CN116402824B - Endocrine abnormality detection method based on children bone age X-ray film - Google Patents
Endocrine abnormality detection method based on children bone age X-ray film Download PDFInfo
- Publication number
- CN116402824B CN116402824B CN202310677468.3A CN202310677468A CN116402824B CN 116402824 B CN116402824 B CN 116402824B CN 202310677468 A CN202310677468 A CN 202310677468A CN 116402824 B CN116402824 B CN 116402824B
- Authority
- CN
- China
- Prior art keywords
- sub
- block
- gray
- radius
- pixel point
- 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.)
- Active
Links
- 210000000988 bone and bone Anatomy 0.000 title claims abstract description 35
- 230000002124 endocrine Effects 0.000 title claims abstract description 23
- 230000005856 abnormality Effects 0.000 title claims abstract description 12
- 238000001514 detection method Methods 0.000 title claims abstract description 9
- 238000000034 method Methods 0.000 claims description 20
- 230000004927 fusion Effects 0.000 claims description 6
- 238000010586 diagram Methods 0.000 claims description 2
- 230000000694 effects Effects 0.000 abstract description 9
- 230000002159 abnormal effect Effects 0.000 abstract description 3
- 210000003275 diaphysis Anatomy 0.000 description 5
- 210000002745 epiphysis Anatomy 0.000 description 4
- 230000009466 transformation Effects 0.000 description 4
- 238000004364 calculation method Methods 0.000 description 3
- 230000011218 segmentation Effects 0.000 description 3
- 238000013528 artificial neural network Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000007689 inspection Methods 0.000 description 2
- 238000013507 mapping Methods 0.000 description 2
- 238000010606 normalization Methods 0.000 description 2
- 208000017701 Endocrine disease Diseases 0.000 description 1
- 102000018997 Growth Hormone Human genes 0.000 description 1
- 108010051696 Growth Hormone Proteins 0.000 description 1
- 206010035664 Pneumonia Diseases 0.000 description 1
- AUYYCJSJGJYCDS-LBPRGKRZSA-N Thyrolar Chemical class IC1=CC(C[C@H](N)C(O)=O)=CC(I)=C1OC1=CC=C(O)C(I)=C1 AUYYCJSJGJYCDS-LBPRGKRZSA-N 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000000903 blocking effect Effects 0.000 description 1
- 230000002950 deficient Effects 0.000 description 1
- 238000003745 diagnosis Methods 0.000 description 1
- 201000010099 disease Diseases 0.000 description 1
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 1
- 208000030172 endocrine system disease Diseases 0.000 description 1
- 239000000122 growth hormone Substances 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000000877 morphologic effect Effects 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 239000005495 thyroid hormone Substances 0.000 description 1
- 229940036555 thyroid hormone Drugs 0.000 description 1
- 210000000707 wrist Anatomy 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0012—Biomedical image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/40—Image enhancement or restoration by the use of histogram techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10116—X-ray image
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30004—Biomedical image processing
- G06T2207/30008—Bone
Abstract
The invention relates to the technical field of computer vision, and provides an endocrine abnormality detection method based on children bone age X-ray films, which comprises the following steps: collecting a radius gray level map; setting an initial sub-block to obtain the side length of the initial sub-block; obtaining the importance degree of each pixel point according to the gradient amplitude of the pixel point; obtaining the sliding step length of the initial sub-block according to the importance degree of all pixel points in the initial sub-block, and sequentially obtaining a plurality of sub-blocks and a plurality of sliding step lengths according to the sliding step length of the initial sub-block to complete the traversal of the radius gray scale map; obtaining a final gray value of each pixel point according to the gray value of the pixel point in each sub-block; and obtaining the enhanced radius image and the bone age according to the final gray value, and judging whether the endocrine of the child is abnormal according to the difference between the bone age and the child age. The invention solves the boundary effect between the sub-blocks, so that the image is clearer after being locally enhanced.
Description
Technical Field
The invention relates to the technical field of computer vision, in particular to an endocrine abnormality detection method based on children bone age X-ray films.
Background
X-ray film is generally referred to as X-ray inspection, which is a clinically common auxiliary imaging inspection method at present, and belongs to electromagnetic waves, and energy can be mutually converted according to the wavelength and the frequency of the electromagnetic waves. Can be used for not only checking fracture, pneumonia, calculus and other diseases, but also checking the bone age of children.
