CN116245841A - Chest digital image heart-chest ratio measuring method for medical image specialty - Google Patents
Chest digital image heart-chest ratio measuring method for medical image specialty Download PDFInfo
- Publication number
- CN116245841A CN116245841A CN202310199278.5A CN202310199278A CN116245841A CN 116245841 A CN116245841 A CN 116245841A CN 202310199278 A CN202310199278 A CN 202310199278A CN 116245841 A CN116245841 A CN 116245841A
- Authority
- CN
- China
- Prior art keywords
- chest
- image
- chest image
- module
- processed
- 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.)
- Withdrawn
Links
- 238000000034 method Methods 0.000 title claims description 55
- 238000003709 image segmentation Methods 0.000 claims abstract description 49
- 238000005259 measurement Methods 0.000 claims abstract description 20
- 230000001965 increasing effect Effects 0.000 claims abstract description 7
- 238000000691 measurement method Methods 0.000 claims abstract description 6
- 238000012545 processing Methods 0.000 claims description 28
- 238000012549 training Methods 0.000 claims description 20
- 230000011218 segmentation Effects 0.000 claims description 16
- 238000000605 extraction Methods 0.000 claims description 13
- 238000002059 diagnostic imaging Methods 0.000 claims description 10
- 241001270131 Agaricus moelleri Species 0.000 claims description 9
- 230000002708 enhancing effect Effects 0.000 claims description 7
- 230000003044 adaptive effect Effects 0.000 claims description 3
- 230000015572 biosynthetic process Effects 0.000 claims description 3
- 238000001514 detection method Methods 0.000 claims description 3
- 238000011176 pooling Methods 0.000 claims description 3
- 238000003786 synthesis reaction Methods 0.000 claims description 3
- 230000000694 effects Effects 0.000 abstract description 5
- 238000003384 imaging method Methods 0.000 abstract 1
- 210000000038 chest Anatomy 0.000 description 260
- 208000006029 Cardiomegaly Diseases 0.000 description 2
- 238000003491 array Methods 0.000 description 2
- 238000012827 research and development Methods 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 230000000747 cardiac effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000008092 positive effect Effects 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 238000002636 symptomatic treatment Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
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
- 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
- G06T3/00—Geometric image transformations in the plane of the image
- G06T3/40—Scaling of whole images or parts thereof, e.g. expanding or contracting
- G06T3/4038—Image mosaicing, e.g. composing plane images from plane sub-images
-
- 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 using histogram techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/90—Dynamic range modification of images or parts thereof
- G06T5/94—Dynamic range modification of images or parts thereof based on local image properties, e.g. for local contrast enhancement
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/764—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/77—Processing image or video features in feature spaces; using data integration or data reduction, e.g. principal component analysis [PCA] or independent component analysis [ICA] or self-organising maps [SOM]; Blind source separation
- G06V10/774—Generating sets of training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/82—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks
-
- 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
- G06T2207/30048—Heart; Cardiac
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Evolutionary Computation (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Health & Medical Sciences (AREA)
- Software Systems (AREA)
- Medical Informatics (AREA)
- Artificial Intelligence (AREA)
- Computing Systems (AREA)
- Multimedia (AREA)
- Databases & Information Systems (AREA)
- Quality & Reliability (AREA)
- Radiology & Medical Imaging (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Life Sciences & Earth Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Molecular Biology (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Image Processing (AREA)
Abstract
The invention belongs to the technical field of heart-chest ratio measurement, and discloses a chest digital image heart-chest ratio measurement method of a medical image specialty, wherein the chest digital image heart-chest ratio measurement system of the medical image specialty comprises: the device comprises a chest image acquisition module, a main control module, a chest image enhancement module, an image segmentation module, a chest identification module, a measurement module and a display module. The layering sense of the chest image is increased through the chest image enhancement module, so that the imaging effect of the chest image is effectively improved; meanwhile, the image segmentation module is used for carrying out scene recognition on the chest image to be processed so as to judge whether the chest image segmentation requirement exists in the scene where the chest image is located, and the accuracy of chest image segmentation can be improved and the situation of chest image segmentation errors can be reduced only on the premise that the chest image segmentation requirement exists in the scene where the chest image is located.
Description
Technical Field
The invention belongs to the technical field of heart-chest ratio measurement, and particularly relates to a chest digital image heart-chest ratio measurement method for medical image profession.
