CN110782450B - Hand carpal development grade determining method and related equipment - Google Patents
Hand carpal development grade determining method and related equipment Download PDFInfo
- Publication number
- CN110782450B CN110782450B CN201911053737.9A CN201911053737A CN110782450B CN 110782450 B CN110782450 B CN 110782450B CN 201911053737 A CN201911053737 A CN 201911053737A CN 110782450 B CN110782450 B CN 110782450B
- Authority
- CN
- China
- Prior art keywords
- hand
- bone
- development level
- carpal
- development
- 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
Images
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
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/18—Complex mathematical operations for evaluating statistical data, e.g. average values, frequency distributions, probability functions, regression analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/20—Analysis of motion
- G06T7/246—Analysis of motion using feature-based methods, e.g. the tracking of corners or segments
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Mathematical Physics (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Computational Mathematics (AREA)
- Computing Systems (AREA)
- Mathematical Optimization (AREA)
- Mathematical Analysis (AREA)
- Molecular Biology (AREA)
- Pure & Applied Mathematics (AREA)
- Evolutionary Computation (AREA)
- Computational Linguistics (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Probability & Statistics with Applications (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Databases & Information Systems (AREA)
- Evolutionary Biology (AREA)
- Algebra (AREA)
- Multimedia (AREA)
- Bioinformatics & Computational Biology (AREA)
- Operations Research (AREA)
- Radiology & Medical Imaging (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Medical Informatics (AREA)
- Quality & Reliability (AREA)
- Image Analysis (AREA)
- Apparatus For Radiation Diagnosis (AREA)
Abstract
The invention discloses a hand carpal development grade determining method and related equipment, which can obtain a characteristic diagram of a first hand bone image; inputting the feature map into a preset hand carpal development level determination model to obtain a hand carpal point position probability map and a hand carpal point development level probability map output by the hand carpal development level determination model; and determining the development level of the hand wrist bone point according to the position probability map of the hand wrist bone point and the development level probability map of the hand wrist bone point. The invention determines the probability that each position in the hand skeleton image is the position of the carpal bone point and determines the probability of the development grade of each position through the hand carpal bone development grade determination model, thereby determining the technical means of the development grade of each hand carpal bone point in the hand skeleton image, overcoming the technical problem of low reliability of the existing bone age prediction method, and further achieving the technical effect of scientifically determining the bone age.
Description
Technical Field
The invention relates to the field of image processing, in particular to a hand carpal development grade determining method and related equipment.
Background
Today, a user can determine the biological age of a customer by calculating the bone age of the customer. Bone age has application value in many fields. For example: a height prediction can be made for a client in conjunction with the bone age of the client. Meanwhile, the bone age also has important reference value in the aspects of athlete selection, judicial judgment and the like.
The prior art can predict bone age in hand bone images by a simple regression analysis method. However, the regression analysis method only predicts the bone age of the hand bone image by using the data fitting capability of machine learning, and the bone age prediction result obtained in actual use has low reliability.
Disclosure of Invention
In view of the above problems, the present invention provides a hand carpal development level determination method and related device, which overcomes or at least partially solves the above problems, and the technical solution is as follows:
a hand carpal development level determination method comprises the following steps:
obtaining a feature map of a first hand bone image;
inputting the feature map into a preset hand carpal development level determination model to obtain a hand carpal point position probability map and a hand carpal point development level probability map output by the hand carpal development level determination model;
and determining the development level of the hand wrist bone point according to the position probability map of the hand wrist bone point and the development level probability map of the hand wrist bone point.
Optionally, the inputting the feature map into a preset hand carpal development level determination model to obtain a position probability map of a hand carpal site and a development level probability map of the hand carpal site output by the hand carpal development level determination model includes:
inputting the feature map into a preset hand carpal development level determination model, so that the preset hand carpal development level determination model convolves the feature map according to a preset first convolution parameter to obtain a position probability map of a first number of hand carpal points, and convolving the feature map according to a preset second convolution parameter by the preset hand carpal development level determination model to obtain a development level probability map of a second number of hand carpal points.
Optionally, the position probability map of each hand wrist bone point in the position probability maps of the hand wrist bone points of the first number corresponds to one hand wrist bone point, the hand wrist bone points corresponding to the position probability maps of the hand wrist bone points are different, any one of the development level probability maps of the hand wrist bone points corresponds to one information combination, the information combinations corresponding to the development level probability maps of the hand wrist bone points are different, and the information combination is composed of one hand wrist bone point and one development level.
Optionally, the determining the development level of the wrist bone point according to the position probability map of the wrist bone point and the development level probability map of the wrist bone point includes:
for any hand wrist bone point: multiplying the position probability maps of the hand wrist bone points with the development level probability maps corresponding to the hand wrist bone points respectively to obtain a third number of development level probability results of the hand wrist bone points, wherein the third number is a quotient obtained by dividing the second number by the first number;
and determining the development grade with the highest probability in the third number of development grade probability results as the development grade of the wrist bone point of the hand.
Optionally, the first number is 7, and the first volume parameter includes: the convolution kernel size is 3 × 3 and the number of output channels is 7;
the second number is 63, the third number is 9, and the second convolution parameters include: the convolution kernel size is 3 × 3 and the number of output channels is 63.
Optionally, after determining the development level of the wrist bone point according to the position probability map of the wrist bone point and the development level probability map of the wrist bone point, the method further includes:
and determining the bone age according to the determined development grade of each hand wrist bone point.
Optionally, the obtaining a feature map of the first hand bone image includes:
obtaining an initial map of a first hand bone image at a plurality of different scales;
and fusing the initial images under the different scales to obtain a characteristic image.
Optionally, the fusing the initial maps at the multiple different scales to obtain a feature map includes:
the initial images under the different scales are up-sampled according to the first up-sampling parameters, and a fourth number of matrix images are obtained, wherein the scales of the fourth number of matrix images are the same;
element summation is carried out on the fourth quantity of matrix diagrams to obtain element diagrams;
and upsampling the element graph according to a second upsampling parameter to obtain a characteristic graph.
A hand carpal development level determination device, comprising: a characteristic map obtaining unit, a probability map obtaining unit and a development level determining unit,
the characteristic map obtaining unit is used for obtaining a characteristic map of a first hand bone image;
the probability map obtaining unit is used for inputting the feature map into a preset hand carpal development level determination model to obtain a position probability map of a hand carpal point and a development level probability map of the hand carpal point output by the hand carpal development level determination model;
the development level determining unit is used for determining the development level of the wrist bone point according to the position probability map of the wrist bone point and the development level probability map of the wrist bone point.