The medical department judges whether the child has endocrine disorder by judging the fusion degree and the actual age of the radius in the child bone age X-ray film, firstly, the child bone age X-ray film image is required to be preprocessed, the radius region is extracted, the gray level map of the radius region is obtained, and the image is enhanced by an optimized local histogram equalization method, so that the epiphyseal diaphysis region in the radius region is more obvious. In the prior art, the image is enhanced by a local histogram equalization method, but in the result of the local histogram equalization, as the original image is divided into different sub-blocks, boundary effects are easy to generate in the sub-blocks and the sub-blocks, and great interference is caused to the judgment of the bone age in the enhanced image.
The method comprises the steps of firstly carrying out segmentation treatment on an image to obtain a radius region. And then carrying out image enhancement by an optimized local histogram equalization method. Finally, the enhanced image is subjected to radius age judgment according to the morphological characteristics of the epiphysis and diaphysis of the radius at different periods and the fusion characteristics of the epiphysis and the diaphysis, and the endocrine condition of the child is judged according to the difference value between the child age and the bone age.
Disclosure of Invention
The invention provides an endocrine abnormality detection method based on children bone age X-ray films, which aims to solve the problem that boundary effects are easy to generate in sub-blocks and sub-blocks in the result of local histogram equalization because the original image is divided into different sub-blocks, and the method greatly interferes with the judgment of bone age in an enhanced image, and adopts the following specific technical scheme:
one embodiment of the invention provides an endocrine abnormality detection method based on children bone age X-ray films, which comprises the following steps:
collecting a radius gray level map;
setting an initial sub-block in the radius gray scale map, and obtaining the side length of the initial sub-block according to the standard deviation of the pixel point gray scale values in the radius gray scale map and the contrast and the side length of the radius gray scale map;
obtaining the gradient amplitude of each pixel in the radius gray scale map, and obtaining the importance degree of each pixel according to the gradient amplitude of each pixel and the gradient amplitude of the pixel in the eight neighborhood of the pixel;
obtaining the sliding step length of the initial sub-block according to the importance degree of all the pixel points in the initial sub-block, marking the sub-block obtained after the initial sub-block slides as a second sub-block, obtaining the sliding step length of the second sub-block according to the importance degree of all the pixel points in the second sub-block, marking the sub-block obtained after the second sub-block slides as a third sub-block, obtaining the sliding step length of the third sub-block according to the importance degree of all the pixel points in the third sub-block, and completing the radius gray scale map traversal by analogy;
equalizing all sub-block pixel points obtained after the initial sub-block sliding respectively to obtain a plurality of equalized gray values corresponding to each pixel point, and taking the average value of the plurality of equalized gray values corresponding to each pixel point as the final gray value of each pixel point;
and obtaining an enhanced radius image according to the final gray value, comparing the enhanced radius image with a fusion grade diagram of the radius to obtain a radius grade, obtaining the bone age according to the radius grade, and judging whether the endocrine of the child is abnormal according to the difference between the bone age and the child age.
Preferably, the method for obtaining the side length of the initial sub-block according to the standard deviation of the pixel gray value in the radius gray scale image, the contrast and the side length of the radius gray scale image comprises the following steps:
in (1) the->Contrast representing radial gray-scale map, +.>Standard deviation of gray value of pixel point in gray map of radius +.>Side length of gray scale of radius, +.>For the side length of the initial sub-block, +.>Representing preset superparameter->Is an exponential function with a base of natural constant.
Preferably, the method for obtaining the importance degree of each pixel point according to the gradient amplitude of each pixel point and the gradient amplitude of the pixel points in the eight neighborhoods of the pixel point comprises the following steps:
each pixel is marked as a central pixel, and is usedThe operator obtains the gradient amplitude of each pixel point, obtains the average value of the gradient amplitudes of all the pixel points in the eight neighborhood of the central pixel point, obtains the maximum value of the gradient amplitude in the radius gray scale map, marks the ratio of the gradient amplitude of the central pixel point to the maximum value of the gradient amplitude as a first ratio, and enables all the pixels in the eight neighborhood of the central pixel point to beThe ratio of the average value of the gradient amplitude values of the points to the maximum value of the gradient amplitude values is recorded as a second ratio, and the product of the first ratio and the second ratio is used as the importance degree of the central pixel point, namely the importance degree of each pixel point.