Background
The heart-chest ratio refers to the ratio of the transverse diameter of the heart to the transverse diameter of the thorax on an X-ray film. The transverse diameter of the heart refers to the sum of the maximum distances from the left and right cardiac edges to the midline, and the transverse diameter of the thorax refers to the inner diameter of the thorax through the right diaphragmatic apex. The heart-chest ratio measurement method is simple, is easy to follow-up comparison, is one of the most commonly used heart measurement methods, and is a commonly used index for evaluating heart enlargement. Heart enlargement is classified into three grades of mild enlargement, moderate enlargement and severe enlargement according to the values, and further symptomatic treatment is carried out. However, chest images acquired by the chest digital image cardiothoracic ratio measurement method of the prior medical image profession are not clear, and the measurement result is affected; meanwhile, when chest images are segmented, a clinician is often required to manually label the chest images, and segmentation errors are easy to occur due to differences in the aspects of relevant medical experience, judgment standards and the like.
Through the above analysis, the problems and defects existing in the prior art are as follows:
(1) The chest image acquired by the chest digital image cardiothoracic ratio measuring method in the prior medical image profession is not clear, and the measuring result is affected.
(2) When chest images are segmented, a clinician is often required to manually label the chest images, and segmentation errors are easy to occur due to the fact that segmentation differences exist due to the fact that relevant medical experiences and judgment standards of the clinician are inconsistent.
(3) The chest image is not accurately identified.
Disclosure of Invention
Aiming at the problems existing in the prior art, the invention provides a chest digital image heart-chest ratio measuring method for medical image profession.
The invention is realized in such a way that a chest digital image heart-chest ratio measuring method for medical image profession comprises the following steps:
step one, acquiring chest image data by using medical image equipment through a chest image acquisition module;
step two, the main control module carries out enhancement treatment on the chest image through the chest image enhancement module;
dividing the chest image through an image dividing module; the chest is identified through the chest identification module; measuring the heart-chest ratio through a measuring module;
the chest recognition module recognition method comprises the following steps:
acquiring a chest image, and marking the chest image to obtain a training sample; constructing a chest image recognition network; the chest image recognition network comprises an input network, a feature extraction network and a classification network; the input network is used for extracting the cavity convolution characteristics of the training samples in a cavity convolution mode;
the feature extraction network comprises an X-DMFF module, a DMS module and an adaptive pooling layer; the feature extraction network is used for extracting local features containing accurate coordinate information and texture information in the cavity convolution features by adopting an X-DMFF module, and obtaining multi-scale spatial features by passing the local features through a DMS module;
the X-DMFF module is a convolution module with adjustable channel number and is used for dividing the input characteristics into two identical channels, convoluting the division results, and performing channel splicing and channel shuffling on the obtained convolution results to obtain local information characteristics; the DMS module comprises two branches taking the self-attention module and the convolution module as main parts respectively, and is used for dividing the input features into two paths with the same channel, inputting the two paths of features into the two branches respectively for feature extraction, and then performing channel splicing and channel shuffling to obtain multi-scale spatial features; the classification network is used for classifying the training samples according to the multi-scale spatial features to obtain sample prediction classification results;
training the chest image recognition network according to the labels of the training samples and sample prediction classification results obtained by inputting the training samples into the chest image recognition network, so as to obtain a trained chest image recognition network; acquiring a chest image to be detected, and inputting the chest image to be detected into a trained chest image recognition network to obtain the category of the chest image;
and step four, displaying chest images, identification results and heart-chest ratio measurement results through a display module.
Another object of the present invention is to provide a chest digital image cardiothoracic ratio measurement system for medical image professionals, comprising:
the chest image acquisition module is used for acquiring chest image data through the medical image equipment;
the chest image enhancement module is used for enhancing the chest image;
the image segmentation module is used for segmenting the chest image;
the chest identification module is used for identifying the chest;
the measuring module is used for measuring the heart-chest ratio;
the display module is used for displaying chest images, identification results and heart-chest ratio measurement results;
and the main control module is used for controlling each module to work normally.
Further, the main control module is electrically connected with the chest image acquisition module, the chest image enhancement module, the image segmentation module, the chest identification module, the measurement module and the display module respectively.
Further, the chest image enhancement module enhancement method comprises the following steps:
(1) Constructing a chest image set, and counting the number of pixel points corresponding to each gray level in the chest image through a counting program, so as to determine the occurrence times of each gray level; obtaining the Lp norm of the times of each gray level, thereby obtaining the Lp norm of the times of each gray level;
(2) In the process of carrying out histogram equalization processing on the chest image, when the number of times of occurrence of a certain gray level is increased, the increasing rate is reduced along with the increase of the number of times of occurrence of the gray level, and the histogram equalization is carried out on the chest image based on the Lp norm of the number of times of occurrence of each gray level so as to realize the enhancement processing on the chest image.
Further, the Lp norm is obtained by:
storing Lp norms corresponding to integers smaller than a first threshold in advance;
and acquiring the Lp norms of the times of occurrence of each gray level of the chest image by searching the Lp norms corresponding to the stored integers.