Optionally, the probability map obtaining unit is specifically configured to input the feature map into a preset hand carpal development level determination model, so that the preset hand carpal development level determination model convolves the feature map according to a preset first convolution parameter to obtain a position probability map of a first number of hand carpal points, and the preset hand carpal development level determination model convolves the feature map according to a preset second convolution parameter to obtain a development level probability map of a second number of hand carpal points.
Optionally, the position probability map of each hand wrist bone point in the position probability maps of the hand wrist bone points of the first number corresponds to one hand wrist bone point, the hand wrist bone points corresponding to the position probability maps of the hand wrist bone points are different, any one of the development level probability maps of the hand wrist bone points corresponds to one information combination, the information combinations corresponding to the development level probability maps of the hand wrist bone points are different, and the information combination is composed of one hand wrist bone point and one development level.
Optionally, the development level determining unit includes: and obtaining a development level probability result subunit and a development level determining subunit.
The development level probability result obtaining subunit is used for obtaining the probability of any hand wrist bone point: multiplying the position probability maps of the hand wrist bone points with the development level probability maps corresponding to the hand wrist bone points respectively to obtain a third number of development level probability results of the hand wrist bone points, wherein the third number is a quotient obtained by dividing the second number by the first number.
And the development level determining subunit is configured to determine the development level with the highest probability in the third number of development level probability results as the development level of the wrist bone point of the hand.
Optionally, the apparatus further comprises: a bone age determination unit for determining the age of a bone,
and the bone age determining unit is used for determining the bone age according to the determined development grade of each hand wrist bone point.
A storage medium having stored thereon computer-executable instructions which, when loaded and executed by a processor, carry out a method of determining a level of development of a wrist bone of a hand as claimed in any preceding claim.
A computer device comprising a processor, a memory and a program stored on the memory and executable on the processor, the processor when executing the program at least implementing a hand carpal development level determination method as claimed in any preceding claim.
A hand carpal development level determination device comprising: a hand skeleton image receiving device, a processor, a memory, a communication bus, and an output device, the memory having stored thereon a program executable on the processor,
the processor is respectively in communication connection with the hand skeleton image receiving device, the memory and the output device through the communication bus;
the hand bone image receiving device receives a first hand bone image;
the processor at least realizes the hand carpal development level determination method as any one of the above when executing the program;
and the output device obtains and outputs the development grade of the hand wrist bone point determined by the processor.
By means of the technical scheme, the hand carpal development grade determining method and the related equipment provided by the invention can obtain the characteristic diagram of the first hand bone image; inputting the feature map into a preset hand carpal development level determination model to obtain a hand carpal point position probability map and a hand carpal point development level probability map output by the hand carpal development level determination model; and determining the development level of the hand wrist bone point according to the position probability map of the hand wrist bone point and the development level probability map of the hand wrist bone point. The invention determines the probability that each position in the hand skeleton image is the position of the carpal bone point and determines the probability of the development grade of each position through the hand carpal bone development grade determination model, thereby determining the technical means of the development grade of each hand carpal bone point in the hand skeleton image, overcoming the technical problem of low reliability of the existing bone age prediction method, and further achieving the technical effect of scientifically determining the bone age.
The foregoing description is only an overview of the technical solutions of the present invention, and the embodiments of the present invention are described below in order to make the technical means of the present invention more clearly understood and to make the above and other objects, features, and advantages of the present invention more clearly understandable.
Drawings
Various other advantages and benefits will become apparent to those of ordinary skill in the art upon reading the following detailed description of the preferred embodiments. The drawings are only for purposes of illustrating the preferred embodiments and are not to be construed as limiting the invention. Also, like reference numerals are used to refer to like parts throughout the drawings. In the drawings:
FIG. 1 is a flow chart of a hand carpal development level determination method according to an embodiment of the present invention;
FIG. 2 illustrates a first hand bone image provided by an embodiment of the present invention;
FIG. 3 is a flow chart of another hand carpal development level determination method provided by the embodiment of the invention;
FIG. 4 is a flow chart of another hand carpal development level determination method provided by the embodiment of the invention;
FIG. 5 is an illustrative diagram of obtaining an elemental map from a matrix map according to an embodiment of the invention;
FIG. 6 is a schematic flow chart illustrating a method for obtaining a hand carpal development level determination model according to an embodiment of the present invention;
FIG. 7 is a flow chart of another hand carpal development level determination method provided by the embodiment of the invention;
FIG. 8 is a graph illustrating a probability of location of a wrist bone point according to an embodiment of the present invention;
FIG. 9 is a chart illustrating a probability of development level of a wrist bone point according to an embodiment of the present invention;
FIG. 10 is a flow chart of another hand carpal development level determination method provided by the embodiment of the present invention;
FIG. 11 is a schematic diagram illustrating a probability result of the development grade of the wrist bone point of the hand according to an embodiment of the present invention;
FIG. 12 is a flow chart illustrating another hand carpal development level determination method provided by an embodiment of the present invention;
FIG. 13 is a schematic diagram illustrating a second image of a hand bone provided by an embodiment of the invention;
FIG. 14 is a schematic structural diagram of a hand carpal development level determination device according to an embodiment of the present invention;
FIG. 15 is a schematic structural diagram of another hand carpal development level determination device provided in an embodiment of the present invention;
FIG. 16 is a schematic view of the extended hand skeleton image holding box of the hand carpal development level determination device provided by the embodiment of the present invention;
FIG. 17 is a schematic diagram illustrating retraction of a hand bone image pod of a hand carpal development level determination device in accordance with an embodiment of the present invention;
fig. 18 is a schematic connection diagram illustrating a processor, a memory and a communication bus of a hand carpal development level determination device according to an embodiment of the present invention.
Detailed Description
Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be embodied in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.
As shown in fig. 1, a method for determining a development grade of a hand carpal bone according to an embodiment of the present invention may include:
and S100, obtaining a characteristic diagram of the first hand bone image.
Specifically, as shown in fig. 2, the first hand skeleton image may be a hand skeleton image obtained by X-ray penetrating irradiation. The scale may be the image size of the image at a certain resolution. The first hand bone image may have a scale of an X-ray image output from an X-ray machine. For example, the resolution corresponding to the scale of the first hand bone image output by the X-ray machine may be 512 × 512. Alternatively, the first hand bone image may be a left hand bone image. The first hand bone image may display a part or all of the carpal bones including the scaphoid, the lunate, the trigonal, the trapezium, the capitate, and the hamate.
Optionally, as shown in fig. 3, in another hand carpal development level determining method provided in the embodiment of the present invention, step S100 may include:
and S110, obtaining initial images of the first hand bone image under a plurality of different scales.