Preferably, the method for obtaining the sliding step length comprises the following steps:
in (1) the->For the side length of the initial sub-block, +.>For the kth pixel point in the c-th sub-block, is->The sliding step of the c-th sub-block.
Preferably, the method for obtaining the final gray value of each pixel point includes:
the initial sub-blocks and the sub-blocks obtained after sliding are collectively called gray sub-blocks, histogram equalization is carried out on the gray sub-blocks to obtain gray values of pixel points in each gray sub-block, different gray values exist in different gray sub-blocks of each pixel point for the pixel points in the radius gray map, the gray values of each pixel point in different gray sub-blocks are counted, and the average value of the gray values of each pixel point in different gray sub-blocks is taken as the final gray value of the pixel point.
The beneficial effects of the invention are as follows: the invention defines a rectangular sub-block in the X-ray film radius region image, utilizes the histogram information of the sub-block to carry out histogram equalization on pixels in the center of the sub-block, moves the center of the sub-block by a certain distance of the sub-block size every time, and finally calculates the optimal gray value of each pixel to output. The boundary effect between the sub-blocks is solved, so that the image is clearer after being locally enhanced.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions of the prior art, the drawings which are used in the description of the embodiments or the prior art will be briefly described, it being obvious that the drawings in the description below are only some embodiments of the invention, and that other drawings can be obtained according to these drawings without inventive faculty for a person skilled in the art.
FIG. 1 is a flow chart of an endocrine abnormality detection method based on a child bone age X-ray film according to an embodiment of the present invention;
fig. 2 is a radius fusion grade.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
The embodiment uses an image processing technology to obtain a radius region by processing the obtained left hand X-ray film image of the child. The optimized local histogram equalization is performed on the child bone age diagnosis method, and the optimized local histogram equalization solves the mutual influence among the sub-blocks, so that the enhancement effect of the clear images of the epiphysis and the diaphysis area is better, doctors are better helped to judge the child bone age, and the endocrine condition is diagnosed.
Referring to fig. 1, a flowchart of a method for detecting endocrine abnormality based on a children bone age X-ray film according to an embodiment of the invention is shown, the method includes the following steps:
and S001, acquiring an X-ray image, and extracting a radius gray scale image by using a neural network.
In this embodiment, the image of the radius region is obtained by using the left-hand X-ray film image, so that it is necessary to extract the radius first and remove the region outside the radius.
Specifically, in this embodiment, a DNN (deep neural network) semantic segmentation method is used to segment the radius in the left-hand X-ray film image, the left-hand X-ray film image of a child in a hospital is collected as a data set to perform semantic segmentation, and the X-ray data set is manually marked, where the pixels to be segmented in the network are two types, namely, a background type and a wrist type, the pixels at the corresponding positions belong to the background type (such as phalanges, metacarps, etc.), the marks belong to the radius type (such as the epiphysis of radius and diaphysis) are 1, and the task of the network is classification, so that a cross entropy loss function is adopted as a loss function of the network. The acquired radius gray scale image is square, and the gray scale values of the rest pixel points in the image are 255 except the radius gray scale image area.
Thus, a segmented radial gray scale map is obtained.
Step S002, setting an initial sub-block, obtaining the side length of the initial sub-block according to the standard deviation and the contrast of the radius gray level image, and obtaining the importance degree of the pixel points according to the gradient amplitude of the pixel points in the radius gray level image.
When local histogram equalization is adopted, an image to be enhanced is divided into non-overlapping sub-blocks, at this time, the enhancement is very easy to cause a block effect, the generation of the block effect is caused by different transformation functions due to the histogram distribution difference of the sub-blocks, the histogram distribution difference of the sub-blocks with different sizes is different, the smaller the sub-block is, the larger the gray level distribution difference of pixel points between the sub-blocks is, the larger the histogram distribution difference of the sub-blocks is, so that the transformation function difference is larger, and serious block effect is caused, and therefore, the size of an initial sub-block is determined according to the gray level distribution of the pixel points.
