CN114359288A - Medical image cerebral aneurysm detection and positioning method based on artificial intelligence - Google Patents
Medical image cerebral aneurysm detection and positioning method based on artificial intelligence Download PDFInfo
- Publication number
- CN114359288A CN114359288A CN202210279446.7A CN202210279446A CN114359288A CN 114359288 A CN114359288 A CN 114359288A CN 202210279446 A CN202210279446 A CN 202210279446A CN 114359288 A CN114359288 A CN 114359288A
- Authority
- CN
- China
- Prior art keywords
- pixel
- image
- confidence
- medical image
- points
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 34
- 201000008450 Intracranial aneurysm Diseases 0.000 title claims abstract description 19
- 238000001514 detection method Methods 0.000 title claims abstract description 15
- 238000013473 artificial intelligence Methods 0.000 title claims abstract description 14
- 230000008859 change Effects 0.000 claims abstract description 22
- 230000001575 pathological effect Effects 0.000 claims abstract description 10
- 238000012545 processing Methods 0.000 claims abstract description 9
- 230000036285 pathological change Effects 0.000 claims abstract description 6
- 231100000915 pathological change Toxicity 0.000 claims abstract description 6
- 230000003902 lesion Effects 0.000 claims description 18
- 238000012549 training Methods 0.000 claims description 11
- 238000007781 pre-processing Methods 0.000 claims description 8
- 230000008569 process Effects 0.000 claims description 7
- 238000011156 evaluation Methods 0.000 claims description 6
- 238000004458 analytical method Methods 0.000 claims description 4
- 238000012937 correction Methods 0.000 claims description 4
- 238000010276 construction Methods 0.000 claims description 3
- 230000001502 supplementing effect Effects 0.000 claims description 3
- 238000002059 diagnostic imaging Methods 0.000 claims description 2
- 230000000717 retained effect Effects 0.000 claims description 2
- 230000003247 decreasing effect Effects 0.000 claims 1
- 206010028980 Neoplasm Diseases 0.000 description 10
- 230000011218 segmentation Effects 0.000 description 9
- 238000010606 normalization Methods 0.000 description 4
- 230000000694 effects Effects 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 238000003062 neural network model Methods 0.000 description 2
- 206010002329 Aneurysm Diseases 0.000 description 1
- 241000287196 Asthenes Species 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
Images
Abstract
The invention relates to a medical image cerebral aneurysm detection method based on artificial intelligence, which obtains a first confidence map of a medical image to be detected; extracting an interested region of the medical image to be detected, obtaining an eight-neighborhood image of the interested region according to the interested region, and further obtaining a reference image; replacing the pixel points of the interested region with the pixel points of the reference image to obtain a reference medical image and a second confidence map; making a difference between the first part confidence coefficient graph and the second part confidence coefficient graph to obtain an eight-neighborhood confidence coefficient change graph; carrying out binarization processing on the image data, and carrying out curve fitting on pixel points with the pixel values of 1 to obtain N straight lines; and judging whether the N straight lines have intersection points in the region of interest, if the intersection points exist and the confidence coefficient of the intersection points is greater than a set threshold value, the intersection points are pathological change points, and the medical image to be detected is a pathological image. The method additionally analyzes the confidence coefficient change of the neighborhood of the target region, improves the judgment accuracy, and reduces the probability of misjudgment.
Description
Technical Field
The invention relates to the field of medical image processing, in particular to a medical image cerebral aneurysm detection and positioning method based on artificial intelligence.
Background
The existing method for detecting and positioning the cerebral aneurysm based on medical image processing adopts a neural network and combines network optimization methods such as a self-attention mechanism and the like to realize the detection and positioning of the cerebral aneurysm. However, enough lesion images are needed as training samples, and it is difficult to acquire a large number of lesion images in practice, and due to different attributes such as lesion degree and lesion position, the generalization ability of the trained detection network is poor when lesion is identified under the condition of a small number of training samples.
Disclosure of Invention
The invention aims to provide a medical image cerebral aneurysm detection and positioning method based on artificial intelligence, which is used for solving the problem that a trained detection network has poor generalization capability when a lesion is identified under the condition of few training samples.
The invention provides a technical scheme of a medical image cerebral aneurysm detection and positioning method based on artificial intelligence, which comprises the following steps:
acquiring a medical image to be detected, and preprocessing the medical image to be detected to obtain a first confidence map;
extracting an interested region of a medical image to be detected, and obtaining an eight-neighborhood image of the interested region according to the interested region; combining the eight neighborhoods to obtain eight-channel neighborhood images, and inputting the eight-channel neighborhood images into a network prediction model to obtain corresponding reference images; replacing the pixel points of the interested region in the medical image to be detected with the pixel points of the reference image to obtain a reference medical image, and preprocessing the reference medical image to obtain a second confidence map;
extracting a first partial confidence map corresponding to an eight-neighborhood image of a medical image to be detected and a second partial confidence map corresponding to an eight-neighborhood image of a reference medical image, and subtracting the first partial confidence map and the second partial confidence map to obtain an eight-neighborhood confidence change map; carrying out binarization processing on the eight-neighborhood confidence coefficient change map, setting the pixel point greater than 0 to be 1, and carrying out curve fitting on the pixel point with the pixel value of 1 to obtain N fitted straight lines, wherein N is greater than or equal to 1;
judging whether the N straight lines have intersection points in the region of interest, if the intersection points exist and the confidence coefficient of the intersection points is greater than a set threshold value, the intersection points are pathological change points, and the medical image to be detected is a pathological image; otherwise, if no lesion point exists, the medical image to be detected is a normal image.