According to the embodiment of the invention, the initial image of the first hand bone image can be obtained through the feature extraction network. The feature extraction network may include convolutional neural networks including Resnext-50, Resnext-101, Resnext-152, and Densenet. For example, with Resnext-50, embodiments of the present invention may obtain initial maps of 1/32, 1/16, 1/8, and 1/4 at the scale of the first hand bone image, respectively. The method and the device can obtain the image characteristics of the same characteristic point of the first hand bone image under different scales. For example, in the initial images of 1/32, 1/16, 1/8 and 1/4, which are scales of the first hand bone image, respectively, the image features of the same feature point in the initial images at four scales can be obtained. Optionally, in the embodiment of the present invention, the first hand bone image may meet the requirement of the feature extraction network by means of black margin filling, so as to prevent the feature extraction network from scaling and deforming the first hand bone image that does not meet the requirement of the scale, thereby ensuring correct implementation of the technical solution of the embodiment of the present invention.
And S120, fusing the initial images under the different scales to obtain a characteristic image.
Specifically, in the embodiment of the present invention, after feature extraction is performed on a plurality of initial images at different scales through an image feature extraction network, information such as edges, shapes, contours, local features, and the like is comprehensively processed to obtain a feature image. The image feature extraction network may include: unet, full convolution network FCN (full volumetric Networks), Panoptic FPN (panoramic Feature Pyramid Networks), and SegNet, among others. For example, based on the above example of obtaining the initial map of the first hand bone image at four scales through resenxt-50, the embodiment of the present invention may fuse the four initial maps of 1/32, 1/16, 1/8 and 1/4, which are scales of the first hand bone image, into one feature map through Panoptic FPN.
Optionally, based on the method shown in fig. 3, as shown in fig. 4, in another hand carpal development level determining method provided in the embodiment of the present invention, step S120 may include:
s121, performing upsampling on the plurality of initial images under different scales according to the first upsampling parameter to obtain a fourth number of matrix images, wherein the scales of the fourth number of matrix images are the same.
Specifically, in the embodiment of the present invention, a plurality of initial graphs under different scales may be upsampled according to different first upsampling parameters, so as to obtain a fourth number of matrix graphs. Wherein the fourth number is related to the number of initial graphs. For example, after the initial maps at four different scales are upsampled according to different first upsampling parameters, four matrix maps are obtained.
The first upsampling parameter may be determined according to a scale of the initial map. For ease of understanding, the description is made herein by way of example: in the embodiment of the present invention, if it is necessary to upsample four initial maps 1/32, 1/16, 1/8 and 1/4, which have respective dimensions of the first hand bone image, according to different first upsampling parameters, the initial map 1/32, which has the dimensions of the first hand bone image, may be upsampled according to the first upsampling parameters of 3 convolution products and 2 × bilinear, the obtained matrix map has the dimensions of 1/4 the first hand bone image, the initial map 1/16, which has the dimensions of the first hand bone image, may be upsampled according to the first upsampling parameters of 2 convolution products and 2 × bilinear, the obtained matrix map has the dimensions of 1/4 the dimensions of the first hand bone image, and the initial map 1/8, which has the dimensions of the first hand bone image, may be upsampled according to the first upsampling parameters of 1 convolution product and 2 × bilinear Sampling, wherein the obtained matrix map has the scale of 1/4 of the scale of the first hand bone image, the initial map of 1/4 having the scale of the first hand bone image can be upsampled according to the first upsampling parameter of 0 convolution and 2 x bilinear, the obtained matrix map has the scale of 1/4 of the scale of the first hand bone image, the four matrix maps have the same scale, and the obtained matrix maps are all 1/4 of the scale of the first hand bone image.
And S122, performing element summation on the matrix diagrams of the fourth quantity to obtain element diagrams.
In particular, the element summation may be adding numbers at the same position in the fourth number of matrix images. The embodiment of the invention can take the result of element summation of the fourth number of matrix diagrams as the element diagram. For ease of understanding, this is illustrated herein in connection with FIG. 5: if the fourth number of matrix maps are fig. 5(a) and 5(b), respectively, then the numbers at the same positions of fig. 5(a) and 5(b) are added, and the resulting elemental map can be as shown in fig. 5 (c). It is noted that the dimensions of the element map are the same as the dimensions of the matrix map. For ease of understanding, the matrix map and the element map shown in fig. 5 are both 3 × 3 in scale, and in practice, the element map may be 1/4 in scale of the first hand bone image.
And S123, performing upsampling on the element diagram according to the second upsampling parameter to obtain a characteristic diagram.
Specifically, the second upsampling parameter may be determined according to a scale of the first hand bone image and a scale of the elemental map. For example, when the scale of the element map is 1/4 of the scale of the first hand bone image, the embodiment of the present invention may perform upsampling on the element map according to the second upsampling parameter, which is 1 convolution and 4 × bilinear, to obtain the feature map with the same scale as the first hand bone image.
S200, inputting the feature map into a preset hand carpal development level determination model, and obtaining a position probability map of a hand carpal bone point and a development level probability map of the hand carpal bone point output by the hand carpal development level determination model.
The preset hand carpal development level determination model can be a convolutional neural network model. The preset hand carpal development level determining model can use the hand skeleton image fusion feature map marked with the hand carpal site position and the hand carpal site development level to perform machine learning, so as to learn the position of the hand carpal site and the image features of the development level of the hand carpal site. The embodiment of the invention can obtain a preset hand carpal development level determination model according to the method.
As shown in fig. 6, a method for obtaining a hand carpal development level determination model according to an embodiment of the present invention may include:
s010, obtaining a training initial image of at least one hand skeleton training image under multiple scales, wherein the hand skeleton training image is marked with a hand carpal site position and a hand carpal site growing grade;
s020, fusing the training initial images under the multiple scales to obtain a training characteristic image;
s030, performing machine learning according to the training characteristic diagram to obtain a hand carpal development level determination model, wherein the hand carpal development level determination model has the following input: and (3) obtaining training characteristic graphs by fusing initial graphs under multiple scales, wherein the output of the hand carpal development level determination model is as follows: a position probability map of the wrist bone point of the hand and a development level probability map of the wrist bone point of the hand.
Specifically, the hand carpal development level determination model performs machine learning on the training feature map, which may be the machine learning on the image features of the positions of the hand carpal points marked in the training feature map of the hand skeletal image and the hand carpal point development levels, and the image features may include image geometric features and gray value distribution. In actual use, the hand carpal development level determination model can output a position probability map of each hand carpal bone point and a development level probability map of each hand carpal bone point corresponding to the fusion feature map of the first hand skeletal image according to the hand carpal bone point obtained by machine learning and the image feature of the position of the hand carpal bone point development level.
Optionally, based on the method shown in fig. 1, as shown in fig. 7, in another hand carpal development level determining method provided in the embodiment of the present invention, step S200 may include:
s210, inputting the feature map into a preset hand carpal development level determination model, so that the preset hand carpal development level determination model convolves the feature map according to a preset first convolution parameter to obtain a position probability map of a first number of hand carpal points, and convolving the feature map according to a preset second convolution parameter to obtain a development level probability map of a second number of hand carpal points.
