阅读说明：本技术 一种前臂骨识别方法及系统 (Forearm bone identification method and system ) 是由 周天丰 崔颖 郭玉冰 周勇 陈山林 刘路 于 2021-08-30 设计创作，主要内容包括：本发明涉及一种前臂骨识别方法及系统,方法包括对各张待测CT影像进行阈值划分和二维连通域计算,得到多个待测连通域；根据待测连通域的属性参数和预测骨骼形状参数确定预设层的连通域；采用逐层搜索的方式向预设层数的远端层的二值图进行连通域匹配,得到第一匹配连通域；根据向预设层数的远端层的二值图进行连通域匹配时的连通域消失情况进行尺骨和桡骨的识别；之后向预设层数的近端层的二值图进行连通域匹配,得到第二匹配连通域；根据匹配连通域得到完整的前臂骨区域,本发明基于二维连通域算法、相邻层骨连通域追踪匹配方法及医学先验知识,能够实现在CT影像中,自动快速获取前臂尺骨及桡骨的所在区域,从而提高前臂骨识别的自动化程度。(The invention relates to a forearm bone identification method and a forearm bone identification system, wherein the method comprises the steps of carrying out threshold division and two-dimensional connected domain calculation on each CT image to be detected to obtain a plurality of connected domains to be detected; determining a connected domain of a preset layer according to the attribute parameters and the predicted skeleton shape parameters of the connected domain to be detected; performing connected domain matching on the binary image of the remote layer with a preset number of layers in a layer-by-layer searching mode to obtain a first matched connected domain; identifying ulna and radius according to the connected domain disappearance condition when the connected domain matching is carried out on the binary image of the far-end layer with the preset number of layers; then, performing connected domain matching on the binary images of the near-end layers with the preset number of layers to obtain a second matching connected domain; the invention can automatically and quickly acquire the areas of the ulna and the radius of the forearm in the CT image based on a two-dimensional connected domain algorithm, a tracking and matching method of adjacent layer bone connected domains and medical priori knowledge, thereby improving the automation degree of forearm bone identification.)

1. A forearm bone identification method is characterized by comprising

Acquiring a CT image set to be detected;

performing threshold division on each CT image to be detected in the CT image set to be detected to obtain a binary image to be detected;

performing two-dimensional connected domain calculation on the binary image with the preset number of layers of the CT image set to be detected to obtain a plurality of connected domains to be detected;

determining a connected domain of a preset layer according to the attribute parameters and the predicted skeleton shape parameters of each connected domain to be detected; the region where the communication domain of the preset layer is located is a forearm bone region corresponding to the preset layer;

performing connected domain matching on the binary image of the preset number of the remote layers in a layer-by-layer searching manner to obtain a first matched connected domain;

identifying the ulna and the radius according to the connected domain disappearance condition when the connected domain matching is carried out on the binary image of the remote layer with the preset number of layers;

performing connected domain matching on the binary image of the near-end layer with the preset number of layers in a layer-by-layer searching mode to obtain a second matching connected domain;

and obtaining a complete forearm bone region according to the first matching connected domain and the second matching connected domain.

2. The forearm bone identification method according to claim 1, wherein the connected domain of a preset layer is determined according to the property parameter and the predicted bone shape parameter of each connected domain to be detected; the region where the connected domain of the preset layer is located is a forearm bone region corresponding to the preset layer, and the method specifically comprises the following steps:

acquiring age data of a tester;

obtaining predicted forearm bone thickness data according to the age data;

calculating a predicted perimeter and a predicted area according to the forearm bone thickness data, and taking the region density of an ideal circle as the predicted region density;

calculating the perimeter to be measured, the area to be measured and the density of the region to be measured of each connected domain to be measured, obtaining the score of the connected domain to be measured according to the predicted perimeter, the predicted area, the density of the predicted region, the perimeter to be measured, the area to be measured, the density of the region to be measured and a preset weight, and judging the score:

if the difference between the magnitudes of the score values of 2 connected domains to be detected is smaller than a first preset judgment threshold value, and the difference between the magnitude of the score value of any one connected domain to be detected in the 2 connected domains to be detected and the magnitudes of the score values of other connected domains to be detected is smaller than a second preset judgment threshold value, determining the CT image set to be detected as a single-side forearm CT image set, and determining the 2 connected domains to be detected as the connected domains of the preset layer;

if the difference between the magnitudes of the scores of any two connected domains to be detected in 4 connected domains to be detected is smaller than the first preset judgment threshold, and the difference between the magnitudes of the scores of any 1 connected domain to be detected in 4 connected domains to be detected and the magnitudes of the scores of other connected domains to be detected is smaller than the second preset judgment threshold, determining the CT image set to be detected as a bilateral forearm CT image set, and determining 4 connected domains to be detected as the connected domains of the preset layer;

if the difference between the magnitudes of the scores of any 1 of the connected domains to be detected and the magnitudes of the scores of other connected domains to be detected is smaller than a second preset judgment threshold, if so, judging whether the relative distance between the connected domains to be detected, of which the magnitudes of the scores are smaller than a third preset judgment threshold, if not, if so, determining each connected domain to be detected with the grade of the score value smaller than a third preset judgment threshold value as a ulna and radius connected domain, and determining the ulna and radius connected domain as the connected domain of the preset layer.

3. The forearm bone recognition method of claim 1, wherein the performing connected component matching on the binary image of the preset number of distal layers in a layer-by-layer search manner to obtain a first matching connected component includes:

calculating a rectangular outer frame according to the two-dimensional coordinate value of the pixel point in the connected domain of the preset layer;

planning adjacent remote layers by using the rectangular outer frames respectively to obtain corresponding areas, and calculating connected areas of the corresponding areas;

matching the connected domain of the corresponding region with the adjacent far-end layer respectively to obtain a matching result;

if the same pixels exist in the matching result, merging the same pixels into a matched connected domain, if the pixel points in the connected domain of the corresponding area appear on the rectangular outer frame, expanding the rectangular outer frame to obtain an expanded area, judging whether the connected domain of the expanded area is matched with the connected domain of the corresponding area, and if so, merging the connected domain of the expanded area into the matched connected domain;

judging whether pixel points exist on the edge of the rectangular outer frame, if not, returning to the step of respectively utilizing the rectangular outer frame to circle corresponding areas on adjacent remote end layers, and calculating connected areas of the corresponding areas, and if so, returning to the step of expanding the rectangular outer frame;

and if the same pixels do not exist in the matching result, determining the connected domain matched in the direction of the far-end layer with the preset number of layers as a first matching connected domain.