Specifically, a size is set at the upper left corner of the radius gray scale mapThe sub-block of size is denoted as an initial sub-block, where the radial gray scale map is described in this embodiment as having a length and width of 2048, and for ease of calculation, the initial sub-block size is typically selected as a result of dividing the size of the image by a multiple of 2, where the image characteristics of the radial image need to be considered in selecting the initial sub-block size, and where texture in the image may typically help determine the sub-block size. Large and largeTexture images require smaller sub-block sizes, while images with less detail may use larger sub-block sizes, lower contrast images require smaller sub-block sizes, and high contrast images may use larger sub-block sizes. Therefore, the contrast of the radius gray level map is calculated and recorded as C, the contrast is calculated as the existing algorithm, and not described in detail herein, all the radius gray level maps are obtained in a hospital database, the contrast of all the historical radius gray level maps is calculated, and the overall normalization method is comprehensively used for all the historical radius gray level maps and the currently detected radius gray level maps, so that the contrast normalization of the current radius gray level map is completed; calculating standard deviation of pixel gray values in the radius gray map, wherein the smaller the standard deviation is, the more obvious texture features are not existed in the radius gray map, the larger the standard deviation is, the more possible the texture features of the radius gray map are, the calculation of the standard deviation is a known technology, and the side length of an initial sub-block is obtained according to the contrast ratio, the standard deviation and the side length of the radius gray map, and the formula is as follows:
in (1) the->Contrast representing radial gray-scale map, +.>Standard deviation of gray value of pixel point in gray map of radius +.>Side length of gray scale of radius, +.>For the side length of the initial sub-block, +.>Representing hyper-parameters, the sub-blocks being radial images in the empirical case +.>Multiple of (I)>The empirical value is between 0.05 and 0.15, the example is not particularly limited, and ++>Is an exponential function with a base of natural constant. The greater the image contrast, the more texture features, the greater the initial selection of sub-blocks.
The side length of the initial sub-block is obtained, and it is worth noting that the initial sub-block size is usually obtained by dividing the size of the image by a multiple of 2, and when the size of the initial sub-block is not satisfied, the radius gray scale map is expanded upwards and to the right until the width of the radius gray scale map can be divided by the width of the initial sub-block.
The higher the detail level in the sub-block in the radius gray map, the more abundant the detail of the region is, the smaller the sliding step length of the sub-block is, the more the number of times of repeated calculation of pixel points in the important region with abundant detail is, the smaller the probability of blocking effect after enhancement is, and meanwhile, the enhancement effect can be optimized, and over enhancement is avoided.
Specifically, if the gradient amplitude of the pixel is larger, it is indicated that the difference between the pixel and surrounding pixels is larger, which means that the pixel is more likely to belong to the detail region, so the gradient amplitude of each pixel is calculated, and the gradient amplitude of each pixel is utilizedThe operator calculates the gradient of the pixel point in the radius gray level image, and the gradient of the ith pixel point in the x direction and the y direction is set as +.>And->The gradient amplitude of the ith pixel point is obtained according to the two direction gradients:
in (1) the->Indicate->Gradient amplitude of each pixel, +.>，/>Respectively represent +.>Individual pixel dot->，/>The larger the gradient magnitude of a pixel, the more likely the pixel belongs to the detail region.
If the gradient amplitude of a pixel is larger, but the gradient amplitude of the pixel in the field is smaller, the pixel is more likely to be a noise point, and the gradient amplitude of some pixels is smaller, but the gradient amplitude of the pixel in the field is larger, that is, the more likely the pixel is a pixel in a detail area, the importance degree of the ith pixel is calculated based on the gradient amplitude, and the formula is as follows:
in (1) the->Indicate->The gradient magnitude of the individual pixels,indicate->Eighth +.>Gradient amplitude of each pixel, +.>Representing the maximum value of the gradient amplitude in the radial gray scale map, < >>Indicate->The importance degree of each pixel point, wherein the larger the gradient amplitude of the ith pixel point is, the larger the average gradient amplitude of the pixel points in the field is, the larger the importance degree of the pixel point is, and the pixel point needs to be repeatedly calculated for a plurality of times, namely the sliding step length of the sub-block of the corresponding region of the pixel point is smaller.