In particular, the first convolution parameters may include a convolution kernel size and a number of output channels. The first number is equal to the number of output channels in the first convolution parameter. For example, if the number of output channels in the first convolution parameter is 5, the hand carpal development level determination model convolves the feature map according to the first convolution parameter, and the first number of the position probability maps of the obtained hand carpal points is 5.
Normally, 7 carpal bones can be shown in the first hand bone image, and therefore, in an alternative embodiment of the invention, the first number is 7, and the first volume parameter comprises: the convolution kernel size is 3 × 3 and the number of output channels is 7.
The position probability maps of each hand wrist bone point in the position probability maps of the hand wrist bone points of the first number correspond to one hand wrist bone point, and the hand wrist bone points corresponding to the position probability maps of the hand wrist bone points are different.
Specifically, the position probability map of each hand wrist bone point in the position probability maps of the first number of hand wrist bone points is only a probability prediction of one hand wrist bone point at each position in the first hand bone image, wherein each position in the first hand bone image may be each pixel point in the first hand bone image. Therefore, when the first number is 7, the position probability maps of 7 hand wrist bone points can respectively predict the probability of 7 hand wrist bone points at each position in the first hand bone image. For example, a position probability map of a corresponding hand wrist bone point in the first hand bone image may be as shown in fig. 8, where a probability at each position in the position probability map is a probability that the position is the hand wrist bone point, and the sum of the probabilities at each position in the position probability map is equal to 1. It should be noted that, for the convenience of understanding, the scale of the position probability map shown in fig. 8 is 5 × 5, and in actual use, the scale of the position probability map of the wrist bone point of the hand may be the same as that of the first hand bone image.
Optionally, in the embodiment of the present invention, the process of convolving the feature map by the preset hand carpal development level determination model according to a preset first convolution parameter to obtain a position probability map of a first number of hand carpal points may include:
convolving the feature map according to a preset first convolution parameter by the preset hand carpal development level determining model to obtain a position information map of a first number of hand carpal points;
and normalizing the position information maps of the first number of hand wrist bone points to obtain a position probability map of the first number of hand wrist bone points.
Optionally, in the embodiment of the present invention, the position information map of the first number of hand wrist bone points may be normalized through a softmax function, so as to obtain the position probability map of the first number of hand wrist bone points.
The second convolution parameters may include convolution kernel size and number of output channels in embodiments of the present invention. The second number is equal to the number of output channels in the second convolution parameter. For example, if the number of output channels in the second convolution parameter is 8, the hand carpal development level determination model convolves the fusion feature map according to the second convolution parameter, and the second number of the development level probability maps of the obtained hand carpal points is 8.
Normally, 7 carpal bones can be shown in the first hand bone image. In the embodiment of the invention, the evaluation of the development grade of the CARPAL bones can be carried out by using the evaluation of the development grade of the CARPAL bones in the TW3-CARPAL bone age evaluation standard. TW3-CARPAL bone age evaluation criteria Each CARPAL bone can be divided into nine developmental levels, A-I. Therefore, in an alternative embodiment of the present invention, the second number is 63, and the second convolution parameter may include: the convolution kernel size is 3 × 3 and the number of output channels is 63. It will be appreciated that the second number may be equal to the product of the number of carpal development levels multiplied by the number of carpal bones in the first hand bone image. Assuming that each carpal bone can be divided into five levels according to other bone age evaluation criteria, the second number may be 35.
The development level probability map of any one hand wrist bone point corresponds to one information combination, the information combinations corresponding to the development level probability maps of the hand wrist bone points are different, and the information combination comprises one hand wrist bone point and one development level.
Specifically, the information combinations corresponding to the development level probability maps of the respective wrist bone points may be that the wrist bone points are the same but have different development levels, or the wrist bone points are different but have the same development levels, or the wrist bone points are different but have different development levels, that is, the development level probability maps of the wrist bone points predict the probability that each position in the first hand bone image is the wrist bone point position, and also predict the development level probability corresponding to each position in the first hand bone image. For example, it is assumed that fig. 9 shows a development level probability map in which the carpal bone is the scaphoid and the development level of the carpal bone is the B level, where 0.38 shows the probability that the position is both the scaphoid and the development level is the B level. It is understood that when the TW3-CARPAL bone age assessment criteria is used, a probability map of development levels for 9 different development levels of scaphoid can be obtained, with the sum of the probabilities on the probability map of development levels for 9 different development levels being equal to 1. It should be noted that, for the convenience of understanding, the development level probability map shown in fig. 9 has a scale of 5 × 5, and in actual use, the development level probability map of the wrist bone point of the hand may have the same scale as that of the first hand bone image.
Therefore, the hand carpal development level determination model can simultaneously output the position probability map of the hand carpal site and the development level probability map of the hand carpal site.
S300, determining the development level of the wrist bone point according to the position probability map of the wrist bone point and the development level probability map of the wrist bone point.
Based on the method shown in fig. 7, as shown in fig. 10, in another hand carpal development level determination method provided in the embodiment of the present invention, step S300 may include:
s310, for any wrist bone point: multiplying the position probability maps of the hand wrist bone points with the development level probability maps corresponding to the hand wrist bone points respectively to obtain a third number of development level probability results of the hand wrist bone points, wherein the third number is a quotient obtained by dividing the second number by the first number.
For ease of understanding, the description is made herein by way of example: assuming that fig. 8 is a position probability map of a certain wrist bone point and fig. 9 is a development level probability map with a development level of the certain wrist bone point being B, the development level probability result of the certain wrist bone point obtained by multiplying the position probability map of the certain wrist bone point by the development level probability map with the development level of the certain wrist bone point being B may be as shown in fig. 11.
When the first hand bone image of the embodiment of the present invention includes 7 CARPAL bones and the TW3-CARPAL age evaluation criterion is adopted, the third number may be 9, that is, the position probability map of any hand CARPAL bone point is multiplied by the 9 development level probability maps corresponding to the hand CARPAL bone point, respectively, to obtain 9 development level probability results of the hand CARPAL bone point.
And S320, determining the development grade with the highest probability in the third number of development grade probability results as the development grade of the wrist bone point of the hand.
According to the method for determining the development grade of the hand carpal bones, provided by the embodiment of the invention, a characteristic diagram of a first hand bone image can be obtained; inputting the feature map into a preset hand carpal development level determination model to obtain a hand carpal point position probability map and a hand carpal point development level probability map output by the hand carpal development level determination model; and determining the development level of the hand wrist bone point according to the position probability map of the hand wrist bone point and the development level probability map of the hand wrist bone point. The invention determines the probability that each position in the hand skeleton image is the position of the carpal bone point and determines the probability of the development grade of each position through the hand carpal bone development grade determination model, thereby determining the technical means of the development grade of each hand carpal bone point in the hand skeleton image, overcoming the technical problem of low reliability of the existing bone age prediction method, and further achieving the technical effect of scientifically determining the bone age.
