longitudinally). After three minimum points of the first circle are selected, three connection modes exist according to the number of the characteristic points. For example, the target point 1 of the first circle may be connected with the target point 1,2 or 3 of the second circle. In this scenario, respective intestinal characteristic points are extracted by determining, as the upward three-dimensional curved surface and the downward three-dimensional curved surface, a point in the target point at which the rotation angle with respect to the central axis of the curved surface satisfies the target central axis rotation amount (for example, 25).
Based on the above-mentioned extracted intestinal band feature points, at step S103, first and second intestinal band lines corresponding to the lying three-dimensional curved surface and the prone three-dimensional curved surface are determined based on the respective intestinal band feature points. In one implementation scenario, a first intestinal band line and a second intestinal band line corresponding to the upward three-dimensional curved surface and the downward three-dimensional curved surface are determined by connecting the shortest distances of the curved surfaces between the respective intestinal band feature points and performing smoothing. Specifically, the shortest distance of the curved surfaces between the respective intestinal characteristic points can be connected by, for example, a geodesic algorithm and smoothed.
Further, at step S104, the first intestinal band line and the second intestinal band line are registered to achieve colorectal curved registration. In one implementation scenario, a target feature value between any one of the first intestinal strip lines and any one of the second intestinal strip lines is calculated, and then a pair of intestinal strip lines corresponding to the minimum target feature value is selected for registration, so that colorectal curved surface registration is realized. From the foregoing, the colonic bands comprise three bands of omentum, tie film and independent bands, whereby the first and second intestinal bands each comprise three, there are three matching methods for which there are three intestinal bands to be matched. By way of example, assume that the first gut-line comprises gut-line 1,2,3 and the second gut-line comprises gut-line 1,2,3, which exists in three schemes 1-1, 2-2, 3-3;1-2, 2-3, 3-1 and 1-3, 2-1, 3-2.
In some embodiments, the aforementioned target feature values may include, but are not limited to, curvature feature values, length feature values, and the like. It will be appreciated that if there is correspondence between the two intestinal bands, the bends are closer. Thus, the similarity between any intestinal bands is calculated based on the target characteristic value, and the total energy between registration is calculated, so that a scheme with the minimum total energy of registration is selected as a final registration result. For example, in one exemplary scenario, characterized by the target feature value being a curvature feature value, curvature feature value differences between arbitrary intestinal lines are calculated, respectively, and a set of intestinal lines with the smallest sum of curvature feature value differences is taken as the registration result. For example, in the above three matching schemes (1-1, 2-2, 3-3;1-2, 2-3, 3-1 and 1-3, 2-1, 3-2), the difference in curvature characteristic values between the intestinal bands in the different matching schemes is calculated, respectively, assuming that the sum of the difference in curvature characteristic values in the 1-1, 2-2, 3-3 schemes is minimum, the 1-1, 2-2, 3-3 schemes are taken as the matching result. Similarly, if the sum of curvature characteristic value differences of the 1-2, 2-3, 3-1 schemes is minimum, the 1-2, 2-3, 3-1 schemes are taken as matching results.
As can be seen from the above description, the embodiment of the present application performs colorectal surface registration by parameterizing the elevation three-dimensional curved surface and the depression three-dimensional curved surface to corresponding planar domains, extracting intestinal band characteristic points in the planar domains, and determining corresponding intestinal band lines in the elevation three-dimensional curved surface and the depression three-dimensional curved surface. Among other things, the intestinal band line identifies one of the important structures of the colon. Therefore, the embodiment of the application enables the curved surface registration to have simplicity, stability and accuracy by mapping the three-dimensional curved surface to the two-dimensional plane and performing colorectal curved surface registration through the important identification structure (intestinal band line).
Fig. 2 is an exemplary schematic diagram illustrating an elevation three-dimensional curved surface and a depression three-dimensional curved surface of a colorectal according to an embodiment of the present application. Fig. 2 (a) shows a three-dimensional curved surface of the colorectal in a upward position, and fig. 2 (b) shows a three-dimensional curved surface of the colorectal in a downward position. In some embodiments, the supine three-dimensional curved surface and the prone three-dimensional curved surface of the colorectal can be obtained by performing three-dimensional reconstruction on the supine and prone CT data acquired by the CT scanning device.