Further, the storing process includes: storing Lp norms corresponding to integers at each interval of the second threshold number;
the first threshold value is set correspondingly according to the resolution of the chest image.
Further, the obtaining the Lp norm of the number of occurrences of each gray level to obtain the Lp norm of the number of occurrences of each gray level includes: when the number of times of occurrence of the gray scale is larger than a first threshold value, the Lp norm value of the number of times of occurrence of the gray scale is the Lp norm value corresponding to the first threshold value;
the value range of p of the Lp norm is more than or equal to 0 and less than 1;
and carrying out histogram equalization on pixel points in the chest image through the following formula:
wherein I (x, y) is a luminance value of a pixel point at a (x, y) position in the chest image, j is an index value of gray scales in the chest image, M is a total number of gray scales in the chest image, H x (j) is an Lp norm value of the number of occurrences of the gray scales j in the chest image, and ψ (I (x, y)) is a luminance value of the pixel point at the (x, y) position after histogram equalization;
the chest image is in YUV format, and the brightness value of the pixel point is the Y component of the pixel point.
Further, the image segmentation module segments the following method:
1) Configuring parameters of medical imaging equipment, and acquiring a chest image to be processed through the medical imaging equipment; extracting a first chest image from a preset area in the chest image to be processed, wherein the preset area comprises specific features in the chest image to be processed, and the specific features are features which are easy to cause confusion of chest image segmentation or features which are subjected to subsequent key processing;
2) Performing enhancement processing on the first chest image to obtain a second chest image; inputting the second chest image into a chest image segmentation network to obtain a segmentation result;
wherein before the first chest image is extracted from the preset area in the chest image to be processed, the method further comprises the steps of;
performing scene recognition on the chest image to be processed to obtain a scene category of the chest image to be processed;
judging whether the scene category of the chest image to be processed is a preset scene category or not;
and performing target recognition on the chest image to be processed; obtaining a result of the target identification;
the extracting the first chest image from the preset area in the chest image to be processed includes:
if the scene category of the chest image to be processed is a preset scene category, extracting the first chest image from a preset area in the chest image to be processed;
after the result of the target identification is obtained, the method further comprises: and detecting whether the target identification result is the specific characteristic or not, and determining the preset area according to the detection result.
Further, before the first chest image is extracted from the preset area in the chest image to be processed, the method further includes:
and determining the position of the specific feature in the chest image to be processed based on the target identification result.
Further, the determining a location of the specific feature in the chest image to be processed based on the result of the target recognition includes;
and if the target recognition result is the specific feature, determining the position of the specific feature in the chest image to be processed based on the target recognition result.
Further, after the inputting the second chest image into the chest image segmentation network to obtain the segmentation result, the method further includes:
acquiring histogram information of the second chest image;
acquiring the dynamic range of the gray value based on the histogram information;
judging whether the dynamic range exceeds a preset dynamic range or not;
if the dynamic range exceeds the preset dynamic range, chest image processing is carried out on the chest image to be processed according to the segmentation result;
after the judging whether the dynamic range exceeds the preset dynamic range, the method further comprises the following steps:
if the dynamic range does not exceed the preset dynamic range, acquiring a third chest image based on the chest image to be processed, wherein the third chest image and the chest image to be processed only have exposure difference;
extracting a fourth chest image from a preset area in the third chest image, wherein the preset area comprises specific features in the chest image to be processed;
performing enhancement processing on the fourth chest image to obtain a fifth chest image;
performing high dynamic range chest image synthesis processing on the fifth chest image to obtain a sixth chest image;
and inputting the sixth chest image into the chest image segmentation network to obtain a new segmentation result.
In combination with the above technical solution and the technical problems to be solved, please analyze the following aspects to provide the following advantages and positive effects:
first, aiming at the technical problems in the prior art and the difficulty in solving the problems, the technical problems solved by the technical proposal of the invention are analyzed in detail and deeply by tightly combining the technical proposal to be protected, the results and data in the research and development process, and the like, and some technical effects brought after the problems are solved have creative technical effects. The specific description is as follows:
according to the method, the Lp norms of the times of occurrence of each gray level of the chest image are obtained through the chest image enhancement module, and then histogram equalization processing is carried out on the chest image based on the Lp norms of the times of occurrence of each gray level, so that the enhancement processing of the chest image is realized; meanwhile, the image segmentation module is used for carrying out scene recognition on the chest image to be processed so as to judge whether the chest image segmentation requirement exists on the scene where the chest image is located, and the chest image region containing specific features is enhanced only on the premise that the chest image segmentation requirement exists on the scene where the chest image is located, for example, the contrast, the color, the brightness, the edges and the like are improved, so that the edge definition of the specific features is improved, the accuracy of the chest image segmentation can be improved, and the situation of chest image segmentation errors can be reduced.