Optionally, as shown in fig. 12, in another hand carpal development level determining method provided in the embodiment of the present invention, after step S300, the method further includes:
and S400, determining the bone age according to the determined development grade of each hand wrist bone point.
According to the embodiment of the invention, the values corresponding to the development levels of the wrist bone points of the hands can be evaluated according to the TW3-CARPAL bone age evaluation standard according to the development levels of the wrist bone points of the hands of the first bone image, and the values corresponding to the development levels of the wrist bone points of the hands are added to determine the calculation result, wherein the calculation result can be the bone age of the client corresponding to the first bone image. For example, the development levels of the wrist bone points of the first hand bone image are: B. d, A, B, B, C, A corresponds to 0, B corresponds to 1, and so on, the calculation result is 7, and the bone age of the client corresponding to the first hand bone image is 7 years old.
Optionally, in the embodiment of the present invention, a second hand bone image obtained by labeling the position of each hand wrist bone point, the development level of each hand wrist bone point, and the bone age determined according to the development level of each hand wrist bone point on the first hand bone image may be output. The second hand bone image may be as shown in fig. 13.
Corresponding to the above method embodiment, an embodiment of the present invention further provides a hand carpal development level determination device, as shown in fig. 14, which may include: a feature map obtaining unit 100, a probability map obtaining unit 200, and a development level determining unit 300.
The feature map obtaining unit 100 is configured to obtain a feature map of a first hand bone image.
Specifically, as shown in fig. 2, the first hand skeleton image may be a hand skeleton image obtained by X-ray penetrating irradiation. The scale may be the image size of the image at a certain resolution. The first hand bone image may have a scale of an X-ray image output from an X-ray machine. For example, the resolution corresponding to the scale of the first hand bone image output by the X-ray machine may be 512 × 512. Alternatively, the first hand bone image may be a left hand bone image. The first hand bone image may display a part or all of the carpal bones including the scaphoid, the lunate, the trigonal, the trapezium, the capitate, and the hamate.
Optionally, the feature map obtaining unit 100 may include: an initial map obtaining unit and a feature map obtaining subunit.
The initial image obtaining unit is used for obtaining initial images of the first hand bone image under a plurality of different scales.
The initial map obtaining unit may obtain an initial map of the first hand bone image through a feature extraction network. The initial image obtaining unit may obtain image features of the same feature point of the first hand bone image at different scales.
And the characteristic map obtaining subunit is used for fusing the initial maps under the different scales to obtain a characteristic map.
The feature map obtaining subunit may extract features from the initial maps at a plurality of different scales through an image feature extraction network, and then perform comprehensive processing on information such as edges, shapes, contours, local features, and the like to obtain a feature map.
Optionally, the feature map obtaining subunit may include: the matrix image acquisition unit comprises a matrix image acquisition subunit, an element image acquisition subunit and an up-sampling subunit.
The matrix map obtaining subunit is configured to perform upsampling on the initial maps at the multiple different scales according to a first upsampling parameter, and obtain a fourth number of matrix maps, where the scales of the fourth number of matrix maps are the same.
Specifically, the matrix map obtaining subunit may perform upsampling on a plurality of initial maps at different scales according to different first upsampling parameters, so as to obtain a fourth number of matrix maps. Wherein the fourth number is related to the number of initial graphs.
The first upsampling parameter may be determined according to a scale of the initial map.
And the element map obtaining subunit is configured to perform element summation on the fourth number of matrix maps to obtain an element map.
In particular, the element summation may be adding numbers at the same position in the fourth number of matrix images. The element map obtaining subunit may obtain, as the element map, a result of element summing the fourth number of matrix maps.
And the up-sampling subunit is used for up-sampling the element diagram according to a second up-sampling parameter to obtain a characteristic diagram.
Specifically, the second upsampling parameter may be determined according to a scale of the first hand bone image and a scale of the elemental map.
The probability map obtaining unit 200 is configured to input the feature map into a preset hand carpal development level determination model, and obtain a position probability map of a hand carpal point and a development level probability map of the hand carpal point output by the hand carpal development level determination model.
The preset hand carpal development level determination model can be a convolutional neural network model. The preset hand carpal development level determining model can use the hand skeleton image fusion feature map marked with the hand carpal site position and the hand carpal site development level to perform machine learning, so as to learn the position of the hand carpal site and the image features of the development level of the hand carpal site.
Optionally, the probability map obtaining unit 200 is specifically configured to input the feature map into a preset hand carpal development level determination model, so that the preset hand carpal development level determination model convolves the feature map according to a preset first convolution parameter to obtain a position probability map of a first number of hand carpal points, and convolves the feature map according to a preset second convolution parameter by the preset hand carpal development level determination model to obtain a development level probability map of a second number of hand carpal points.
In particular, the first convolution parameters may include a convolution kernel size and a number of output channels. The first number is equal to the number of output channels in the first convolution parameter. For example, if the number of output channels in the first convolution parameter is 5, the hand carpal development level determination model convolves the feature map according to the first convolution parameter, and the first number of the position probability maps of the obtained hand carpal points is 5.
Normally, 7 carpal bones can be shown in the first hand bone image, and therefore, in an alternative embodiment of the invention, the first number is 7, and the first volume parameter comprises: the convolution kernel size is 3 × 3 and the number of output channels is 7.
The position probability maps of each hand wrist bone point in the position probability maps of the hand wrist bone points of the first number correspond to one hand wrist bone point, and the hand wrist bone points corresponding to the position probability maps of the hand wrist bone points are different.
Specifically, the position probability map of each hand wrist bone point in the position probability maps of the first number of hand wrist bone points is only a probability prediction of one hand wrist bone point at each position in the first hand bone image, wherein each position in the first hand bone image may be each pixel point in the first hand bone image. Therefore, when the first number is 7, the position probability maps of 7 hand wrist bone points can respectively predict the probability of 7 hand wrist bone points at each position in the first hand bone image.
Optionally, the preset hand carpal development level determination model performs convolution on the feature map according to a preset first convolution parameter to obtain a position probability map of a first number of hand carpal points.
And normalizing the position information maps of the first number of hand wrist bone points to obtain a position probability map of the first number of hand wrist bone points.
Optionally, in the embodiment of the present invention, the position information map of the first number of hand wrist bone points may be normalized through a softmax function, so as to obtain the position probability map of the first number of hand wrist bone points.
The second convolution parameters may include convolution kernel size and number of output channels in embodiments of the present invention. The second number is equal to the number of output channels in the second convolution parameter. For example, if the number of output channels in the second convolution parameter is 8, the hand carpal development level determination model convolves the fusion feature map according to the second convolution parameter, and the second number of the development level probability maps of the obtained hand carpal points is 8.