Fig. 3 is an exemplary schematic diagram showing planar domains corresponding to an elevation three-dimensional curved surface and a depression three-dimensional curved surface according to an embodiment of the present application. Fig. 3 (a) shows a planar area corresponding to the upward three-dimensional curved surface, and fig. 3 (b) shows a planar area corresponding to the downward three-dimensional curved surface. In some embodiments, the planar surface may be parameterized to a corresponding planar domain using a method such as conformal mapping or conformal mapping. Preferably, the planar domain is a rectangular area. By parameterizing the three-dimensional surface to the planar domain, subsequent computations can be greatly simplified.
After the upward three-dimensional curved surface and the downward three-dimensional curved surface are parameterized to the corresponding plane domains, the respective central axes can be extracted first. In some embodiments, multiple groups of sampling point sets including multiple sampling points can be collected along the longitudinal direction of the rectangular domain corresponding to the upward three-dimensional curved surface and the downward three-dimensional curved surface, and the coordinate mean points corresponding to the sampling point sets are restored to the upward three-dimensional curved surface or the downward three-dimensional curved surface for connection, so that the respective central axes of the upward three-dimensional curved surface and the downward three-dimensional curved surface can be obtained. Based on the extracted central axis, the extracted central axis can be registered by adopting a DTW method, for example, so as to improve the subsequent calculation accuracy.
Fig. 4 is an exemplary coordinate diagram illustrating the registration of respective central axes of an elevation three-dimensional curved surface and a depression three-dimensional curved surface before and after according to an embodiment of the present application. As shown in fig. 4, the upper part shows a schematic coordinate diagram before the registration of the respective central axes of the three-dimensional curved surface in the upward position and the three-dimensional curved surface in the downward position, and the lower part in fig. 4 shows a schematic coordinate diagram after the registration of the respective central axes of the three-dimensional curved surface in the upward position and the three-dimensional curved surface in the downward position. Wherein the abscissa represents the pixel coordinates of the sequence, the ordinate represents the height information, and the needled position in the figure represents the intestinal band. In the figure, A and B respectively correspond to the coordinate information curves of the respective central axes of the upward three-dimensional curved surface and the downward three-dimensional curved surface.
Fig. 5 is a schematic image showing the respective central axes of the elevation three-dimensional curved surface and the depression three-dimensional curved surface after registration. As shown in fig. 5, where L1 and L2 correspond to the respective central axes of the supine-position three-dimensional curved surface and the prone-position three-dimensional curved surface, respectively. In some embodiments, the registered central axes may also be subjected to a filtering operation such that partial registration of the rotated (e.g., as shown at the circles in fig. 5) portions of the intestines is accomplished.
Fig. 6 is an exemplary schematic diagram illustrating a filtered central axis according to an embodiment of the present application. A schematic view of the filtered central axis is shown in fig. 6. In some embodiments, the coordinate systems may be calculated for points on the central axis using, for example, a discrete method, and the rotation angles of adjacent two coordinate systems may be calculated to determine the amount of rotation of the central axis, and the central axis may be repeatedly filtered until the amount of rotation is below a target rotation (e.g., 25), resulting in a filtered central axis. Based on the respective filtered central axes of the upward three-dimensional curved surface and the downward three-dimensional curved surface, the respective corresponding height maps of the upward three-dimensional curved surface and the downward three-dimensional curved surface can be constructed by calculating the distance between the intestinal wall point and the central axis.
Fig. 7 is an exemplary schematic diagram illustrating a height map according to an embodiment of the present application. Fig. 7 is a partial height view corresponding to the elevation three-dimensional curved surface. As previously described, the distance function of the point to the point of the nearest central axis is constructed by finding the point corresponding to the nearest central axis to the point on all the intestinal walls on the supine three-dimensional curved surface. The height map is then obtained by, for example, subjecting the distance function to, for example, laplace filtering, and applying, for example, gabor filters to emphasize the differences in the distance function in the horizontal and vertical directions. Similarly, a height map corresponding to the depression-position three-dimensional curved surface can be obtained. Next, respective feature lines are extracted from the height maps corresponding to the upward three-dimensional curved surface and the downward three-dimensional curved surface, respectively. Specifically, each row of the height map is accumulated to obtain one-dimensional dataThe feature line is obtained by extracting the local maximum therein.