The method comprises the steps of obtaining a chest image through a chest identification module, and marking the chest image to obtain a training sample; constructing a chest image recognition network; the chest image recognition network comprises an input network, a feature extraction network and a classification network; the input network is used for extracting the cavity convolution characteristics of the training sample in a cavity convolution mode, so that chest images can be accurately identified
Secondly, the technical scheme is regarded as a whole or from the perspective of products, and the technical scheme to be protected has the following technical effects and advantages:
according to the method, the Lp norms of the times of occurrence of each gray level of the chest image are obtained through the chest image enhancement module, and then histogram equalization processing is carried out on the chest image based on the Lp norms of the times of occurrence of each gray level, so that the enhancement processing of the chest image is realized; meanwhile, the image segmentation module is used for carrying out scene recognition on the chest image to be processed so as to judge whether the chest image segmentation requirement exists on the scene where the chest image is located, and the chest image region containing specific features is enhanced only on the premise that the chest image segmentation requirement exists on the scene where the chest image is located, for example, the contrast, the color, the brightness, the edges and the like are improved, so that the edge definition of the specific features is improved, the accuracy of the chest image segmentation can be improved, and the situation of chest image segmentation errors can be reduced.
Drawings
Fig. 1 is a flowchart of a method for measuring a chest digital image cardiothoracic ratio of a medical image specialty according to an embodiment of the present invention.
Fig. 2 is a block diagram of a chest digital image cardiothoracic ratio measurement system for medical image profession according to an embodiment of the present invention.
Fig. 3 is a flowchart of a chest image enhancement module enhancement method according to an embodiment of the present invention.
Fig. 4 is a flowchart of a method for dividing an image by using an image dividing module according to an embodiment of the present invention.
In fig. 2: 1. a chest image acquisition module; 2. a main control module; 3. a chest image enhancement module; 4. an image segmentation module; 5. a chest identification module; 6. a measurement module; 7. and a display module.
Detailed Description
The present invention will be described in further detail with reference to the following examples in order to make the objects, technical solutions and advantages of the present invention more apparent. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the invention.
1. The embodiments are explained. In order to fully understand how the invention may be embodied by those skilled in the art, this section is an illustrative embodiment in which the claims are presented for purposes of illustration.
As shown in fig. 1, the method for measuring the heart-chest ratio of the chest digital image in the medical image profession provided by the embodiment of the invention comprises the following steps:
s101, acquiring chest image data by using medical image equipment through a chest image acquisition module;
s102, a main control module enhances the chest image through a chest image enhancement module;
s103, dividing the chest image through an image dividing module; the chest is identified through the chest identification module; measuring the heart-chest ratio through a measuring module;
the chest recognition module recognition method comprises the following steps:
acquiring a chest image, and marking the chest image to obtain a training sample; constructing a chest image recognition network; the chest image recognition network comprises an input network, a feature extraction network and a classification network; the input network is used for extracting the cavity convolution characteristics of the training samples in a cavity convolution mode;
the feature extraction network comprises an X-DMFF module, a DMS module and an adaptive pooling layer; the feature extraction network is used for extracting local features containing accurate coordinate information and texture information in the cavity convolution features by adopting an X-DMFF module, and obtaining multi-scale spatial features by passing the local features through a DMS module;
the X-DMFF module is a convolution module with adjustable channel number and is used for dividing the input characteristics into two identical channels, convoluting the division results, and performing channel splicing and channel shuffling on the obtained convolution results to obtain local information characteristics; the DMS module comprises two branches taking the self-attention module and the convolution module as main parts respectively, and is used for dividing the input features into two paths with the same channel, inputting the two paths of features into the two branches respectively for feature extraction, and then performing channel splicing and channel shuffling to obtain multi-scale spatial features; the classification network is used for classifying the training samples according to the multi-scale spatial features to obtain sample prediction classification results;
training the chest image recognition network according to the labels of the training samples and sample prediction classification results obtained by inputting the training samples into the chest image recognition network, so as to obtain a trained chest image recognition network; acquiring a chest image to be detected, and inputting the chest image to be detected into a trained chest image recognition network to obtain the category of the chest image;
s104, displaying chest images, identification results and heart-chest ratio measurement results through a display module.
As shown in fig. 2, the chest digital image cardiothoracic ratio measurement system for medical image profession provided by the embodiment of the invention comprises: the device comprises a chest image acquisition module 1, a main control module 2, a chest image enhancement module 3, an image segmentation module 4, a chest identification module 5, a measurement module 6 and a display module 7.