4. The forearm bone identification method according to claim 1, wherein identifying the ulna and the radius according to a connected component matching condition when the connected component matching is performed on the binary image of the preset number of distal layers specifically includes:

and judging whether the lost matching is in the same layer, if not, determining the pixel area which is lost matching firstly as an ulna area, and determining the pixel area which is lost matching later as a radius area, if so, determining the area containing the most voxels as the radius area, and determining the area containing the least voxels as the ulna area.

5. The forearm bone identification method according to claim 1, wherein the matching of the connected component to the binary image of the preset number of proximal end layers is performed in a layer-by-layer search manner to obtain a second matching connected component, and specifically includes:

calculating a rectangular outer frame according to the two-dimensional coordinate value of the pixel point in the connected domain of the preset layer;

planning adjacent remote layers by using the rectangular outer frames respectively to obtain corresponding areas, and calculating connected areas of the corresponding areas;

matching the connected domains of the corresponding areas with the adjacent near-end layers respectively to obtain matching results;

if the same pixels exist in the matching result, merging the same pixels into a matched connected domain, if the pixel points in the connected domain of the corresponding area appear on the rectangular outer frame, expanding the rectangular outer frame to obtain an expanded area, judging whether the connected domain of the expanded area is matched with the connected domain of the corresponding area, and if so, merging the connected domain of the expanded area into the matched connected domain;

judging whether pixel points exist on the edge of the rectangular outer frame, if not, returning to the step of respectively utilizing the rectangular outer frame to circle corresponding areas on adjacent near-end layers, calculating connected areas of the corresponding areas, and if so, returning to expand the rectangular outer frame;

and if the same pixels do not exist in the matching result, determining the connected domain matched in the direction of the near-end layer with the preset number of layers as a second matching connected domain.

6. The forearm bone identification method according to claim 5, wherein after the matching of the connected components of the corresponding regions with the adjacent proximal layers respectively to obtain matching results, the method further comprises:

when a connected domain which is communicated with both of the two connected domains of the adjacent layers in the proximal direction appears, the occurrence of the ulnar-radial fusion deformity is determined.

7. The forearm bone identification method according to claim 1, wherein the predetermined number of layers is a middle number of layers of the CT image set to be measured.

8. The forearm bone recognition method of claim 1, wherein the threshold division of each CT image to be detected in the CT image set to obtain a binary image to be detected includes:

determining a preset bone density threshold;

and obtaining the binary image to be detected according to the preset bone density threshold value and the CT image to be detected.

9. An forearm bone recognition system to be applied to the forearm bone recognition method according to any one of claims 1 to 8, the forearm bone recognition system comprising:

the acquisition module is used for acquiring a CT image set to be detected;

the threshold dividing module is used for performing threshold dividing on each CT image to be detected in the CT image set to be detected to obtain a binary image to be detected;

the calculation module is used for performing two-dimensional connected domain calculation on the binary image with the preset number of layers of the CT image set to be detected to obtain a plurality of connected domains to be detected;

the preset connected domain determining module is used for determining the connected domain of a preset layer according to the attribute parameters and the predicted skeleton shape parameters of each connected domain to be detected; the region where the communication domain of the preset layer is located is a forearm bone region corresponding to the preset layer;

the first matching module is used for matching the connected domain to the binary image of the preset number of the remote layers in a layer-by-layer searching mode to obtain a first matching connected domain;

the identification module is used for identifying the ulna and the radius according to the connected domain disappearance condition when the connected domain matching is carried out on the binary image of the preset number of the far-end layers;

the second matching module is used for matching the connected domain to the binary image of the near-end layer with the preset number of layers in a layer-by-layer searching mode to obtain a second matching connected domain;

and the forearm bone acquisition module is used for acquiring a complete forearm bone region according to the first matching connected domain and the second matching connected domain.

10. The forearm bone identification system of claim 9, wherein the preset connected component determination module includes:

an acquisition unit configured to acquire age data of a tester;

a bone thickness estimation unit for obtaining predicted forearm bone thickness data from the age data;

the predicted data calculation unit is used for calculating predicted perimeter and predicted area according to the forearm bone thickness data, and taking the region density of an ideal circle as the predicted region density;

the to-be-detected data calculation unit is used for calculating the to-be-detected perimeter, the to-be-detected area and the to-be-detected region density of each to-be-detected connected domain, and obtaining the score value of the to-be-detected connected domain according to the predicted perimeter, the predicted area, the predicted region density, the to-be-detected perimeter, the to-be-detected area, the to-be-detected region density and a preset weight value;

a judging unit, configured to judge the score value:

if the difference between the magnitudes of the score values of 2 connected domains to be detected is smaller than a first preset judgment threshold value, and the difference between the magnitude of the score value of any one connected domain to be detected in the 2 connected domains to be detected and the magnitudes of the score values of other connected domains to be detected is smaller than a second preset judgment threshold value, determining the CT image set to be detected as a single-side forearm CT image set, and determining the 2 connected domains to be detected as the connected domains of the preset layer;

if the difference between the magnitudes of the scores of any two connected domains to be detected in 4 connected domains to be detected is smaller than the first preset judgment threshold, and the difference between the magnitudes of the scores of any 1 connected domain to be detected in 4 connected domains to be detected and the magnitudes of the scores of other connected domains to be detected is smaller than the second preset judgment threshold, determining the CT image set to be detected as a bilateral forearm CT image set, and determining 4 connected domains to be detected as the connected domains of the preset layer;

if the difference between the magnitudes of the scores of any 1 of the connected domains to be detected and the magnitudes of the scores of other connected domains to be detected is smaller than a second preset judgment threshold, if so, judging whether the relative distance between the connected domains to be detected, of which the magnitudes of the scores are smaller than a third preset judgment threshold, if not, if so, determining each connected domain to be detected with the grade of the score value smaller than a third preset judgment threshold value as a ulna and radius connected domain, and determining the ulna and radius connected domain as the connected domain of the preset layer.

Technical Field

The invention relates to the technical field of medical image recognition, in particular to a forearm bone recognition method and a forearm bone recognition system.

Background

Congenital forearm deformity refers to various deformities occurring in the upper limbs, with an incidence rate of about 1/600. It is characterized by various manifestations, including malaevelopment of radius, megalimb, congenital ulnar deformity, angulation and rotation deformity of ulnar and radius.