Fig. 8 is an exemplary schematic diagram illustrating extraction of feature rows according to an embodiment of the present application. The red circle as shown in fig. 8 is at a local maximum, which corresponds to a feature row. Wherein the abscissa represents a sequence of pixel coordinates and the ordinate represents height information. After the feature lines are obtained, extracting the upward three-dimensional curved surface and the downward three-dimensional curved surface based on the respective feature lines, and extracting respective intestinal characteristic points. In one implementation scenario, first, points with local minima smaller than 20% of the surroundings thereof are extracted from the feature line as target points, and the three-dimensional curved surface in the target points and the three-dimensional curved surface in the depression position are extracted with the rotation angle with respect to the central axis of the curved surface satisfying the determination that the rotation amount is 25, for example. Further, connecting the characteristic points of the intestinal band can obtain an intestinal band line. The upward three-dimensional curved surface and the downward three-dimensional curved surface respectively correspond to three intestinal strip lines.
Fig. 9 is an exemplary schematic diagram showing intestinal band feature points of a supine three-dimensional curved surface and corresponding three intestinal band lines according to an embodiment of the present application. Small circles as shown in fig. 9 represent intestinal band feature points, and intestinal band lines including three intestinal band lines of omentum band, tie band, and independent band (for example, shown by red, green, and blue three lines in the figure) can be obtained by connecting the intestinal band feature points. In some embodiments, the present application further provides for determining final intestinal tracts corresponding to the supine three-dimensional curved surface and the prone three-dimensional curved surface by connecting the shortest distances of the curved surfaces between the respective intestinal tracts feature points and smoothing the curved surfaces, as shown in fig. 10.
Fig. 10 is a diagram showing a final intestinal band corresponding to a supine three-dimensional curved surface according to an embodiment of the present application. The final intestinal band corresponding to the supine three-dimensional curved surface (i.e., the first intestinal band) is shown in fig. 10. Similarly, a final intestinal band corresponding to the prone three-dimensional curved surface (i.e., a second intestinal band) may be obtained. Colorectal curved registration is achieved by registering the first intestinal band line and the second intestinal band line. In one implementation scenario, a set of intestinal lines with the smallest sum of curvature characteristic value differences is used as a registration result by respectively calculating curvature characteristic value differences between any intestinal lines.
Fig. 11 is an exemplary schematic diagram illustrating a supine three-dimensional curved surface and a prone three-dimensional curved surface after intestinal band registration according to an embodiment of the present application. Fig. 11 (a) shows a three-dimensional curved surface in a upward position after the registration of intestinal lines, and fig. 11 (b) shows a three-dimensional curved surface in a downward position after the registration of intestinal lines. In some embodiments, the registration results of colorectal curved registration may also be optimized by constructing a differential homoembryo function to be based on the differential homoembryo function. It is understood that a differential homoembryo function means that the function satisfies a one-to-one mapping (i.e., satisfies both single and full shots) and is continuous, infinitely differentiable.