The chest image acquisition module 1 is connected with the main control module 2 and is used for acquiring chest image data through medical image equipment;
the main control module 2 is connected with the chest image acquisition module 1, the chest image enhancement module 3, the image segmentation module 4, the chest identification module 5, the measurement module 6 and the display module 7 and is used for controlling the normal work of each module;
the chest image enhancement module 3 is connected with the main control module 2 and is used for enhancing chest images;
the image segmentation module 4 is connected with the main control module 2 and is used for segmenting chest images;
the chest identification module 5 is connected with the main control module 2 and is used for identifying the chest;
the measuring module 6 is connected with the main control module 2 and is used for measuring the heart-chest ratio;
and the display module 7 is connected with the main control module 2 and is used for displaying chest images, identification results and heart-chest ratio measurement results.
As shown in fig. 3, the method for enhancing the chest image enhancement module 3 according to the embodiment of the present invention is as follows:
s201, constructing a chest image set, and counting the number of pixel points corresponding to each gray level in the chest image through a counting program, so as to determine the occurrence times of each gray level; obtaining the Lp norm of the times of each gray level, thereby obtaining the Lp norm of the times of each gray level;
s202, in the process of carrying out histogram equalization processing on the chest image, when the number of times of occurrence of a certain gray level is increased, the increasing rate is reduced along with the increase of the number of times of occurrence of the gray level, and the histogram equalization is carried out on the chest image based on the Lp norm of the number of times of occurrence of each gray level so as to realize the enhancement processing on the chest image.
The Lp norm is obtained by the following method:
storing Lp norms corresponding to integers smaller than a first threshold in advance;
and acquiring the Lp norms of the times of occurrence of each gray level of the chest image by searching the Lp norms corresponding to the stored integers.
The storage process provided by the embodiment of the invention comprises the following steps: storing Lp norms corresponding to integers at each interval of the second threshold number;
the first threshold value is set correspondingly according to the resolution of the chest image.
The invention provides a method for obtaining Lp norms of times of occurrence of each gray level, thereby obtaining the Lp norms of the times of occurrence of each gray level, comprising the following steps: when the number of times of occurrence of the gray scale is larger than a first threshold value, the Lp norm value of the number of times of occurrence of the gray scale is the Lp norm value corresponding to the first threshold value;
the value range of p of the Lp norm is more than or equal to 0 and less than 1;
and carrying out histogram equalization on pixel points in the chest image through the following formula:
wherein I (x, y) is a luminance value of a pixel point at a (x, y) position in the chest image, j is an index value of gray scales in the chest image, M is a total number of gray scales in the chest image, H x (j) is an Lp norm value of the number of occurrences of the gray scales j in the chest image, and ψ (I (x, y)) is a luminance value of the pixel point at the (x, y) position after histogram equalization;
the chest image is in YUV format, and the brightness value of the pixel point is the Y component of the pixel point.
As shown in fig. 4, the image segmentation module 4 provided in the embodiment of the present invention has the following segmentation methods:
s301, configuring medical imaging equipment parameters, and acquiring a chest image to be processed through medical imaging equipment; extracting a first chest image from a preset area in the chest image to be processed, wherein the preset area comprises specific features in the chest image to be processed, and the specific features are features which are easy to cause confusion of chest image segmentation or features which are subjected to subsequent key processing;
s302, enhancing the first chest image to obtain a second chest image; inputting the second chest image into a chest image segmentation network to obtain a segmentation result;
wherein before the first chest image is extracted from the preset area in the chest image to be processed, the method further comprises the steps of;
performing scene recognition on the chest image to be processed to obtain a scene category of the chest image to be processed;
judging whether the scene category of the chest image to be processed is a preset scene category or not;
and performing target recognition on the chest image to be processed; obtaining a result of the target identification;
the extracting the first chest image from the preset area in the chest image to be processed includes:
if the scene category of the chest image to be processed is a preset scene category, extracting the first chest image from a preset area in the chest image to be processed;
after the result of the target identification is obtained, the method further comprises: and detecting whether the target identification result is the specific characteristic or not, and determining the preset area according to the detection result.
The method provided by the embodiment of the invention further comprises the following steps before the first chest image is extracted from the preset area in the chest image to be processed:
and determining the position of the specific feature in the chest image to be processed based on the target identification result.
The method for determining the position of the specific feature in the chest image to be processed based on the target recognition result provided by the embodiment of the invention comprises the following steps of;
and if the target recognition result is the specific feature, determining the position of the specific feature in the chest image to be processed based on the target recognition result.