Further, the method for extracting the region of interest of the medical image to be detected comprises the following steps:
selecting the first k pixels with high confidence coefficients in the first confidence coefficient image to form a first pixel point set;
calculating the Euclidean distance between any two pixels in the first pixel set, and when the Euclidean distance is smaller than a preset threshold value, combining the two pixels into a single pixel, wherein the pixel value of the single pixel is the pixel value of any one of the two pixels; supplementing the pixels which do not form the set with the first pixel set, and continuing merging judgment until the number of elements in the first pixel set is k, so as to obtain a new first pixel set;
construction of
And a sliding window with the size is adopted to extract the surrounding information of each pixel point of the new first pixel point set, and the confidence coefficient mean value of the extracted area is used as an evaluation index
If the index is evaluated
Then to
For optimal template size, otherwise, order
，
Until obtaining the optimal sliding window size;
and extracting the region of the medical image to be detected according to the optimal sliding window size to obtain the region of interest with the same size as the optimal sliding window.
Further, the training process of the network prediction model is as follows:
acquiring a plurality of real pathological medical images and non-pathological medical images, randomly selecting the central point of each medical image, acquiring a central image and eight neighborhood images according to the optimal sliding window size, taking the central image as annotation data, and taking the eight neighborhood images as input training samples of a network prediction model to train the network prediction model.
Further, the obtaining process of the fitted N straight lines is as follows:
1) respectively performing straight line fitting on pixel points with pixel values of 1 in the eight-neighborhood confidence coefficient change image to obtain a first straight line;
2) respectively calculating the distance from each pixel point to the first straight line, obtaining the distance mean value of all the pixel points, when the distance mean value is larger than or equal to the fitting degree threshold value, sequentially extracting the pixel points with the largest distance until the fitting degree of the remaining pixel points is smaller than the fitting degree threshold value, counting the number of the extracted pixel points, when the number is larger than the set number, performing straight line fitting on the extracted pixel points to obtain a second straight line, calculating to obtain the corresponding distance mean value, judging whether the fitting degree threshold value is larger than or equal to the fitting degree threshold value, and repeating until the pixel points of all the fitted straight lines meet the fitting degree threshold value, and obtaining N straight lines.
And further, screening the lesion points according to the confidence coefficient change value gradient of pixel points included in the fitted straight line, if the gradient from the intersection point in the region of interest to the outside is in a descending trend, keeping the intersection point, and if not, deleting the intersection point. Further, the method also comprises the following step of correcting the first confidence corresponding to each pixel point in the region of interest according to the intersection point:
1) calculating Euclidean distance between the intersection point and each pixel point to be set as
，
Is shown as
The number of the intersection points is equal to or greater than the number of the intersection points,
is shown as
Each pixel point;
2) confidence correction value of each intersection
Comprises the following steps:
the corrected first confidence of each pixel point is
Wherein
Is the first confidence of each pixel point in the first confidence map,
the total number of the intersection points.
Further, before comparing the confidence of the intersection point with a set threshold, normalization processing is further performed on the confidence of the pixel point.
Further, a category image is obtained according to the determined lesion points and the corrected first confidence coefficient map, connected domain analysis is conducted on the category image to obtain a plurality of connected domains, the connected domains with the number of pixel points smaller than the set number are screened out, and the central points of the surrounding frames of the rest connected domains are used as locating points.
The invention has the following beneficial effects: compared with a mode of judging directly through the confidence coefficient, the confidence coefficient graphs before and after replacement are obtained through replacement operation, the confidence coefficient change of the target area neighborhood is additionally analyzed, the tendency based on the neighborhood confidence coefficient change assists the confidence coefficient of the pixel point of the target area to correct to a certain degree, the judgment accuracy rate is improved, and meanwhile the probability of misjudgment and misjudgment is reduced.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions and advantages of the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and other drawings can be obtained by those skilled in the art without creative efforts.
Fig. 1 is a flowchart of a method for detecting cerebral aneurysm based on artificial intelligence medical imaging according to the present invention.
Detailed Description
To further explain the technical means and effects of the present invention adopted to achieve the predetermined objects, the embodiments, structures, features and effects thereof according to the present invention will be described in detail below with reference to the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" refers to not necessarily the same embodiment. Furthermore, the particular features, structures, or characteristics may be combined in any suitable manner in one or more embodiments.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
The following describes a specific scheme of the medical image cerebral aneurysm detection method based on artificial intelligence in detail with reference to the accompanying drawings.
Specifically, referring to fig. 1, a flowchart of a method for detecting a cerebral aneurysm based on artificial intelligence in a medical image according to an embodiment of the present invention is shown, where the method includes the following steps:
step 1, acquiring a medical image to be detected, and preprocessing the medical image to be detected to obtain a first confidence map.