Step S003, traversing the radius gray level map to obtain a final gray level value according to the plurality of sub-blocks obtained by the initial sub-block and the sliding step length, and obtaining the enhanced radius image according to the final gray level value.
The larger the average importance of the pixel points in the adjacent sub-blocks of the initial sub-block, the more details in the adjacent sub-blocks, the smaller the corresponding sliding sub-block step length.
The method comprises the steps of sliding an initial sub-block from left to right, sliding the initial sub-block from top to bottom, moving the initial sub-block downwards by one step length after each row of sliding is finished, returning to the leftmost side of a radius gray scale image, continuing sliding, obtaining the sliding step length of the initial sub-block according to the importance degree of pixel points in the initial sub-block, marking the initial sub-block as a second sub-block after the initial sub-block slides once, calculating the sliding step length of the second sub-block, marking the sub-block after the second sub-block slides as a third sub-block, and so on until the radius gray scale image is traversed, wherein the sliding step length is obtained in the following manner:
in (1) the->For the side length of the initial sub-block, +.>For the importance of the kth pixel point in the c-th sub-block, < >>The sliding step of the c-th sub-block is represented by the initial sub-block when c is 1, and the higher the importance of each sub-block is, the higher the detail of the sub-block is, the smaller the sliding step of the sliding sub-block is.
It should be noted that if the sub-block exceeds the image in the sliding process, the last column of the sub-block is the same as the last column of the radius gray-scale image.
Each time a sub-block slides, a sub-block image is obtained, histogram equalization operation is performed on the sub-block image, the histogram equalization operation is not summarized in detail herein, all sub-block images are subjected to histogram equalization operation, when all sub-block histogram equalization is completed, the number of times that all pixels are equalized is counted because part of the number of times that all pixels are equalized is more than once, and the counted number of times is divided by the accumulated result of histogram equalization, so that a mapping result of a transformation function of each pixel is obtained.
The final result of the mapping after pixel equalization is:
in (1) the->For the final gray value after the ith pixel point is the ith equalization, +.>For the final gray value result of the i-th pixel point mapped by the transformation function,/and/or>Indicating the number of times the ith pixel is equalized. And (5) recording an image formed by the final gray value of each pixel point as an enhanced radius image.
Thus, an enhanced image of the posterior radius is obtained.
Step S004, obtaining the bone age of the child according to the enhanced radius image, and detecting endocrine abnormality according to the difference of the bone age and the age of the child.
The biological age of a person is bone age, the physiological age is actual age, if the bone age is less than 2 years old, the endocrine insufficiency of the child is predicted, and the growth hormone or thyroid hormone is deficient in the body; if the bone age is greater than the age of 2 years, the endocrine of the children is predicted to be too high; if the bone age is greater than 1.3 years old, the endocrine of the children is slightly low, so that the children can intervene as early as possible; if the bone age is less than 1.3 years old, the endocrine of the children is slightly high; the fact that the bone age and the actual age of the children are different and not more than one year old indicates that the endocrine of the children is normal. As shown in fig. 2, the radius fusion grade is shown, the epiphyseal soft plate is observed by using the enhanced radius image to obtain the corresponding radius grade, and the corresponding bone ages of different radius grades are searched in the radius age grade table, wherein the radius age grade table is as follows:
checking the bone age x corresponding to the grade, setting the actual age of the child as y, and setting the difference value between the bone age and the actual age as z, and then:
if z fluctuates between-1.3 and +1.3, the endocrine of the child is normal, and if greater or less than this interval, the endocrine of the child is abnormal.
The foregoing description of the preferred embodiments of the invention is not intended to be limiting, but rather is intended to cover all modifications, equivalents, alternatives, and improvements that fall within the spirit and scope of the invention.