Among them, Congenital fusion of ulna and radius (CRS) is a rare Congenital malformation of bone development of upper limbs, mainly manifested by Congenital bony or cartilaginous connection in the proximal end of the ulna and radius. Congenital fusion of the ulna will result in the loss of active and passive rotation function of the forearm, resulting in supination dysfunction of the upper limb of the patient, and as the patient ages, the skeletal muscular system will gradually mature, the motor ability will further decline, and the symptoms will gradually worsen, thereby seriously affecting the patient's daily life. Meanwhile, the disease is autosomal dominant inheritance.

At present, the diagnosis of forearm deformity is mainly realized by means of manual observation of doctors, ulna and radius deformity generally comprises ulna/radius angulation deformity and rotation deformity, the diagnosis is generally realized by marking a bone medial axis and a key point landmark by doctors by using Mimics commercial software, the diagnosis is not automated, and great difficulty exists for doctors with less experience.

Disclosure of Invention

In order to overcome the defects of the prior art, the invention aims to provide a forearm bone identification method and a forearm bone identification system, which can improve the automation degree of forearm bone identification.

In order to achieve the purpose, the invention provides the following scheme:

a forearm bone identification method comprises

Acquiring a CT image set to be detected;

performing threshold division on each CT image to be detected in the CT image set to be detected to obtain a binary image to be detected;

performing two-dimensional connected domain calculation on the binary image with the preset number of layers of the CT image set to be detected to obtain a plurality of connected domains to be detected;

determining a connected domain of a preset layer according to the attribute parameters and the predicted skeleton shape parameters of each connected domain to be detected; the region where the communication domain of the preset layer is located is a forearm bone region corresponding to the preset layer;

performing connected domain matching on the binary image of the preset number of the remote layers in a layer-by-layer searching manner to obtain a first matched connected domain;

identifying the ulna and the radius according to the connected domain disappearance condition when the connected domain matching is carried out on the binary image of the remote layer with the preset number of layers;

performing connected domain matching on the binary image of the near-end layer with the preset number of layers in a layer-by-layer searching mode to obtain a second matching connected domain;

and obtaining a complete forearm bone region according to the first matching connected domain and the second matching connected domain.

Preferably, the connected domain of the preset layer is determined according to the attribute parameters and the predicted bone shape parameters of each connected domain to be detected; the region where the connected domain of the preset layer is located is a forearm bone region corresponding to the preset layer, and the method specifically comprises the following steps:

acquiring age data of a tester;

obtaining predicted forearm bone thickness data according to the age data;

calculating a predicted perimeter and a predicted area according to the forearm bone thickness data, and taking the region density of an ideal circle as the predicted region density;

calculating the perimeter to be measured, the area to be measured and the density of the region to be measured of each connected domain to be measured, obtaining the score of the connected domain to be measured according to the predicted perimeter, the predicted area, the density of the predicted region, the perimeter to be measured, the area to be measured, the density of the region to be measured and a preset weight, and judging the score:

if the difference between the magnitudes of the score values of 2 connected domains to be detected is smaller than a first preset judgment threshold value, and the difference between the magnitude of the score value of any one connected domain to be detected in the 2 connected domains to be detected and the magnitudes of the score values of other connected domains to be detected is smaller than a second preset judgment threshold value, determining the CT image set to be detected as a single-side forearm CT image set, and determining the 2 connected domains to be detected as the connected domains of the preset layer;

if the difference between the magnitudes of the scores of any two connected domains to be detected in 4 connected domains to be detected is smaller than the first preset judgment threshold, and the difference between the magnitudes of the scores of any 1 connected domain to be detected in 4 connected domains to be detected and the magnitudes of the scores of other connected domains to be detected is smaller than the second preset judgment threshold, determining the CT image set to be detected as a bilateral forearm CT image set, and determining 4 connected domains to be detected as the connected domains of the preset layer;

if the difference between the magnitudes of the scores of any 1 of the connected domains to be detected and the magnitudes of the scores of other connected domains to be detected is smaller than a second preset judgment threshold, if so, judging whether the relative distance between the connected domains to be detected, of which the magnitudes of the scores are smaller than a third preset judgment threshold, if not, if so, determining each connected domain to be detected with the grade of the score value smaller than a third preset judgment threshold value as a ulna and radius connected domain, and determining the ulna and radius connected domain as the connected domain of the preset layer.

Preferably, the performing connected domain matching on the binary image of the preset number of far-end layers in a layer-by-layer search manner to obtain a first matching connected domain includes:

calculating a rectangular outer frame according to the two-dimensional coordinate value of the pixel point in the connected domain of the preset layer;

planning adjacent remote layers by using the rectangular outer frames respectively to obtain corresponding areas, and calculating connected areas of the corresponding areas;

matching the connected domain of the corresponding region with the adjacent far-end layer respectively to obtain a matching result;

if the same pixels exist in the matching result, merging the same pixels into a matched connected domain, if the pixel points in the connected domain of the corresponding area appear on the rectangular outer frame, expanding the rectangular outer frame to obtain an expanded area, judging whether the connected domain of the expanded area is matched with the connected domain of the corresponding area, and if so, merging the connected domain of the expanded area into the matched connected domain;

judging whether pixel points exist on the edge of the rectangular outer frame, if not, returning to the step of respectively utilizing the rectangular outer frame to circle corresponding areas on adjacent remote end layers, and calculating connected areas of the corresponding areas, and if so, returning to the step of expanding the rectangular outer frame;

and if the same pixels do not exist in the matching result, determining the connected domain matched in the direction of the far-end layer with the preset number of layers as a first matching connected domain.

Preferably, the identification of the ulna and the radius is performed according to a connected domain matching condition when the connected domain matching is performed on the binary image of the preset number of the distal layers, specifically including:

and judging whether the lost matching is in the same layer, if not, determining the pixel area which is lost matching firstly as an ulna area, and determining the pixel area which is lost matching later as a radius area, if so, determining the area containing the most voxels as the radius area, and determining the area containing the least voxels as the ulna area.