Normally, 7 carpal bones can be shown in the first hand bone image. In the embodiment of the invention, the evaluation of the development grade of the CARPAL bones can be carried out by using the evaluation of the development grade of the CARPAL bones in the TW3-CARPAL bone age evaluation standard. TW3-CARPAL bone age evaluation criteria Each CARPAL bone can be divided into nine developmental levels, A-I. Therefore, in an alternative embodiment of the present invention, the second number is 63, and the second convolution parameter may include: the convolution kernel size is 3 × 3 and the number of output channels is 63. It will be appreciated that the second number may be equal to the product of the number of carpal development levels multiplied by the number of carpal bones in the first hand bone image. Assuming that each carpal bone can be divided into five levels according to other bone age evaluation criteria, the second number may be 35.
The development level probability map of any one hand wrist bone point corresponds to one information combination, the information combinations corresponding to the development level probability maps of the hand wrist bone points are different, and the information combination comprises one hand wrist bone point and one development level.
Specifically, the information combinations corresponding to the development level probability maps of the respective wrist bone points may be that the wrist bone points are the same but have different development levels, or the wrist bone points are different but have the same development levels, or the wrist bone points are different but have different development levels, that is, the development level probability maps of the wrist bone points predict the probability that each position in the first hand bone image is the wrist bone point position, and also predict the development level probability corresponding to each position in the first hand bone image.
Therefore, the hand carpal development level determination model can simultaneously output the position probability map of the hand carpal site and the development level probability map of the hand carpal site.
The development level determining unit 300 is configured to determine a development level of the hand wrist bone point according to the position probability map of the hand wrist bone point and the development level probability map of the hand wrist bone point.
Optionally, the development level determining unit 300 includes: and obtaining a development level probability result subunit and a development level determining subunit.
The development level probability result obtaining subunit is used for obtaining the probability of any hand wrist bone point: multiplying the position probability maps of the hand wrist bone points with the development level probability maps corresponding to the hand wrist bone points respectively to obtain a third number of development level probability results of the hand wrist bone points, wherein the third number is a quotient obtained by dividing the second number by the first number.
And the development level determining subunit is configured to determine the development level with the highest probability in the third number of development level probability results as the development level of the wrist bone point of the hand.
According to the hand carpal development grade determining device provided by the embodiment of the invention, a characteristic diagram of a first hand bone image can be obtained; inputting the feature map into a preset hand carpal development level determination model to obtain a hand carpal point position probability map and a hand carpal point development level probability map output by the hand carpal development level determination model; and determining the development level of the hand wrist bone point according to the position probability map of the hand wrist bone point and the development level probability map of the hand wrist bone point. The invention determines the probability that each position in the hand skeleton image is the position of the carpal bone point and determines the probability of the development grade of each position through the hand carpal bone development grade determination model, thereby determining the technical means of the development grade of each hand carpal bone point in the hand skeleton image, overcoming the technical problem of low reliability of the existing bone age prediction method, and further achieving the technical effect of scientifically determining the bone age.
Optionally, as shown in fig. 15, another hand carpal development level determination device provided in the embodiment of the present invention may further include: the age of the bone determination unit 400 is,
the bone age determining unit 400 is configured to determine the bone age according to the determined development level of each wrist bone point.
Bone age determination section 400 may evaluate the values corresponding to the development levels of the respective wrist bone points based on the development levels of the respective wrist bone points of the first hand bone image according to the TW3-CARPAL bone age evaluation criteria, and determine the calculation result, which may be the bone age of the client corresponding to the first hand bone image, by adding the values corresponding to the development levels of the respective wrist bone points.
Alternatively, the bone age determination unit 400 may output a second hand bone image in which the position of each hand wrist bone point, the development level of each hand wrist bone point, and the bone age determined from the development level of each hand wrist bone point are marked on the first hand bone image. The second hand bone image may be as shown in fig. 13.
In the storage medium provided in the embodiment of the present invention, a computer-executable instruction is stored, and when the computer-executable instruction is loaded and executed by a processor, the method for determining a development level of a wrist bone of a hand according to any one of the above-mentioned embodiments is implemented.
The computer device provided by the embodiment of the invention comprises a processor, a memory and a program which is stored on the memory and can run on the processor, wherein when the processor executes the program, at least the hand carpal development level determination method is realized.
Alternatively, as shown in fig. 16 to 18, a hand carpal development level determination device according to an embodiment of the present invention includes: a hand skeleton image receiving device, a processor 50, a memory 60, a communication bus 70 and an output device, the memory 60 having stored thereon a program executable on the processor 50,
the processor 50 is communicatively connected to the hand skeleton image receiving device, the memory 60 and the output device through the communication bus 70, respectively;
the hand bone image receiving device receives a first hand bone image;
the processor 50, when executing a program, at least implements the hand carpal development level determination method as described in any of the above;
the output device obtains and outputs the development level of the wrist bone point of the hand determined by the processor 50.
Fig. 18 is a schematic diagram illustrating a connection between the processor 50, the memory 60, and the communication bus 70 according to an embodiment of the present invention.
As shown in fig. 16 to 17, the hand skeleton image receiving apparatus may include: a hand skeleton image containing box 10 and an operating mechanism, wherein the hand skeleton image receiving device can also comprise a scanning device or a photographing device. The operating mechanism of the embodiment of the invention can drive the hand skeleton image containing box 10 to extend out of the hand carpal development grade determining device. Specifically, the hand carpal development level determining device may be provided with an access 20, and the hand skeleton image holding box 10 may be extended or retracted from the access 20. After the hand skeleton image holding box 10 is extended from the entrance 20 of the hand carpal development level determination device, the user may first put the hand skeleton image into the hand skeleton image holding box 10. The processor 50 then retracts the hand skeleton image holding box 10 to the hand carpal development level determination device by controlling the operation mechanism. Then, the scanning device or the photographing device can scan or photograph the hand bone image in the hand bone image holding box 10, so as to obtain the hand bone image and send the hand bone image to the processor 50, and after the processor 50 executes the hand carpal bone development level determining method on the hand bone image, the development level of the hand carpal bone point of the hand bone image is output through the output device.
Alternatively, the output device may be a display screen 30 and/or a printer.
Optionally, the hand carpal development level determination device may further include an input device, wherein the input device may be a key. It will be appreciated that the input device may also be a touch screen.
The user can control the hand carpal development level determination device by means of keys or a touch screen, for example: the user firstly controls the hand carpal development level determining device to extend the hand skeleton image containing box 10 out, and after the hand skeleton image is placed in the hand skeleton image containing box 10, the hand skeleton image containing box is controlled to be retracted and processed by the hand carpal development level determining method.