After the second chest image is input to the chest image segmentation network to obtain the segmentation result, the method further includes:
acquiring histogram information of the second chest image;
acquiring the dynamic range of the gray value based on the histogram information;
judging whether the dynamic range exceeds a preset dynamic range or not;
if the dynamic range exceeds the preset dynamic range, chest image processing is carried out on the chest image to be processed according to the segmentation result;
after the judging whether the dynamic range exceeds the preset dynamic range, the method further comprises the following steps:
if the dynamic range does not exceed the preset dynamic range, acquiring a third chest image based on the chest image to be processed, wherein the third chest image and the chest image to be processed only have exposure difference;
extracting a fourth chest image from a preset area in the third chest image, wherein the preset area comprises specific features in the chest image to be processed;
performing enhancement processing on the fourth chest image to obtain a fifth chest image;
performing high dynamic range chest image synthesis processing on the fifth chest image to obtain a sixth chest image;
and inputting the sixth chest image into the chest image segmentation network to obtain a new segmentation result.
2. Application example. In order to prove the inventive and technical value of the technical solution of the present invention, this section is an application example on specific products or related technologies of the claim technical solution.
According to the embodiment of the invention, the Lp norms of the times of each gray level of the chest image are obtained through the chest image enhancement module, and then the histogram equalization processing is carried out on the chest image based on the Lp norms of the times of each gray level, so that the enhancement processing of the chest image is realized; meanwhile, the image segmentation module is used for carrying out scene recognition on the chest image to be processed so as to judge whether the chest image segmentation requirement exists on the scene where the chest image is located, and the chest image region containing specific features is enhanced only on the premise that the chest image segmentation requirement exists on the scene where the chest image is located, for example, the contrast, the color, the brightness, the edges and the like are improved, so that the edge definition of the specific features is improved, the accuracy of the chest image segmentation can be improved, and the situation of chest image segmentation errors can be reduced.
It should be noted that the embodiments of the present invention can be realized in hardware, software, or a combination of software and hardware. The hardware portion may be implemented using dedicated logic; the software portions may be stored in a memory and executed by a suitable instruction execution system, such as a microprocessor or special purpose design hardware. Those of ordinary skill in the art will appreciate that the apparatus and methods described above may be implemented using computer executable instructions and/or embodied in processor control code, such as provided on a carrier medium such as a magnetic disk, CD or DVD-ROM, a programmable memory such as read only memory (firmware), or a data carrier such as an optical or electronic signal carrier. The device of the present invention and its modules may be implemented by hardware circuitry, such as very large scale integrated circuits or gate arrays, semiconductors such as logic chips, transistors, etc., or programmable hardware devices such as field programmable gate arrays, programmable logic devices, etc., as well as software executed by various types of processors, or by a combination of the above hardware circuitry and software, such as firmware.
3. Evidence of the effect of the examples. The embodiment of the invention has a great advantage in the research and development or use process, and has the following description in combination with data, charts and the like of the test process.
According to the embodiment of the invention, the Lp norms of the times of each gray level of the chest image are obtained through the chest image enhancement module, and then the histogram equalization processing is carried out on the chest image based on the Lp norms of the times of each gray level, so that the enhancement processing of the chest image is realized; meanwhile, the image segmentation module is used for carrying out scene recognition on the chest image to be processed so as to judge whether the chest image segmentation requirement exists on the scene where the chest image is located, and the chest image region containing specific features is enhanced only on the premise that the chest image segmentation requirement exists on the scene where the chest image is located, for example, the contrast, the color, the brightness, the edges and the like are improved, so that the edge definition of the specific features is improved, the accuracy of the chest image segmentation can be improved, and the situation of chest image segmentation errors can be reduced.
The foregoing is merely illustrative of specific embodiments of the present invention, and the scope of the invention is not limited thereto, but any modifications, equivalents, improvements and alternatives falling within the spirit and principles of the present invention will be apparent to those skilled in the art within the scope of the present invention.
Claims (10)
1. The chest digital image cardiothoracic ratio measuring method for medical image profession is characterized by comprising the following steps:
step one, acquiring chest image data by using medical image equipment through a chest image acquisition module;
step two, the main control module carries out enhancement treatment on the chest image through the chest image enhancement module;
dividing the chest image through an image dividing module; the chest is identified through the chest identification module; measuring the heart-chest ratio through a measuring module;
the chest recognition module recognition method comprises the following steps:
acquiring a chest image, and marking the chest image to obtain a training sample; constructing a chest image recognition network; the chest image recognition network comprises an input network, a feature extraction network and a classification network; the input network is used for extracting the cavity convolution characteristics of the training samples in a cavity convolution mode;
the feature extraction network comprises an X-DMFF module, a DMS module and an adaptive pooling layer; the feature extraction network is used for extracting local features containing accurate coordinate information and texture information in the cavity convolution features by adopting an X-DMFF module, and obtaining multi-scale spatial features by passing the local features through a DMS module;
the X-DMFF module is a convolution module with adjustable channel number and is used for dividing the input characteristics into two identical channels, convoluting the division results, and performing channel splicing and channel shuffling on the obtained convolution results to obtain local information characteristics; the DMS module comprises two branches taking the self-attention module and the convolution module as main parts respectively, and is used for dividing the input features into two paths with the same channel, inputting the two paths of features into the two branches respectively for feature extraction, and then performing channel splicing and channel shuffling to obtain multi-scale spatial features; the classification network is used for classifying the training samples according to the multi-scale spatial features to obtain sample prediction classification results;
training the chest image recognition network according to the labels of the training samples and sample prediction classification results obtained by inputting the training samples into the chest image recognition network, so as to obtain a trained chest image recognition network; acquiring a chest image to be detected, and inputting the chest image to be detected into a trained chest image recognition network to obtain the category of the chest image;
and step four, displaying chest images, identification results and heart-chest ratio measurement results through a display module.