Preferably, the performing connected domain matching on the binary image of the near-end layer with the preset number of layers in a layer-by-layer search manner to obtain a second matching connected domain specifically includes:

calculating a rectangular outer frame according to the two-dimensional coordinate value of the pixel point in the connected domain of the preset layer;

planning adjacent remote layers by using the rectangular outer frames respectively to obtain corresponding areas, and calculating connected areas of the corresponding areas;

matching the connected domains of the corresponding areas with the adjacent near-end layers respectively to obtain matching results;

if the same pixels exist in the matching result, merging the same pixels into a matched connected domain, if the pixel points in the connected domain of the corresponding area appear on the rectangular outer frame, expanding the rectangular outer frame to obtain an expanded area, judging whether the connected domain of the expanded area is matched with the connected domain of the corresponding area, and if so, merging the connected domain of the expanded area into the matched connected domain;

judging whether pixel points exist on the edge of the rectangular outer frame, if not, returning to the step of respectively utilizing the rectangular outer frame to circle corresponding areas on adjacent near-end layers, calculating connected areas of the corresponding areas, and if so, returning to expand the rectangular outer frame;

and if the same pixels do not exist in the matching result, determining the connected domain matched in the direction of the near-end layer with the preset number of layers as a second matching connected domain.

Preferably, after the matching the connected domain of the corresponding region with the adjacent near-end layer respectively to obtain a matching result, the method further includes:

when a connected domain which is communicated with both of the two connected domains of the adjacent layers in the proximal direction appears, the occurrence of the ulnar-radial fusion deformity is determined.

Preferably, the preset number of layers is the number of middle layers of the CT image set to be detected.

Preferably, the threshold division is performed on each to-be-detected CT image in the to-be-detected CT image set to obtain a to-be-detected binary image, and the threshold division includes:

determining a preset bone density threshold;

and obtaining the binary image to be detected according to the preset bone density threshold value and the CT image to be detected.

An forearm bone identification system applied to the forearm bone identification method, the forearm bone identification system comprising:

the acquisition module is used for acquiring a CT image set to be detected;

the threshold dividing module is used for performing threshold dividing on each CT image to be detected in the CT image set to be detected to obtain a binary image to be detected;

the calculation module is used for performing two-dimensional connected domain calculation on the binary image with the preset number of layers of the CT image set to be detected to obtain a plurality of connected domains to be detected;

the preset connected domain determining module is used for determining the connected domain of a preset layer according to the attribute parameters and the predicted skeleton shape parameters of each connected domain to be detected; the region where the communication domain of the preset layer is located is a forearm bone region corresponding to the preset layer;

the first matching module is used for matching the connected domain to the binary image of the preset number of the remote layers in a layer-by-layer searching mode to obtain a first matching connected domain;

the identification module is used for identifying the ulna and the radius according to the connected domain disappearance condition when the connected domain matching is carried out on the binary image of the preset number of the far-end layers;

the second matching module is used for matching the connected domain to the binary image of the near-end layer with the preset number of layers in a layer-by-layer searching mode to obtain a second matching connected domain;

and the forearm bone acquisition module is used for acquiring a complete forearm bone region according to the first matching connected domain and the second matching connected domain.

Preferably, the preset connected domain determining module includes:

an acquisition unit configured to acquire age data of a tester;

a bone thickness estimation unit for obtaining predicted forearm bone thickness data from the age data;

the predicted data calculation unit is used for calculating predicted perimeter and predicted area according to the forearm bone thickness data, and taking the region density of an ideal circle as the predicted region density;

the to-be-detected data calculation unit is used for calculating the to-be-detected perimeter, the to-be-detected area and the to-be-detected region density of each to-be-detected connected domain, and obtaining the score value of the to-be-detected connected domain according to the predicted perimeter, the predicted area, the predicted region density, the to-be-detected perimeter, the to-be-detected area, the to-be-detected region density and a preset weight value;

a judging unit, configured to judge the score value:

if the difference between the magnitudes of the score values of 2 connected domains to be detected is smaller than a first preset judgment threshold value, and the difference between the magnitude of the score value of any one connected domain to be detected in the 2 connected domains to be detected and the magnitudes of the score values of other connected domains to be detected is smaller than a second preset judgment threshold value, determining the CT image set to be detected as a single-side forearm CT image set, and determining the 2 connected domains to be detected as the connected domains of the preset layer;

if the difference between the magnitudes of the scores of any two connected domains to be detected in 4 connected domains to be detected is smaller than the first preset judgment threshold, and the difference between the magnitudes of the scores of any 1 connected domain to be detected in 4 connected domains to be detected and the magnitudes of the scores of other connected domains to be detected is smaller than the second preset judgment threshold, determining the CT image set to be detected as a bilateral forearm CT image set, and determining 4 connected domains to be detected as the connected domains of the preset layer;

if the difference between the magnitudes of the scores of any 1 of the connected domains to be detected and the magnitudes of the scores of other connected domains to be detected is smaller than a second preset judgment threshold, if so, judging whether the relative distance between the connected domains to be detected, of which the magnitudes of the scores are smaller than a third preset judgment threshold, if not, if so, determining each connected domain to be detected with the grade of the score value smaller than a third preset judgment threshold value as a ulna and radius connected domain, and determining the ulna and radius connected domain as the connected domain of the preset layer.

According to the specific embodiment provided by the invention, the invention discloses the following technical effects:

the method provided by the invention mainly solves the problems of realizing automatic identification of the ulna and radius region and automatic diagnosis of the ulna and radius deformity in the forearm deformity diagnosis. The invention can automatically and quickly acquire the areas of the ulna and the radius of the forearm in the CT image based on a two-dimensional connected domain algorithm, a tracking and matching method of adjacent layer bone connected domains and medical priori knowledge, thereby improving the automation degree of forearm bone identification. In a specific embodiment, the present invention provides a diagnosis of the presence or absence of a ulnar fusion deformity. By giving the area of the ulna and radius of the forearm, it is possible to provide the necessary basis for the diagnosis of the presence of other deformities in the ulna and radius.

Drawings

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings needed to be used in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings without inventive exercise.