The development grade of the wrist bone point of the hand can be output automatically or according to the operation of the user. For example, the printed result is output from a printer, which may be provided inside the hand carpal development level determination device and output the printing paper through a printing paper output port 40 as shown in fig. 16 to 17.
Of course, the hand carpal development level determination device may also comprise a data interface to obtain hand bone images from other devices via the data interface. The data interface may be a USB interface, a bluetooth communication interface, a Wi-Fi communication interface, or the like, which is not limited in the present invention.
The hand carpal development level determination device comprises a processor and a memory, wherein the characteristic map obtaining unit 100, the probability map obtaining unit 200, the development level determination unit 300 and the like are stored in the memory as program units, and the processor executes the program units stored in the memory to realize corresponding functions.
The processor comprises a kernel, and the kernel calls the corresponding program unit from the memory. The kernel can be set to be one or more, and the development level of each hand carpal shop is determined by adjusting the parameters of the kernel, so that the bone age is determined.
The memory may include volatile memory in a computer readable medium, Random Access Memory (RAM) and/or nonvolatile memory such as Read Only Memory (ROM) or flash memory (flash RAM), and the memory includes at least one memory chip.
An embodiment of the present invention provides a storage medium on which a program is stored, the program implementing the hand carpal development level determination method when executed by a processor.
The embodiment of the invention provides a processor, which is used for running a program, wherein the program executes the hand carpal development level determination method during running.
The embodiment of the invention provides equipment, which comprises a processor, a memory and a program stored on the memory and capable of running on the processor, wherein the processor executes the program to realize the steps of any one of the hand carpal development grade determination methods.
The device herein may be a server, a PC, a PAD, a mobile phone, etc.
The present application further provides a computer program product adapted to perform a program when executed on a data processing device, the program being initialized with the steps of any of the above-described hand carpal development level determination methods.
As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
In a typical configuration, a computing device includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory.
The memory may include forms of volatile memory in a computer readable medium, Random Access Memory (RAM) and/or non-volatile memory, such as Read Only Memory (ROM) or flash memory (flash RAM). The memory is an example of a computer-readable medium.
Computer-readable media, including both non-transitory and non-transitory, removable and non-removable media, may implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), Static Random Access Memory (SRAM), Dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), Read Only Memory (ROM), Electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), Digital Versatile Discs (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information that can be accessed by a computing device. As defined herein, a computer readable medium does not include a transitory computer readable medium such as a modulated data signal and a carrier wave.
It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in the process, method, article, or apparatus that comprises the element.
As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The above are merely examples of the present application and are not intended to limit the present application. Various modifications and changes may occur to those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the claims of the present application.
Claims (12)
1. A hand carpal development level determination method, comprising:
obtaining a feature map of a first hand bone image;
inputting the feature map into a preset hand carpal development level determination model, so that the preset hand carpal development level determination model convolves the feature map according to a preset first convolution parameter to obtain a position probability map of a first number of hand carpal points, and convolving the feature map according to a preset second convolution parameter to obtain a development level probability map of a second number of hand carpal points;
for any hand wrist bone point: multiplying the position probability maps of the hand wrist bone points with the development level probability maps corresponding to the hand wrist bone points respectively to obtain a third number of development level probability results of the hand wrist bone points, wherein the third number is a quotient obtained by dividing the second number by the first number; and determining the development grade with the highest probability in the third number of development grade probability results as the development grade of the wrist bone point of the hand.
2. The method of claim 1, wherein the position probability map of each of the first number of the position probability maps of the wrist bone points corresponds to one of the wrist bone points, the position probability maps of the wrist bone points correspond to different ones of the wrist bone points, the development level probability map of any one of the wrist bone points corresponds to one of the information combinations, and the development level probability map of each of the wrist bone points corresponds to different ones of the information combinations, each of the information combinations comprising one of the wrist bone points and one of the development levels.
3. The method of claim 1, wherein the first number is 7, and wherein the first convolution parameter comprises: the convolution kernel size is 3 × 3 and the number of output channels is 7;
the second number is 63, the third number is 9, and the second convolution parameters include: the convolution kernel size is 3 × 3 and the number of output channels is 63.
4. The method of claim 1, further comprising:
and determining the bone age according to the determined development grade of each hand wrist bone point.
5. The method of any one of claims 1 to 4, wherein the obtaining a feature map of a first hand bone image comprises:
obtaining an initial map of a first hand bone image at a plurality of different scales;
and fusing the initial images under the different scales to obtain a characteristic image.
6. The method according to claim 5, wherein the fusing the initial maps at the plurality of different scales to obtain a feature map comprises:
the initial images under the different scales are up-sampled according to the first up-sampling parameters, and a fourth number of matrix images are obtained, wherein the scales of the fourth number of matrix images are the same;
element summation is carried out on the fourth quantity of matrix diagrams to obtain element diagrams;
and upsampling the element graph according to a second upsampling parameter to obtain a characteristic graph.
7. A hand carpal development level determination device, comprising: the device comprises a feature map obtaining unit, a probability map obtaining unit and a development level determining unit, wherein the development level determining unit comprises: a development grade probability result obtaining subunit and a development grade determining subunit;
the characteristic map obtaining unit is used for obtaining a characteristic map of a first hand bone image;
the probability map obtaining unit is configured to input the feature map into a preset hand carpal development level determination model, so that the preset hand carpal development level determination model convolves the feature map according to a preset first convolution parameter to obtain a position probability map of a first number of hand carpal points, and the preset hand carpal development level determination model convolves the feature map according to a preset second convolution parameter to obtain a development level probability map of a second number of hand carpal points;
the development level probability result obtaining subunit is used for obtaining the probability of any hand wrist bone point: multiplying the position probability maps of the hand wrist bone points with the development level probability maps corresponding to the hand wrist bone points respectively to obtain a third number of development level probability results of the hand wrist bone points, wherein the third number is a quotient obtained by dividing the second number by the first number;
and the development level determining subunit is configured to determine the development level with the highest probability in the third number of development level probability results as the development level of the wrist bone point of the hand.
8. The apparatus according to claim 7, wherein the position probability map of each of the first number of the position probability maps of the wrist bone points corresponds to one of the wrist bone points, the position probability maps of the wrist bone points correspond to different ones of the wrist bone points, the development level probability map of any one of the wrist bone points corresponds to one of the information combinations, and the development level probability map of each of the wrist bone points corresponds to different ones of the information combinations, each of the information combinations comprising one of the wrist bone points and one of the development levels.
9. The apparatus of claim 7, further comprising: a bone age determination unit for determining the age of a bone,
and the bone age determining unit is used for determining the bone age according to the determined development grade of each hand wrist bone point.