A chest digital image cardiothoracic measurement system for medical imaging profession comprising:
the chest image acquisition module is used for acquiring chest image data through the medical image equipment;
the chest image enhancement module is used for enhancing the chest image;
the image segmentation module is used for segmenting the chest image;
the chest identification module is used for identifying the chest;
the measuring module is used for measuring the heart-chest ratio;
the display module is used for displaying chest images, identification results and heart-chest ratio measurement results;
and the main control module is used for controlling each module to work normally.
2. The method for measuring the heart-chest ratio of the chest digital image in the medical image specialty of claim 1, wherein the main control module is electrically connected with the chest image acquisition module, the chest image enhancement module, the image segmentation module, the chest identification module, the measurement module and the display module respectively.
3. The method for measuring the heart-chest ratio of the chest digital image of the medical image specialty according to claim 1, wherein the method for enhancing the chest image by the enhancing module is as follows:
(1) Constructing a chest image set, and counting the number of pixel points corresponding to each gray level in the chest image through a counting program, so as to determine the occurrence times of each gray level; obtaining the Lp norm of the times of each gray level, thereby obtaining the Lp norm of the times of each gray level;
(2) In the process of carrying out histogram equalization processing on the chest image, when the number of times of occurrence of a certain gray level is increased, the increasing rate is reduced along with the increase of the number of times of occurrence of the gray level, and the histogram equalization is carried out on the chest image based on the Lp norm of the number of times of occurrence of each gray level so as to realize the enhancement processing on the chest image.
4. A method for measuring a chest digital image cardiothoracic ratio in a medical image specialty according to claim 3, wherein said Lp norm is obtained by:
storing Lp norms corresponding to integers smaller than a first threshold in advance;
and acquiring the Lp norms of the times of occurrence of each gray level of the chest image by searching the Lp norms corresponding to the stored integers.
5. A chest digital image cardiology ratio measurement method for medical image professionals according to claim 3, wherein said storage procedure comprises: storing Lp norms corresponding to integers at each interval of the second threshold number;
the first threshold value is set correspondingly according to the resolution of the chest image.
6. A method for measuring a chest digital image cardiothoracic ratio in a medical image specialty according to claim 3, wherein said obtaining the Lp norm of the number of occurrences of each gray level to obtain the Lp norm of the number of occurrences of each gray level comprises: when the number of times of occurrence of the gray scale is larger than a first threshold value, the Lp norm value of the number of times of occurrence of the gray scale is the Lp norm value corresponding to the first threshold value;
the value range of p of the Lp norm is more than or equal to 0 and less than 1;
and carrying out histogram equalization on pixel points in the chest image through the following formula:
wherein I (x, y) is a luminance value of a pixel point at a (x, y) position in the chest image, j is an index value of gray scales in the chest image, M is a total number of gray scales in the chest image, H x (j) is an Lp norm value of the number of occurrences of the gray scales j in the chest image, and ψ (I (x, y)) is a luminance value of the pixel point at the (x, y) position after histogram equalization;
the chest image is in YUV format, and the brightness value of the pixel point is the Y component of the pixel point.
7. The method for measuring the heart-chest ratio of the chest digital image of the medical image specialty according to claim 1, wherein the image segmentation module segments as follows:
1) Configuring parameters of medical imaging equipment, and acquiring a chest image to be processed through the medical imaging equipment; extracting a first chest image from a preset area in the chest image to be processed, wherein the preset area comprises specific features in the chest image to be processed, and the specific features are features which are easy to cause confusion of chest image segmentation or features which are subjected to subsequent key processing;
2) Performing enhancement processing on the first chest image to obtain a second chest image; inputting the second chest image into a chest image segmentation network to obtain a segmentation result;
wherein before the first chest image is extracted from the preset area in the chest image to be processed, the method further comprises the steps of;
performing scene recognition on the chest image to be processed to obtain a scene category of the chest image to be processed;
judging whether the scene category of the chest image to be processed is a preset scene category or not;
and performing target recognition on the chest image to be processed; obtaining a result of the target identification;
the extracting the first chest image from the preset area in the chest image to be processed includes:
if the scene category of the chest image to be processed is a preset scene category, extracting the first chest image from a preset area in the chest image to be processed;
after the result of the target identification is obtained, the method further comprises: and detecting whether the target identification result is the specific characteristic or not, and determining the preset area according to the detection result.