FIG. 1 is a flow chart of an identification method in an embodiment provided by the present invention;

FIG. 2 is a flow chart of automatic forearm ulnar radius identification and automatic forearm ulnar fusion diagnosis in an embodiment provided by the invention;

FIG. 3 is a two-valued graph corresponding to an intermediate CT image in an embodiment of the invention;

FIG. 4 is a schematic diagram of a connected domain of the automatically selected mid-level forearm radius in accordance with an embodiment of the present invention;

FIG. 5 is a schematic diagram of an initial rectangular detection box in an embodiment of the present invention;

FIG. 6 is a schematic diagram of a rectangular detection box automatically expanded according to the calculation condition of the current layer connected domain in the embodiment of the present invention;

fig. 7 is a partial diagram of a left-side remote 227 layer communication domain binary map in an embodiment of the present invention;

fig. 8 is a left-side remote-end 228 layer communication-domain binary map partial diagram in an embodiment provided by the present invention;

fig. 9 is a partial diagram of a left-side remote 229-level communication domain binary map in an embodiment of the present invention;

fig. 10 is a partial diagram of a left-side far-end 230 layer communication domain binary map in an embodiment of the present invention;

FIG. 11 is a partial diagram of a binary map of the 233 layers of the left forearm distal end in accordance with an embodiment of the invention;

FIG. 12 is a partial diagram of a left forearm distal 234 layer binary map in accordance with an embodiment of the invention;

FIG. 13 is a partial diagram of a binary map of the left forearm distal 235 layers in accordance with an embodiment of the invention;

FIG. 14 is a partial diagram of a left forearm distal 236 layer binary map in accordance with an embodiment of the invention;

FIG. 15 is a left radial access domain of the distal half in accordance with an embodiment of the present invention;

FIG. 16 is a left ulnar communication domain of the distal half in accordance with an embodiment of the present invention;

FIG. 17 illustrates a distal half segment right radius approach in accordance with an exemplary embodiment of the present invention;

FIG. 18 illustrates the distal half right ulnar communication area in accordance with an embodiment of the present invention;

FIG. 19 illustrates a layer 117 of the proximal left ulnar radius in accordance with an embodiment of the present invention;

FIG. 20 illustrates a layer 116 of the proximal left ulnar radius in accordance with an embodiment of the present invention;

FIG. 21 illustrates 115 layers of the proximal left ulnar radius in accordance with an embodiment of the present invention;

FIG. 22 illustrates a layer 114 of the proximal left ulnar radius in accordance with an embodiment of the present invention;

FIG. 23 illustrates the left ulnar radius region in accordance with an embodiment of the present invention;

FIG. 24 is a right ulnar region in accordance with an embodiment of the present invention;

fig. 25 is a right radial region in accordance with an embodiment of the present invention.

Detailed Description

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.

The invention aims to provide a forearm bone identification method and a forearm bone identification system, which can realize automatic forearm bone identification.

In order to make the aforementioned objects, features and advantages of the present invention comprehensible, embodiments accompanied with figures are described in further detail below.

Fig. 1 and 2 are a flowchart of an identification method and a flowchart of automatic identification of forearm ulna and automatic diagnosis of forearm ulna fusion in an embodiment of the present invention, and as shown in fig. 1 and 2, the present invention provides a forearm bone identification method, including

Step 100: acquiring a CT image set to be detected;

step 200: performing threshold division on each CT image to be detected in the CT image set to be detected to obtain a binary image to be detected;

step 300: performing two-dimensional connected domain calculation on the binary image with the preset number of layers of the CT image set to be detected to obtain a plurality of connected domains to be detected;

step 400: determining a connected domain of a preset layer according to the attribute parameters and the predicted skeleton shape parameters of each connected domain to be detected; the region where the communication domain of the preset layer is located is a forearm bone region corresponding to the preset layer;

step 500: performing connected domain matching on the binary image of the preset number of the remote layers in a layer-by-layer searching manner to obtain a first matched connected domain;

step 600: identifying the ulna and the radius according to the connected domain disappearance condition when the connected domain matching is carried out on the binary image of the remote layer with the preset number of layers;

step 700: performing connected domain matching on the binary image of the near-end layer with the preset number of layers in a layer-by-layer searching mode to obtain a second matching connected domain;

step 800: and obtaining a complete forearm bone region according to the first matching connected domain and the second matching connected domain.

Optionally, acquiring the set of CT images to be tested includes acquiring CT images of DICOM files of the forearm of the patient, which should contain the entire forearm bone (i.e., ulna and radius), which may contain part of the forearm (humerus) and the hand (carpal, metacarpal and phalanges). The direction of the forearm should be approximately parallel to the direction of the CT image slice number (not too far from the spectrum), and the CT slice number at the proximal end of the forearm is smaller than that at the distal end of the forearm.

Preferably, the CT image set includes a long bone to be detected.

Optionally, the preset number of layers is a middle number of layers of the CT image set to be detected.

Preferably, the threshold division is performed on each to-be-detected CT image in the to-be-detected CT image set to obtain a to-be-detected binary image, and the threshold division includes:

determining a preset bone density threshold;

and obtaining the binary image to be detected according to the preset bone density threshold value and the CT image to be detected.

Specifically, initial threshold segmentation is performed to extract a bone region in the CT image, and the whole CT image is saved as a binary three-dimensional image (bone region is 1, background region is 0) with the same size as the original CT image, which is referred to as binary image hereinafter.

Preferably, the connected domain of the preset layer is determined according to the attribute parameters and the predicted bone shape parameters of each connected domain to be detected; the region where the connected domain of the preset layer is located is a forearm bone region corresponding to the preset layer, and the method specifically comprises the following steps:

acquiring age data of a tester;

obtaining predicted forearm bone thickness data according to the age data;

calculating a predicted perimeter and a predicted area according to the forearm bone thickness data, and taking the region density of an ideal circle as the predicted region density;

calculating the perimeter to be measured, the area to be measured and the density of the region to be measured of each connected domain to be measured, obtaining the score of the connected domain to be measured according to the predicted perimeter, the predicted area, the density of the predicted region, the perimeter to be measured, the area to be measured, the density of the region to be measured and a preset weight, and judging the score:

if the difference between the magnitudes of the score values of 2 connected domains to be detected is smaller than a first preset judgment threshold value, and the difference between the magnitude of the score value of any one connected domain to be detected in the 2 connected domains to be detected and the magnitudes of the score values of other connected domains to be detected is smaller than a second preset judgment threshold value, determining the CT image set to be detected as a single-side forearm CT image set, and determining the 2 connected domains to be detected as the connected domains of the preset layer;

if the difference between the magnitudes of the scores of any two connected domains to be detected in 4 connected domains to be detected is smaller than the first preset judgment threshold, and the difference between the magnitudes of the scores of any 1 connected domain to be detected in 4 connected domains to be detected and the magnitudes of the scores of other connected domains to be detected is smaller than the second preset judgment threshold, determining the CT image set to be detected as a bilateral forearm CT image set, and determining 4 connected domains to be detected as the connected domains of the preset layer;

if the difference between the magnitudes of the scores of any 1 of the connected domains to be detected and the magnitudes of the scores of other connected domains to be detected is smaller than a second preset judgment threshold, if so, judging whether the relative distance between the connected domains to be detected, of which the magnitudes of the scores are smaller than a third preset judgment threshold, if not, if so, determining each connected domain to be detected with the grade of the score value smaller than a third preset judgment threshold value as a ulna and radius connected domain, and determining the ulna and radius connected domain as the connected domain of the preset layer.