10. A storage medium having stored thereon computer-executable instructions which, when loaded and executed by a processor, carry out a method of determining a level of development of a wrist bone of a hand according to any one of claims 1 to 6.
11. A computer device comprising a processor, a memory and a program stored on the memory and executable on the processor, the processor when executing the program at least implementing a hand carpal development level determination method as claimed in any one of the preceding claims 1 to 6.
12. A hand carpal development level determination device, comprising: a hand skeleton image receiving device, a processor, a memory, a communication bus, and an output device, the memory having stored thereon a program executable on the processor,
the processor is respectively in communication connection with the hand skeleton image receiving device, the memory and the output device through the communication bus;
the hand bone image receiving device receives a first hand bone image;
the processor executes a program to implement at least the hand carpal development level determination method as claimed in any one of the preceding claims 1 to 6;
and the output device obtains and outputs the development grade of the hand wrist bone point determined by the processor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911053737.9A CN110782450B (en) | 2019-10-31 | 2019-10-31 | Hand carpal development grade determining method and related equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911053737.9A CN110782450B (en) | 2019-10-31 | 2019-10-31 | Hand carpal development grade determining method and related equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110782450A CN110782450A (en) | 2020-02-11 |
| CN110782450B true CN110782450B (en) | 2020-09-29 |
Family
ID=69388086
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911053737.9A Active CN110782450B (en) | 2019-10-31 | 2019-10-31 | Hand carpal development grade determining method and related equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110782450B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111833298B (en) * | 2020-06-04 | 2022-08-19 | 石家庄喜高科技有限责任公司 | Skeletal development grade detection method and terminal equipment |
| CN113436145A (en) * | 2021-05-24 | 2021-09-24 | 北京医准智能科技有限公司 | Bone age determination method and device based on artificial intelligence and electronic equipment |
| CN114601483B (en) * | 2022-05-11 | 2022-08-16 | 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 |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107895367A (en) * | 2017-11-14 | 2018-04-10 | 中国科学院深圳先进技术研究院 | A kind of stone age recognition methods, system and electronic equipment |
| CN109949280A (en) * | 2019-03-06 | 2019-06-28 | 上海联影智能医疗科技有限公司 | Image processing method, device, equipment storage medium and growth and development assessment system |
Family Cites Families (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP1293925A1 (en) * | 2001-09-18 | 2003-03-19 | Agfa-Gevaert | Radiographic scoring method |
| US10991093B2 (en) * | 2016-09-21 | 2021-04-27 | The General Hospital Corporation | Systems, methods and media for automatically generating a bone age assessment from a radiograph |
| CN107944496A (en) * | 2017-12-06 | 2018-04-20 | 电子科技大学 | Stone age automatic identification system based on improved residual error network |
| CN108334899A (en) * | 2018-01-28 | 2018-07-27 | 浙江大学 | Quantify the bone age assessment method of information integration based on hand bone X-ray bone and joint |
| CN108968991B (en) * | 2018-05-08 | 2022-10-11 | 平安科技（深圳）有限公司 | Hand bone X-ray film bone age assessment method, device, computer equipment and storage medium |
| CN109285154A (en) * | 2018-09-30 | 2019-01-29 | 杭州依图医疗技术有限公司 | A kind of method and device detecting the stone age |
| CN109741309B (en) * | 2018-12-27 | 2021-04-02 | 北京深睿博联科技有限责任公司 | Bone age prediction method and device based on deep regression network |
| CN109948522B (en) * | 2019-03-18 | 2020-12-01 | 浙江工业大学 | X-ray hand bone maturity interpretation method based on deep neural network |
| CN109948614B (en) * | 2019-03-26 | 2022-06-28 | 浙江康体汇科技有限公司 | Wrist bone interest area cutting method based on machine learning |
-
2019
- 2019-10-31 CN CN201911053737.9A patent/CN110782450B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107895367A (en) * | 2017-11-14 | 2018-04-10 | 中国科学院深圳先进技术研究院 | A kind of stone age recognition methods, system and electronic equipment |
| CN109949280A (en) * | 2019-03-06 | 2019-06-28 | 上海联影智能医疗科技有限公司 | Image processing method, device, equipment storage medium and growth and development assessment system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110782450A (en) | 2020-02-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110782450B (en) | Hand carpal development grade determining method and related equipment | |
| CN111667011B (en) | Damage detection model training and vehicle damage detection method, device, equipment and medium | |
| CN110853003B (en) | Hand epiphysis development grade determination method and related equipment | |
| US9147120B2 (en) | Analog utility meter reading | |
| CN107977665A (en) | The recognition methods of key message and computing device in a kind of invoice | |
| CN110443250A (en) | A kind of classification recognition methods of contract seal, device and calculate equipment | |
| CN109165645A (en) | A kind of image processing method, device and relevant device | |
| CN110827345B (en) | Cardiothoracic ratio determining method, cardiothoracic ratio determining device, cardiothoracic ratio determining equipment, storage medium and computer equipment | |
| CN111291825A (en) | Focus classification model training method and device, computer equipment and storage medium | |
| CN107578367B (en) | Method and device for generating stylized image | |
| CN109118456B (en) | Image processing method and device | |
| CN110879972B (en) | Face detection method and device | |
| CN108776791A (en) | Client identity identifies method of calibration, device, computer equipment and storage medium | |
| CN110689134A (en) | Method, apparatus, device and storage medium for performing machine learning process | |
| WO2019034965A1 (en) | Breast cancer detection | |
| CN109447080A (en) | A kind of character identifying method and device | |
| CN111753031A (en) | Map area integration method and device, storage medium and electronic equipment | |
| CN111310758A (en) | Text detection method and device, computer equipment and storage medium | |
| JPWO2020240808A1 (en) | Learning device, classification device, learning method, classification method, learning program, and classification program | |
| CN113704313A (en) | Map annotation gland automatic identification method, device and medium based on topology inspection | |
| CN114387289A (en) | Semantic segmentation method and device for three-dimensional point cloud of power transmission and distribution overhead line | |
| CN110705633B (en) | Target object detection method and device and target object detection model establishing method and device | |
| CN114240949A (en) | Cervical cell segmentation network training method, cervical cell segmentation method and cervical cell segmentation device | |
| CN115294618A (en) | Image processing method and device | |
| CN116468804B (en) | Laser radar and camera external parameter calibration precision evaluation method and device |
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 | ||
| CP03 | Change of name, title or address | ||
| CP03 | Change of name, title or address |
Address after: Room B401, floor 4, building 1, Shangdi Information 12, Haidian District, Beijing 100085 Patentee after: Tuxiang Medical Technology Co., Ltd Address before: Room B401, floor 4, building 1, No. 12, Shangdi Information Road, Haidian District, Beijing 100085 Patentee before: Beijing Tuoxiang Technology Co.,Ltd. |