8. The method for measuring the heart-to-chest ratio of a chest digital image for medical imaging profession according to claim 7, wherein said extracting a first chest image from a predetermined area in said chest image to be processed further comprises:
and determining the position of the specific feature in the chest image to be processed based on the target identification result.
9. The method for measuring the heart-to-chest ratio of a chest digital image specialized for medical imaging according to claim 7, wherein the determining the position of the specific feature in the chest image to be processed based on the result of the target recognition comprises;
and if the target recognition result is the specific feature, determining the position of the specific feature in the chest image to be processed based on the target recognition result.
10. The method for measuring the heart-to-chest ratio of a chest digital image for medical imaging profession of claim 7, wherein after said inputting said second chest image into the chest image segmentation network to obtain a segmentation result, further comprising:
acquiring histogram information of the second chest image;
acquiring the dynamic range of the gray value based on the histogram information;
judging whether the dynamic range exceeds a preset dynamic range or not;
if the dynamic range exceeds the preset dynamic range, chest image processing is carried out on the chest image to be processed according to the segmentation result;
after the judging whether the dynamic range exceeds the preset dynamic range, the method further comprises the following steps:
if the dynamic range does not exceed the preset dynamic range, acquiring a third chest image based on the chest image to be processed, wherein the third chest image and the chest image to be processed only have exposure difference;
extracting a fourth chest image from a preset area in the third chest image, wherein the preset area comprises specific features in the chest image to be processed;
performing enhancement processing on the fourth chest image to obtain a fifth chest image;
performing high dynamic range chest image synthesis processing on the fifth chest image to obtain a sixth chest image;
and inputting the sixth chest image into the chest image segmentation network to obtain a new segmentation result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310199278.5A CN116245841A (en) | 2023-03-03 | 2023-03-03 | Chest digital image heart-chest ratio measuring method for medical image specialty |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310199278.5A CN116245841A (en) | 2023-03-03 | 2023-03-03 | Chest digital image heart-chest ratio measuring method for medical image specialty |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116245841A true CN116245841A (en) | 2023-06-09 |
Family
ID=86632880
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310199278.5A Withdrawn CN116245841A (en) | 2023-03-03 | 2023-03-03 | Chest digital image heart-chest ratio measuring method for medical image specialty |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116245841A (en) |
-
2023
- 2023-03-03 CN CN202310199278.5A patent/CN116245841A/en not_active Withdrawn
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109447154B (en) | Picture similarity detection method, device, medium and electronic equipment | |
| CN109886928B (en) | Target cell marking method, device, storage medium and terminal equipment | |
| CN111598913B (en) | Image segmentation method and system based on robot vision | |
| CN108764352B (en) | Method and device for detecting repeated page content | |
| CN115205223B (en) | Visual inspection method and device for transparent object, computer equipment and medium | |
| CN111325717B (en) | Mobile phone defect position identification method and equipment | |
| CN111444807B (en) | Target detection method, device, electronic equipment and computer readable medium | |
| US20230214989A1 (en) | Defect detection method, electronic device and readable storage medium | |
| CN111429482A (en) | Target tracking method and device, computer equipment and storage medium | |
| CN113609984A (en) | Pointer instrument reading identification method and device and electronic equipment | |
| US11164327B2 (en) | Estimation of human orientation in images using depth information from a depth camera | |
| CN118279304B (en) | Abnormal recognition method, device and medium for special-shaped metal piece based on image processing | |
| CN113255516A (en) | Living body detection method and device and electronic equipment | |
| Lauridsen et al. | Reading circular analogue gauges using digital image processing | |
| CN113537026B (en) | Method, device, equipment and medium for detecting graphic elements in building plan | |
| CN114359160A (en) | Screen detection method and device, electronic equipment and storage medium | |
| CN112287905A (en) | Vehicle damage identification method, device, equipment and storage medium | |
| CN116524477A (en) | Method and system for identifying detection result of detection box | |
| CN115222653B (en) | Test method and device | |
| CN116245841A (en) | Chest digital image heart-chest ratio measuring method for medical image specialty | |
| CN113034449B (en) | Target detection model training method and device and communication equipment | |
| CN111753625B (en) | Pedestrian detection method, device, equipment and medium | |
| CN112146834B (en) | Method and device for measuring structural vibration displacement | |
| CN114220011A (en) | Goods quantity identification method and device, electronic equipment and storage medium | |
| CN113627611A (en) | Model training method and device, electronic equipment and storage medium |
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 | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20230609 |