Specifically, the age data of the patient in the DICOM file is read, and the thickness of the forearm bone of the patient is estimated. Because the size of the ulna and the size and the shape of other objects (such as a CT bed and the like) are different greatly in a CT plan, the prediction error of the thickness of the ulna and the radius of the forearm of a patient is insensitive to the identification of the ulna and the radius of the middle layer, and can be roughly taken as d_predict10-15 mm. And calculating the corresponding predicted circumference P_predict＝π·d_predictPredicted area

Optionally, an intermediate layer in the binary image (generally, the intermediate layer corresponds to a certain layer in the middle segment of the ulna and the radius, and is embodied in a state that the ulna and the radius are not fused) is extracted, and if there is a difference in the CT image shooting range, the layer can be manually specified, but the labor workload is hardly spent. Two-dimensional connected domain calculation is carried out on the intermediate layer, and the perimeter P of each obtained connected domain is calculated_cArea S_cAnd region density R_c＝P_c ²/S_c. Calculating the score of each connected domain based on the following formula:

wherein, w_P、w_SAnd w_RThe weight of the cost corresponding to the perimeter, the area and the density of the regions is represented by the attention degree of three judgment indexes when searching for an ulna connected domain and a radius connected domain in a plurality of connected domains, and the proportion of the three judgment indexes to the forearm bone can be taken as w_P:w_S:w_R＝1:1:10。

Sorting the cost values of all the connected domains in size, comparing the magnitude of the cost values of all the connected domains, and if the magnitude of the cost values of 2 connected domains is close to and far smaller than the magnitude of the cost values of other connected domains, indicating that the connected domains are single-side forearm CT images; if the magnitude of the cost values of 4 connected domains is close to and much smaller than that of the other connected domains, the two-sided forearm CT image is shown. If the other conditions exist, comparing the relative distances of the connected domains with smaller cost value magnitude, and selecting the predicted value d of the radius diameter with the relative distance close to 1-3 times of the radius diameter_predictThe connected domain of (a) is considered as a set of ulnar and radial connected domains.

Specifically, the four initial connected domains obtained were filled with the cavities (i.e., the supplemented bone marrow fraction) and divided into left and right forearms according to their central positions. Are designated as middle layer left forearm 1, middle layer left forearm 2, middle layer right forearm 1 and middle layer right forearm 2.

Preferably, the performing connected domain matching on the binary image of the preset number of far-end layers in a layer-by-layer search manner to obtain a first matching connected domain includes:

calculating a rectangular outer frame according to the two-dimensional coordinate value of the pixel point in the connected domain of the preset layer;

planning adjacent remote layers by using the rectangular outer frames respectively to obtain corresponding areas, and calculating connected areas of the corresponding areas;

matching the connected domain of the corresponding region with the adjacent far-end layer respectively to obtain a matching result;

if the same pixels exist in the matching result, merging the same pixels into a matched connected domain, if the pixel points in the connected domain of the corresponding area appear on the rectangular outer frame, expanding the rectangular outer frame to obtain an expanded area, judging whether the connected domain of the expanded area is matched with the connected domain of the corresponding area, and if so, merging the connected domain of the expanded area into the matched connected domain;

judging whether pixel points exist on the edge of the rectangular outer frame, if not, returning to the step of respectively utilizing the rectangular outer frame to circle corresponding areas on adjacent remote end layers, and calculating connected areas of the corresponding areas, and if so, returning to the step of expanding the rectangular outer frame;

and if the same pixels do not exist in the matching result, determining the connected domain matched in the direction of the far-end layer with the preset number of layers as a first matching connected domain.

As an alternative embodiment, the left side is used as an example to illustrate how to find the complete forearm bone. Recording the middle layer left forearm 1 and the middle layer left forearm 2 as the adjacent upper layer left forearm 1 and the adjacent upper layer left forearm 2, and calculating the minimum rectangular outer frame based on the two-dimensional coordinate values of the pixel points in the two connected domains. The minimum rectangular outer frame is outwards expanded by n pixels to obtain an expanded rectangular outer frame area (the minimum rectangular outer frame area can be not expanded, and the calculation efficiency can be further improved by properly expanding the minimum rectangular outer frame area);

in this embodiment, a rectangular frame is calculated from the left/right side regions of adjacent layers, a connected component is calculated in the rectangular frame, and the rectangular frame is enlarged according to the calculation, so that the bone connected component of the layer is a complete forearm bone region.

Optionally, tracking the distal ulna radius layer by layer, circling a corresponding region by using an enlarged rectangular outer frame region on one layer of the adjacent distal direction, calculating a connected domain inside the region, matching each connected domain with the adjacent upper left forearm 1 and the adjacent upper left forearm 2, and recording the connected domains as the left forearm 1 or the left forearm 2 if the same pixel exists; if partial pixel points in the two connected domains appear on the enlarged rectangular outer frame, moving the corresponding rectangular edge outwards by 1 unit to obtain a modified rectangular outer frame; then, calculating a connected domain of the increased area, judging whether the connected domain is communicated with the previous connected domain, and if so, adding the left forearm 1 or 2 of the layer communicated with the connected domain; this procedure is cycled until no pixel points appear on the edges of the rectangular box for the layer of left forearm 1 connected fields and the layer of left forearm 2 connected fields, and the layer of left forearm 1 and the layer of left forearm 2 connected fields are recorded as the previous layer of left forearm 1 and the previous layer of left forearm 2 and added to left forearm 1 and left forearm 2, respectively.

As an optional implementation manner, in this embodiment, the preset detection step is 1, and the preset detection step may also be a value greater than 1, that is, in this embodiment, tracking may also be performed by an even layer in the adjacent far-end direction.

Optionally, the current matching connected domain is an N × 3-dimensional matrix, where N — N1+ N2+ N3 … corresponds to the number of pixels included in each layer of matched bone.

The above steps are cycled until either left forearm 1 or left forearm 2 has no matching connected domain. When there is no match between the left forearm 1 or the left forearm 2, the other forearm bone is continued to be matched in the same way until there is no match between the two forearm bones.