In one implementation scenario, one intestinal track (e.g., one intestinal track in a planar three-dimensional surface) may be selected as the reference information and noted asThe remaining intestinal band line information (e.g. intestinal band line in the prone three-dimensional curved surface) is recorded asThe registration result is recorded as. In this scenario, an objective registration function may be obtained=WhereinRepresenting the gradient of the reference information. Next, a registration function is updated based on the target registration function:
(1)
Wherein, The weight is represented by a weight that,The reference information is represented by a reference number,Representing the remaining intestinal band line information,Representing constant coefficients. In this scenario, by registering the targets with a functionSubstituting into the above formula (1) can obtain an updated registration function, e.g. noted as. Further, an initial Beltrami coefficient is calculated based on the updated registration function. In one implementation, the initial Beltrami coefficients may be calculated based on the interpolated registration function by linearly interpolating the registration function. Specifically, the aforementioned initial Beltrami coefficientCan be expressed as follows:
(2)
Wherein, ,,And is also provided with,、、Coordinate information representing triangle vertices in the reference information (i.e. vertices of the surface mesh in the curved surface of the colon),、、Coordinate information representing vertices of other triangles under the intestinal band line,、、The barycentric coordinates ("barycentric") coefficient of a triangle is represented, which is expressed by the ratio of the Area of the triangle ("Area"). By barycentric coefficientsIn the case of an example of this,,Representing the center of gravity. Similarly, it is possible to obtainAnd. Thereby, the foregoingAs is known, by bringing into contactSubstituting the above formula (2) can obtain the initial Beltrami coefficient.
Further, a final registration function is obtained from the initial Beltrami coefficients and the target registration function. In one embodiment, the initial beltermi coefficients are first modified so that the final registration function satisfies the differential homoembryo (i.e., a differential homoembryo function is obtained), then the target registration function is reconstructed from the modified beltermi coefficients, and then the final registration function is determined based on the reconstructed registration function and the target registration function. In one implementation scenario, the initial Beltrami coefficients may be modified specifically based on the following equation so that the final registration function satisfies the differential homoembryo:
(3)
Wherein the method comprises the steps of Representing the modified Beltrami coefficients,Representing the Beltrami coefficients prior to correction (e.g., the initial Beltrami coefficients). From the modified Beltrami coefficients, the target registration function can be reconstructed. In one exemplary scenario, assume that modified Beltrami coefficientsBased on the above formula (1), the following formula can be obtained:
(4)
By sorting this formula (4), it is possible to obtain:
Wherein, ,,,Substitution of the aforementioned parameters into the formulas (5) and (6) can be obtainedThereby, the target registration function can be reconstructed, and the reconstructed registration function can be obtained. Then, a final registration function is determined based on the reconstructed registration function and the target registration function, the final registration function being a differential homoembryo function.
In one embodiment, the reconstructed registration function is taken as the final registration function in response to a maximum value of an absolute value of a difference between the reconstructed registration function and the target registration function meeting a preset threshold. Specifically, when the maximum value of the absolute value of the difference between the reconstructed registration function and the final registration function is smaller than a preset threshold value, the reconstructed registration function is used as the final registration function. As an example, assume that the target registration function is scored asThe registration function after reconstruction is noted asThe preset threshold is recorded asReconstructed registration functionRegistering a function with an objectThe maximum value of the absolute value of the difference is recorded asThen. When (when)<At the time, the reconstructed registration functionAs a final registration function. Conversely, when≥When using the reconstructed registration functionUpdating the registration function (i.e. using the reconstructed registration functionAs the target registration function for the next iteration), repeating the foregoing operations including calculating Beltrami coefficients based on the updated registration function, modifying the Beltrami coefficients, reconstructing the registration function, calculating the maximum value of the absolute value of the difference between the current registration function and the previous (or last iteration) registration functionUp to<And stopping at that time, obtaining a target registration function (namely a differential homoembryo function) so as to further finely optimize the registration result of colorectal curved surface registration based on the differential homoembryo function.
Fig. 12 is an exemplary flow diagram illustrating an ensemble for colorectal curved registration according to an embodiment of the present application. As shown in fig. 12, at step S1101 and step S1102, an elevation three-dimensional curved surface and a depression three-dimensional curved surface of the colorectal are respectively acquired, and the elevation three-dimensional curved surface and the depression three-dimensional curved surface are respectively parameterized to corresponding planar domains. Next, at step S1103 and step S1104, respective central axes of the elevation three-dimensional curved surface and the depression three-dimensional curved surface are extracted, and at step S1105, central axis registration and filtering are performed. Based on the filtered central axis, at step S1106 and step S1107, height maps corresponding to the elevation three-dimensional curved surface and the depression three-dimensional curved surface are respectively constructed.