Claims (2)
1. The endocrine abnormality detection method based on the children bone age X-ray film is characterized by comprising the following steps of:
collecting a radius gray level map;
setting an initial sub-block in the radius gray scale map, and obtaining the side length of the initial sub-block according to the standard deviation of the pixel point gray scale values in the radius gray scale map and the contrast and the side length of the radius gray scale map;
obtaining the gradient amplitude of each pixel in the radius gray scale map, and obtaining the importance degree of each pixel according to the gradient amplitude of each pixel and the gradient amplitude of the pixel in the eight neighborhood of the pixel;
obtaining the sliding step length of the initial sub-block according to the importance degree of all the pixel points in the initial sub-block, marking the sub-block obtained after the initial sub-block slides as a second sub-block, obtaining the sliding step length of the second sub-block according to the importance degree of all the pixel points in the second sub-block, marking the sub-block obtained after the second sub-block slides as a third sub-block, obtaining the sliding step length of the third sub-block according to the importance degree of all the pixel points in the third sub-block, and completing the radius gray scale map traversal by analogy;
equalizing all sub-block pixel points obtained after the initial sub-block slides respectively to obtain the equalized gray value of each pixel point for a plurality of times, and taking the average value of the equalized gray values of each pixel point for a plurality of times as the final gray value of each pixel point;
obtaining an enhanced radius image according to the final gray value, comparing the enhanced radius image with a fusion grade diagram of the radius to obtain a radius grade, obtaining bone age according to the radius grade, and judging whether the child has endocrine abnormality according to the difference of the bone age and the child age;
the method for obtaining the side length of the initial sub-block according to the standard deviation of the pixel point gray value in the radius gray scale image, the contrast and the side length of the radius gray scale image comprises the following steps:
in (1) the->Contrast representing radial gray-scale map, +.>Standard deviation of gray value of pixel point in gray map of radius +.>Side length of gray scale of radius, +.>For the side length of the initial sub-block, +.>Representing preset superparameter->Is an exponential function with a natural constant as a base;
the method for obtaining the importance degree of each pixel point according to the gradient amplitude of each pixel point and the gradient amplitude of the pixel points in eight neighborhoods of the pixel point comprises the following steps:
each pixel is marked as a central pixel, and is usedThe operator obtains the gradient amplitude value of each pixel point, obtains the average value of the gradient amplitude values of all the pixel points in the eight neighborhood of the central pixel point, obtains the maximum value of the gradient amplitude values in the radius gray scale map, marks the ratio of the gradient amplitude value of the central pixel point to the maximum value of the gradient amplitude value as a first ratio, marks the ratio of the average value of the gradient amplitude values of all the pixel points in the eight neighborhood of the central pixel point to the maximum value of the gradient amplitude value as a second ratio, and marks the product of the first ratio and the second ratio as the importance degree of the central pixel point, namely the importance degree of each pixel point;
the sliding step length obtaining method comprises the following steps:
in (1) the->For the side length of the initial sub-block, +.>For the importance of the kth pixel point in the c-th sub-block, < >>The sliding step of the c-th sub-block.
2. The endocrine abnormality detection method based on the children bone age X-ray film according to claim 1, wherein the method for obtaining the final gray value of each pixel point is as follows:
the initial sub-blocks and the sub-blocks obtained after sliding are collectively called gray sub-blocks, histogram equalization is carried out on the gray sub-blocks to obtain gray values of pixel points in each gray sub-block, different gray values exist in different gray sub-blocks of each pixel point for the pixel points in the radius gray map, the gray values of each pixel point in different gray sub-blocks are counted, and the average value of the gray values of each pixel point in different gray sub-blocks is taken as the final gray value of the pixel point.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310677468.3A CN116402824B (en) | 2023-06-09 | 2023-06-09 | Endocrine abnormality detection method based on children bone age X-ray film |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310677468.3A CN116402824B (en) | 2023-06-09 | 2023-06-09 | Endocrine abnormality detection method based on children bone age X-ray film |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116402824A CN116402824A (en) | 2023-07-07 |