Preferably, the identification of the ulna and the radius is performed according to a connected domain matching condition when the connected domain matching is performed on the binary image of the preset number of the distal layers, specifically including:

and judging whether the lost matching is in the same layer, if not, determining the pixel area which is lost matching firstly as an ulna area, and determining the pixel area which is lost matching later as a radius area, if so, determining the area containing the most voxels as the radius area, and determining the area containing the least voxels as the ulna area.

In the present embodiment, the ulna/radius is automatically judged according to medical knowledge: marking the left forearm which disappears first as the left ulna and the left forearm which disappears later as the left radius; and if the two components and the same layer disappear, recording the obtained half of the left forearm 1 and the left forearm 2 as the left radius bone with a larger number of voxels, and recording the obtained half of the left forearm 1 and the obtained half of the left forearm 2 as the left ulna bone with a smaller number of voxels, thereby completing the identification and acquisition of the distal half segment of the left ulna radius bone.

In this embodiment, the matching for the near-end layer is as follows:

the method for matching the connected domains to the binary image of the near-end layer with the preset number of layers by adopting a layer-by-layer searching mode to obtain a second matching connected domain specifically comprises the following steps:

calculating a rectangular outer frame according to the two-dimensional coordinate value of the pixel point in the connected domain of the preset layer;

planning adjacent remote layers by using the rectangular outer frames respectively to obtain corresponding areas, and calculating connected areas of the corresponding areas;

matching the connected domains of the corresponding areas with the adjacent near-end layers respectively to obtain matching results;

if the same pixels exist in the matching result, merging the same pixels into a matched connected domain, if the pixel points in the connected domain of the corresponding area appear on the rectangular outer frame, expanding the rectangular outer frame to obtain an expanded area, judging whether the connected domain of the expanded area is matched with the connected domain of the corresponding area, and if so, merging the connected domain of the expanded area into the matched connected domain;

judging whether pixel points exist on the edge of the rectangular outer frame, if not, returning to the step of respectively utilizing the rectangular outer frame to circle corresponding areas on adjacent near-end layers, calculating connected areas of the corresponding areas, and if so, returning to expand the rectangular outer frame;

and if the same pixels do not exist in the matching result, determining the connected domain matched in the direction of the near-end layer with the preset number of layers as a second matching connected domain. .

Preferably, after the matching the connected domain of the corresponding region with the adjacent near-end layer respectively to obtain a matching result, the method further includes:

when a connected domain which is communicated with both of the two connected domains of the adjacent layers in the proximal direction appears, the occurrence of the ulnar-radial fusion deformity is determined.

Specifically, the matching and acquisition of the proximal ulna and radius are performed as follows. Before ulnar fusion occurs, it is matched and acquired in the same manner as the distal calculation. When a layer at the near end is tracked, the connected domain in the rectangular frame of the layer is matched with the connected domains of the two bones in the adjacent layer, the ulnar fusion deformity is identified, the other sections are continuously tracked until no matching exists, the condition that the forearm ulnar reaches the near end point is indicated, and the identification of the ulnar is completed; and if the condition that the communicating region of one layer is communicated with the two adjacent bone communicating regions of the previous layer does not occur in the tracking process of the proximal half section till the proximal end point, the patient is considered to have no ulna and radius deformity.

Optionally, the proximal ulna half-segment communication domain is merged with the distal half-segment communication domain. And obtaining the conclusion of whether the proximal ulna and radius fusion deformity and the position and the area of the ulna and radius exist.

In the practical application process, a set of CT image data of forearms of male five-year-old children is taken, the storage format of the CT image data is DICOM format, 357 CT images are contained, and the CT image data completely comprises bilateral forearm ulnar radius, wherein the left forearm ulnar radius has fusion deformity, and the right forearm ulnar radius is normal.

First, the middle layer CT image is selected, and the number of layers is selected to be 179 th layer in this embodiment. Setting the bone density threshold of the CT image to 226, and obtaining a binary image of the CT image. As shown in fig. 3. Two-dimensional connected domain analysis is carried out on the obtained data to obtain 28 connected domains.

The predicted bone diameter was set to 12mm, and the circumference and area of the forearm bone predicted therefrom are shown in table 1.

TABLE 1

Calculating the size of the connected domain, excluding the undersized connected domain smaller than 10 pixels, considering the connected domain as interference, and then respectively calculating the perimeter, the area and the density of the region of each other connected domain, as shown in the table. The weighted distance of each connected domain to the predicted value is then calculated according to the following formula.

Wherein P, S and R respectively represent the perimeter, area and density of the connected domain,andrespectively representing the predicted values; w_P、W_SAnd W_RAnd respectively representing the cost weight of judging whether the connected domain belongs to the forearm bone connected domain or not with respect to the perimeter, the area and the density of the region, wherein T is the transpose of the matrix. In this embodiment, the cost weight is taken as W_P＝1、W_S1 and W_R＝10。

And sorting the weighted distances from small to large, and automatically selecting the connected domains corresponding to the minimum four weighted distances, namely the connected domains corresponding to the numbers 5, 2, 3 and 1. The representation of the corresponding connected component on the binary map is given in fig. 4.

And dividing the four forearm ulna and radius communication areas into a left forearm bone and a right forearm bone according to the x coordinate size of the four forearm ulna and radius communication areas, wherein each side comprises two forearm bones which respectively correspond to an ulna or a radius.

The distal layer is searched layer by layer (left side is taken as an example, and the right side calculation method is the same), the left ulna and radius connected domains are merged, and a rectangular frame of the initial range is given, as shown in fig. 5. For clarity of display, each side of the rectangular box is shifted outward by one pixel.

And calculating connected domains of pixels in the gray rectangular frame, respectively matching each obtained connected domain with the two ulna/radius connected domains of the previous layer, if the connected domains are matched, adding the corresponding bone connected domains, and otherwise, ignoring the connected domains.

If both forearm bones are matched, it is detected whether there are pixels on the connected domain on the edge of the rectangular frame, as shown in fig. 5. As can be seen from fig. 5, pixel points of the ulna and radius connected domain of the current layer exist on the upper, lower and right boundaries of the gray rectangular frame, and if the corresponding edge needs to move outward, the corresponding edge moves outward one line/one line of pixels in sequence, the connected domain on the newly added line/one line of pixels is calculated, and the connected domain is merged with the ulna/radius connected domain in the gray rectangular frame according to the connectivity, so as to detect whether a pixel point matched with the ulna and radius exists on the edge of the new rectangular frame. If the corresponding edges exist, the loop expands the corresponding edges again, if the corresponding edges do not exist, the corresponding edges of the initial rectangular frame are expanded by 1 pixel outwards, and the loop is ended. The resulting rectangular frame area is shown in fig. 6.