Further, at step S1108 and step S1109, the feature line and the intestinal band feature point of each of the lying three-dimensional curved surface and the lying three-dimensional curved surface are extracted, respectively. After extracting the intestinal band feature points, at step S1110 and step S1111, first and second intestinal band lines corresponding to the upward three-dimensional curved surface and the downward three-dimensional curved surface, respectively, are obtained based on the intestinal band feature points. Next, at step S1112, the first and second intestinal straps are intestinal strap registered to achieve colorectal curved registration. For more details regarding colorectal curved registration, reference may be made to the description of fig. 1 above, and the present application is not repeated here.
Fig. 13 is an exemplary block diagram illustrating an apparatus 1200 for colorectal curved registration according to an embodiment of the application. It is to be appreciated that the device 1200 may include apparatus of embodiments of the present application, and that the device implementing aspects of the present application may be a single device (e.g., a computing device) or a multi-function device including various peripheral devices.
As shown in fig. 13, the apparatus of the present application may further include a central processing unit or central processing unit ("CPU") 1211, which may be a general purpose CPU, a special purpose CPU, or other information processing and program running execution unit. Further, the device 1200 may also include a mass memory 1212 and a read only memory ("ROM") 1213, wherein the mass memory 1212 may be configured to store various types of data, including various supine and prone three-dimensional surfaces with the colorectal, central axes, altitude maps, intestinal tract, algorithm data, intermediate results, and various programs needed to run the device 1200. ROM 1213 may be configured to store data and instructions necessary to power up self-test device 1200, initialize functional modules in the system, drive basic input/output of the system, and boot the operating system.
Optionally, the device 1200 may also include other hardware platforms or components, such as a tensor processing unit ("TPU") 1214, a graphics processing unit ("GPU") 1215, a field programmable gate array ("FPGA") 1216, and a machine learning unit ("MLU") 1217, as shown. It will be appreciated that while various hardware platforms or components are shown in device 1200, this is by way of example only and not limitation, and that one of skill in the art may add or remove corresponding hardware as desired. For example, device 1200 may include only a CPU, associated memory device, and interface device to implement the method for colorectal curved registration of the present application.
In some embodiments, to facilitate the transfer and interaction of data with external networks, the device 1200 of the present application further comprises a communication interface 1218, whereby the communication interface 1218 may be connected to a local area network/wireless local area network ("LAN/WLAN") 1205, and further, the local server 1206 or Internet ("Internet") 1207. Alternatively or additionally, the device 1200 of the present application may also be directly connected to the internet or cellular network via the communication interface 1218 based on wireless communication technology, such as wireless communication technology based on generation 3 ("3G"), generation 4 ("4G"), or generation 5 ("5G"). In some application scenarios, the device 1200 of the present application may also access the server 1208 and database 1209 of the external network as needed to obtain various known algorithms, data, and modules, and may remotely store various data, such as various types of data or instructions for presenting, for example, the supine and prone three-dimensional surfaces of the colorectal, the central axis, the altitude map, the intestinal band, etc.
Peripheral devices of the apparatus 1200 may include a display device 1202, an input device 1203, and a data transfer interface 1204. In one embodiment, the display device 1202 may, for example, include one or more speakers and/or one or more visual displays configured for voice prompts and/or image video display of the present application for colorectal curved registration. The input device 1203 may include other input buttons or controls, such as a keyboard, mouse, microphone, gesture-capture camera, etc., configured to receive input of audio data and/or user instructions. The data transfer interface 1204 may include, for example, a serial interface, a parallel interface, or a universal serial bus interface ("USB"), a small computer system interface ("SCSI"), serial ATA, fireWire ("FireWire"), PCI Express, and high definition multimedia interface ("HDMI"), etc., configured for data transfer and interaction with other devices or systems. According to aspects of the application, the data transmission interface 1204 may receive the supine and prone three-dimensional surfaces from the colorectal acquired by the CT device and transmit data or results including the supine and prone three-dimensional surfaces of the colorectal or various other types to the device 1200.
The above-described CPU 1211, mass memory 1212, ROM 1213, TPU 1214, GPU 1215, FPGA 1216, MLU 1217, and communication interface 1218 of the device 1200 of the present application may be interconnected by a bus 1219, and data interaction with peripheral devices may be achieved by the bus. In one embodiment, through the bus 1219, the cpu 1211 may control other hardware components in the device 1200 and its peripherals.