| CN116402824B true CN116402824B (en) | 2023-10-03 |
Family
ID=87012796
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310677468.3A Active CN116402824B (en) | 2023-06-09 | 2023-06-09 | Endocrine abnormality detection method based on children bone age X-ray film |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116402824B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116843692B (en) * | 2023-09-04 | 2023-12-01 | 青岛冠宝林活性炭有限公司 | Regenerated active carbon state detection method based on artificial intelligence |
| CN117437236B (en) * | 2023-12-21 | 2024-03-01 | 中国人民解放军总医院第一医学中心 | Gastrointestinal mirror image anomaly detection method and system |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110223772A (en) * | 2019-06-10 | 2019-09-10 | 青岛大学附属医院 | A kind for the treatment of endocrine diseases diagnostic device and data information processing method |
| CN114601483A (en) * | 2022-05-11 | 2022-06-10 | 山东第一医科大学第一附属医院(山东省千佛山医院) | Bone age analysis method and system based on image processing |
| CN116027018A (en) * | 2021-10-26 | 2023-04-28 | 孙雅军 | Children endocrine detects blood sugar with intelligent blood glucose meter that has test paper storage mechanism |
| CN116152505A (en) * | 2023-04-20 | 2023-05-23 | 深圳市亿康医疗技术有限公司 | Bone target identification and segmentation method based on X-ray data |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2016196687A1 (en) * | 2015-06-01 | 2016-12-08 | Massachusetts Institute Of Technology | System and method for neuroendocrine control |
-
2023
- 2023-06-09 CN CN202310677468.3A patent/CN116402824B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110223772A (en) * | 2019-06-10 | 2019-09-10 | 青岛大学附属医院 | A kind for the treatment of endocrine diseases diagnostic device and data information processing method |
| CN116027018A (en) * | 2021-10-26 | 2023-04-28 | 孙雅军 | Children endocrine detects blood sugar with intelligent blood glucose meter that has test paper storage mechanism |
| CN114601483A (en) * | 2022-05-11 | 2022-06-10 | 山东第一医科大学第一附属医院(山东省千佛山医院) | Bone age analysis method and system based on image processing |
| CN116152505A (en) * | 2023-04-20 | 2023-05-23 | 深圳市亿康医疗技术有限公司 | Bone target identification and segmentation method based on X-ray data |
Non-Patent Citations (1)
| Title |
|---|
| 骨龄测试在儿童生长发育中的应用分析;闫荣;《中国农村卫生》;第2021年卷(第06期);42，46 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116402824A (en) | 2023-07-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN116402824B (en) | Endocrine abnormality detection method based on children bone age X-ray film | |
| CN111798462B (en) | Automatic delineation method of nasopharyngeal carcinoma radiotherapy target area based on CT image | |
| CN108898160B (en) | Breast cancer histopathology grading method based on CNN and imaging omics feature fusion | |
| CN109636808B (en) | Lung lobe segmentation method based on full convolution neural network | |
| JP5941674B2 (en) | Cell contour forming apparatus and method, and cell contour forming program | |
| CN112184617B (en) | Spine MRI image key point detection method based on deep learning | |
| CN112488234B (en) | End-to-end histopathology image classification method based on attention pooling | |
| CN112150428A (en) | Medical image segmentation method based on deep learning | |
| Liu et al. | A fully automatic segmentation algorithm for CT lung images based on random forest | |
| CN111681230A (en) | System and method for scoring high-signal of white matter of brain | |
| CN113223005B (en) | Thyroid nodule automatic segmentation and grading intelligent system | |
| US20230005140A1 (en) | Automated detection of tumors based on image processing | |
| CN112308822A (en) | Intervertebral disc CT image detection method based on deep convolutional neural network | |
| CN112150477B (en) | Full-automatic segmentation method and device for cerebral image artery | |
| Kuang et al. | Skull R-CNN: A CNN-based network for the skull fracture detection | |
| CN117291843B (en) | Efficient management method for image database | |
| Sharma et al. | Review Paper on Brain Tumor Detection Using Pattern Recognition Techniques. | |
| CN114926396A (en) | Mental disorder magnetic resonance image preliminary screening model construction method | |
| Jaware et al. | Automatic Segmentation of Infant Brain MRI Using Soft Computing Techniques | |
| CN116758068B (en) | Marrow picture cell morphology analysis method based on artificial intelligence | |
| CN111369490A (en) | Method for judging osteoporosis by extracting bone X-ray image characteristics | |
| Mahajan et al. | Enhanced Optimized MSVM Classification of Lung Cancer in Biomedical CT Image Database | |
| CN115239688B (en) | Brain metastasis recognition method and system based on magnetic resonance contrast enhancement 3D-T1WI image | |
| CN113627231B (en) | Automatic segmentation method for liquid region in retina OCT image based on machine vision | |
| Hwang et al. | RBCA-Net: Reverse boundary channel attention network for kidney tumor segmentation in CT images |
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 |