And searching ulna and radius connected domains layer by adopting the same method, wherein in the 229 th layer, only one bone on the left side is matched, and if the left ulna reaches the far-end point, recording the voxel connected domain corresponding to the bone which loses matching as the left ulna connected domain, and recording the other voxel connected domain as the left radius connected domain. The binary CT images of the slice are shown in fig. 7 to 10, which can prove the correctness of the algorithm.

At level 235, the left radial connected domain is out of match, the left forearm reaches the distal end point, and the CT image binary maps for this and the adjacent levels are shown in fig. 11-14. Here the lower right is the left radial communicating domain. From layer 234, the communication domain appearing above the left is the greater arm humerus communication domain.

The resulting distal half of the forearm bone is shown in fig. 15-18. Left radius, left ulna, right radius and right ulna, respectively.

In a similar way to the layers at the distal end, as shown in fig. 19 to 22, at the 115 th layer, a connected domain appears in the local area of the left ulna radius, and the connected domain is communicated with both the left ulna connected domain and the left radius connected domain of the adjacent layer in the distal direction, so that the fusion deformity of the left ulna radius is identified.

And the characteristics of ulna and radius fusion do not appear until the ulna and radius disappear from the near end of the right end, and the corresponding fusion flag is 0.

The resulting automatically generated left ulnar radius area is shown in fig. 23-25, the right ulna as shown and the right radius as shown in fig. 10.

The present invention also provides a forearm bone recognition system, comprising:

the acquisition module is used for acquiring a CT image set to be detected;

the threshold dividing module is used for performing threshold dividing on each CT image to be detected in the CT image set to be detected to obtain a binary image to be detected;

the calculation module is used for performing two-dimensional connected domain calculation on the binary image with the preset number of layers of the CT image set to be detected to obtain a plurality of connected domains to be detected;

the preset connected domain determining module is used for determining the connected domain of a preset layer according to the attribute parameters and the predicted skeleton shape parameters of each connected domain to be detected; the region where the communication domain of the preset layer is located is a forearm bone region corresponding to the preset layer;

the first matching module is used for matching the connected domain to the binary image of the preset number of the remote layers in a layer-by-layer searching mode to obtain a first matching connected domain;

the identification module is used for identifying the ulna and the radius according to the connected domain disappearance condition when the connected domain matching is carried out on the binary image of the preset number of the far-end layers;

the second matching module is used for matching the connected domain to the binary image of the near-end layer with the preset number of layers in a layer-by-layer searching mode to obtain a second matching connected domain;

and the forearm bone acquisition module is used for acquiring a complete forearm bone region according to the first matching connected domain and the second matching connected domain.

Preferably, the preset connected domain determining module includes:

an acquisition unit configured to acquire age data of a tester;

a bone thickness estimation unit for obtaining predicted forearm bone thickness data from the age data;

the predicted data calculation unit is used for calculating predicted perimeter and predicted area according to the forearm bone thickness data, and taking the region density of an ideal circle as the predicted region density;

the to-be-detected data calculation unit is used for calculating the to-be-detected perimeter, the to-be-detected area and the to-be-detected region density of each to-be-detected connected domain, and obtaining the score value of the to-be-detected connected domain according to the predicted perimeter, the predicted area, the predicted region density, the to-be-detected perimeter, the to-be-detected area, the to-be-detected region density and a preset weight value;

a judging unit, configured to judge the score value:

if the difference between the magnitudes of the score values of 2 connected domains to be detected is smaller than a first preset judgment threshold value, and the difference between the magnitude of the score value of any one connected domain to be detected in the 2 connected domains to be detected and the magnitudes of the score values of other connected domains to be detected is smaller than a second preset judgment threshold value, determining the CT image set to be detected as a single-side forearm CT image set, and determining the 2 connected domains to be detected as the connected domains of the preset layer;

if the difference between the magnitudes of the scores of any two connected domains to be detected in 4 connected domains to be detected is smaller than the first preset judgment threshold, and the difference between the magnitudes of the scores of any 1 connected domain to be detected in 4 connected domains to be detected and the magnitudes of the scores of other connected domains to be detected is smaller than the second preset judgment threshold, determining the CT image set to be detected as a bilateral forearm CT image set, and determining 4 connected domains to be detected as the connected domains of the preset layer;

if the difference between the magnitudes of the scores of any 1 of the connected domains to be detected and the magnitudes of the scores of other connected domains to be detected is smaller than a second preset judgment threshold, if so, judging whether the relative distance between the connected domains to be detected, of which the magnitudes of the scores are smaller than a third preset judgment threshold, if not, if so, determining each connected domain to be detected with the grade of the score value smaller than a third preset judgment threshold value as a ulna and radius connected domain, and determining the ulna and radius connected domain as the connected domain of the preset layer.

The invention has the following beneficial effects:

(1) according to the scheme for automatically identifying the forearm ulna and radius, CT image data of the forearm ulna and radius are read in, the ulna and radius area in the middle layer of the forearm is automatically identified, then the ulna and radius area is tracked layer by layer, and the ulna and the radius are automatically classified according to medical priori knowledge.

(2) The automatic forearm ulna and radius identification and automatic forearm ulna and radius fusion diagnosis scheme provided by the invention realizes full-automatic forearm ulna and radius identification and fusion deformity identification, hardly relates to manual operation, and saves a large amount of manpower and material resources.

(3) The scheme for automatically identifying the forearm ulna radius provided by the invention can provide a basis for the diagnosis and treatment of other subsequent malformations.

(4) The method proposed by the invention has repeatability, which is crucial to the medical field. Since this gives rise to the reliability and stability of the calculation results.

(5) The invention adopts the connected domain of the local two-dimensional CT image to calculate, thereby having higher calculating speed and higher efficiency.

The embodiments in the present description are described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments are referred to each other.

The principles and embodiments of the present invention have been described herein using specific examples, which are provided only to help understand the method and the core concept of the present invention; meanwhile, for a person skilled in the art, according to the idea of the present invention, the specific embodiments and the application range may be changed. In view of the above, the present disclosure should not be construed as limiting the invention.