An apparatus for colorectal curved registration that may be used to perform the present application is described above in connection with fig. 13. It is to be understood that the device structure or architecture herein is merely exemplary and that the implementation and implementation entities of the present application are not limited thereto, but that changes may be made without departing from the spirit of the present application.
Those skilled in the art will also appreciate from the foregoing description, taken in conjunction with the accompanying drawings, that embodiments of the present application may also be implemented in software programs. The present application thus also provides a computer readable storage medium having stored thereon computer readable instructions for colorectal surface registration, which when executed by one or more processors, may be used to implement the method for colorectal surface registration described in connection with fig. 1, 12 of the present application.
It should be noted that although the operations of the method of the present application are depicted in the drawings in a particular order, this does not require or imply that the operations must be performed in that particular order or that all of the illustrated operations be performed in order to achieve desirable results. Rather, the steps depicted in the flowcharts may change the order of execution. Additionally or alternatively, certain steps may be omitted, multiple steps combined into one step to perform, and/or one step decomposed into multiple steps to perform.
It should be understood that when the terms "first," "second," "third," and "fourth," etc. are used in the claims, the specification and the drawings of the present application, they are used merely to distinguish between different objects, and not to describe a particular order. The terms "comprises" and "comprising" when used in the specification and claims of the present application are taken to specify the presence of stated features, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and/or groups thereof.
It is also to be understood that the terminology used in the description of the application herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the application. As used in the specification and claims, the singular forms "a", "an" and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It should be further understood that the term "and/or" as used in the present specification and claims refers to any and all possible combinations of one or more of the associated listed items, and includes such combinations.
Although the embodiments of the present application are described above, the descriptions are merely examples for facilitating understanding of the present application, and are not intended to limit the scope and application of the present application. Any person skilled in the art can make any modification and variation in form and detail without departing from the spirit and scope of the present disclosure, but the scope of the present disclosure is defined by the appended claims.
In addition, the application collects and acquires various data, accords with the relevant legal requirements and passes the authorization of the data provider. Any organization or individual needs to obtain external data, should lawfully gain authorization and ensure data security, must illegally collect, use, process, transmit unauthorized or unprotected data, must illegally sell, provide or disclose unauthorized or unprotected data.
Claims (8)
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202510295165.4A CN119810167B (en) | 2025-03-13 | 2025-03-13 | Method, device and storage medium for colorectal surface registration |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202510295165.4A CN119810167B (en) | 2025-03-13 | 2025-03-13 | Method, device and storage medium for colorectal surface registration |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN119810167A true CN119810167A (en) | 2025-04-11 |
| CN119810167B CN119810167B (en) | 2025-08-12 |
Family
ID=95276600
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202510295165.4A Active CN119810167B (en) | 2025-03-13 | 2025-03-13 | Method, device and storage medium for colorectal surface registration |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN119810167B (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20070270682A1 (en) * | 2006-05-17 | 2007-11-22 | The Gov't Of The U.S., As Represented By The Secretary Of Health & Human Services, N.I.H. | Teniae coli guided navigation and registration for virtual colonoscopy |
| CN115510526A (en) * | 2022-09-02 | 2022-12-23 | 同济大学 | BIM-based three-dimensional design method and system for assembly type cement concrete pavement |
| CN118037810A (en) * | 2024-02-18 | 2024-05-14 | 首都医科大学附属北京友谊医院 | Method, device and storage medium for extracting the central axis of a three-dimensional colon curvature |
| CN118154789A (en) * | 2024-04-02 | 2024-06-07 | 煤炭科学研究总院有限公司 | Texture calculation method for parametric modeling of 3D shafts and tunnels |
| CN118737392A (en) * | 2024-09-04 | 2024-10-01 | 天津御锦智能医学装备技术有限公司 | A method, device and product for recognizing and positioning colonoscopy images |
-
2025
- 2025-03-13 CN CN202510295165.4A patent/CN119810167B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20070270682A1 (en) * | 2006-05-17 | 2007-11-22 | The Gov't Of The U.S., As Represented By The Secretary Of Health & Human Services, N.I.H. | Teniae coli guided navigation and registration for virtual colonoscopy |
| CN115510526A (en) * | 2022-09-02 | 2022-12-23 | 同济大学 | BIM-based three-dimensional design method and system for assembly type cement concrete pavement |
| CN118037810A (en) * | 2024-02-18 | 2024-05-14 | 首都医科大学附属北京友谊医院 | Method, device and storage medium for extracting the central axis of a three-dimensional colon curvature |
| CN118154789A (en) * | 2024-04-02 | 2024-06-07 | 煤炭科学研究总院有限公司 | Texture calculation method for parametric modeling of 3D shafts and tunnels |
| CN118737392A (en) * | 2024-09-04 | 2024-10-01 | 天津御锦智能医学装备技术有限公司 | A method, device and product for recognizing and positioning colonoscopy images |
Also Published As
| Publication number | Publication date |
|---|---|
| CN119810167B (en) | 2025-08-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Long et al. | Neuraludf: Learning unsigned distance fields for multi-view reconstruction of surfaces with arbitrary topologies | |
| Bartroli et al. | Nonlinear virtual colon unfolding | |
| US9561004B2 (en) | Automated 3-D orthopedic assessments | |
| US20150254857A1 (en) | Image processing system with registration mechanism and method of operation thereof | |
| CN111066065A (en) | System and method for hybrid depth regularization | |
| WO2023207266A9 (en) | Image registration method, apparatus and device, and storage medium | |
| CN111626927B (en) | Binocular image super-resolution method, system and device using parallax constraints | |
| CN107895349B (en) | A Synthesis-Based Deblurring Method for Endoscopic Video | |
| CN118781012B (en) | 3D ultra-high-definition fluorescence medical endoscope imaging method and device | |
| CN112183541A (en) | Contour extraction method and device, electronic device, and storage medium | |
| WO2024098240A1 (en) | Gastrointestinal endoscopy visual reconstruction navigation system and method | |
| WO2011109710A1 (en) | Hierarchical atlas-based segmentation | |
| CN112767340B (en) | Equipment and related products for evaluating lesion areas based on neural network models | |
| CN111402221B (en) | Image processing method and device and electronic equipment | |
| CN112785562A (en) | System for evaluating based on neural network model and related products | |
| CN117115286A (en) | Subtraction image acquisition method, device and computer equipment | |
| Ren et al. | Shape recovery of endoscopic videos by shape from shading using mesh regularization | |
| CN119810167B (en) | Method, device and storage medium for colorectal surface registration | |
| CN115829947A (en) | Model processing apparatus and method | |
| CN118037963B (en) | Method, device, equipment and medium for reconstructing three-dimensional model of digestive cavity inner wall | |
| CN110752004A (en) | Voxel model-based respiratory characteristic characterization method | |
| CN112884817B (en) | Dense optical flow calculation method, device, electronic device and storage medium | |
| CN113658106A (en) | Liver focus automatic diagnosis system based on abdomen enhanced CT | |
| Yung et al. | Journal of visual communication and image representation | |
| JP4887491B2 (en) | MEDICAL IMAGE PROCESSING METHOD, DEVICE THEREOF, AND PROGRAM |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20251205 Address after: 116024 Liaoning, Dalian, Ganjingzi Ling Road, No. 2 Patentee after: DALIAN University OF TECHNOLOGY Country or region after: China Patentee after: BEIJING FRIENDSHIP HOSPITAL, CAPITAL MEDICAL University Patentee after: Capital Normal University Patentee after: Ningbo Mingshang Medical Technology Co.,Ltd. Address before: 116024 Liaoning Province, Dalian City, Ganjingzi District, Linggong Road No. 2 Patentee before: DALIAN University OF TECHNOLOGY Country or region before: China Patentee before: BEIJING FRIENDSHIP HOSPITAL, CAPITAL MEDICAL University Patentee before: Capital Normal University Patentee before: Beijing Mingshang Medical Technology Co.,Ltd. |