The medical image to be measured in this embodiment is a CT image.
The CT image preprocessing method comprises the following steps:
1) constructing a semantic segmentation network, wherein the semantic segmentation network is of an encoder-decoder structure; training the semantic segmentation network by using the label data to obtain a trained semantic segmentation network;
the trained label data were: the plurality of CT images are composed of a plurality of CT images with cerebral aneurysm and a plurality of CT images without cerebral aneurysm, and the label data is marked in a pixel level.
The loss function of the semantic segmentation network adopts binary cross entropy as the loss function.
Since the semantic segmentation network is a well-known network, the training process thereof will not be described in detail here.
2) Inputting an image to be processed into a trained semantic segmentation network, and outputting a semantic segmentation image to obtain a first confidence map;
it should be noted that, when the semantic segmentation image is obtained, each pixel point has a corresponding category, and the process of obtaining the category outputs the category confidence of each pixel point for the networkEach pixel point has two category confidence levels (tumor category and non-tumor category), and the category corresponding to the maximum confidence level is output as the category of the pixel point after being processed by a Softmax function; and extracting the confidence degrees corresponding to the tumor body categories of the pixel points, and forming a first confidence degree graph by the confidence degrees corresponding to the tumor body categories of all the pixel points, wherein the pixel value range of each pixel point in the first confidence degree graph is
。
Step 2, extracting an interested region of the medical image to be detected, and obtaining an eight-neighborhood image of the interested region according to the interested region; combining the eight neighborhoods to obtain eight-channel neighborhood images, and inputting the eight-channel neighborhood images into a network prediction model to obtain corresponding reference images; and replacing the pixel points of the interested region in the medical image to be detected with the pixel points of the reference image to obtain a reference medical image, and preprocessing the reference medical image to obtain a second confidence map.
In this embodiment, the extraction of the region of interest may be performed by using a connected domain analysis method; as another embodiment, the following steps may also be adopted to extract the region of interest, specifically:
1) selecting the first k pixels with high confidence coefficients in the first confidence coefficient image to form a first pixel point set;
2) calculating the Euclidean distance between any two pixels in the first pixel set, and when the Euclidean distance is smaller than a preset threshold value, combining the two pixels into a single pixel, wherein the pixel value of the single pixel is the pixel value of any one of the two pixels; supplementing the pixels which do not form the set with the first pixel set, and continuing merging judgment until the number of elements in the first pixel set is k, so as to obtain a new first pixel set;
it should be noted that, here
May be based on the implementerThe detection precision is adaptively adjusted, and the higher the required detection precision is, the higher the detection precision is
The larger the value, the set to 10 in this application; preset threshold value
The method is used for merging the pixels with the closer distance in the first pixel point set, so that the pixels in the same aneurysm region can be prevented from being processed for multiple times, and a threshold value is preset in the embodiment
Is arranged as
Wherein, in the step (A),
width and height of the first confidence map.
3) Construction of
And a sliding window with the size is adopted to extract the surrounding information of each pixel point of the new first pixel point set, and the confidence coefficient mean value of the extracted area is used as an evaluation index
If the index is evaluated
Then to
For optimal template size, otherwise, order
，
Until obtaining the optimal sliding window size;
in the above-mentioned steps, the step of,
the initial value is set to be 5 and,
is shown in
The evaluation index obtained after the information is extracted by the size template,
for evaluation index threshold, 0.2 is set in this application.
4) And extracting the region of the medical image to be detected according to the optimal sliding window size to obtain the region of interest with the same size as the optimal sliding window.
Specifically, the extraction of the region of interest of the present invention may use a neural network model to extract features, which is not described herein too much since the neural network model is a prior art for extracting features. In addition, the present invention can define the size of the region of interest by determining the optimal sliding window size when extracting the region of interest.
Based on the extracted interested region, obtaining an eight-neighborhood image of the interested region, specifically, extracting the coordinates of a central pixel point of the interested region as
Then, the same size template is used to base on the coordinates of the pixel points
，
，
，
，
，
，
，
Taking the CT image as a center, and extracting an eight-neighborhood image from the CT image; and simultaneously, after Concat is carried out on the eight-channel neighborhood images, the eight-channel neighborhood images are obtained.
The reference image in this embodiment is obtained as follows: and outputting the reference image by taking the eight-channel neighborhood image as the input of the network prediction model.
The training of the network prediction model in the above is as follows: acquiring a plurality of real pathological medical images and non-pathological medical images, randomly selecting the central point of each medical image, acquiring a central image and eight neighborhood images according to the optimal sliding window size, taking the central image as annotation data, and taking the eight neighborhood images as input training samples of a network prediction model to train the network prediction model.
In this embodiment, the method for obtaining the second confidence map corresponding to the reference medical image is the same as the method for preprocessing in step 1, and redundant description is omitted here.
Step 3, extracting a first partial confidence map corresponding to the eight-neighborhood image of the medical image to be detected and a second partial confidence map corresponding to the eight-neighborhood image of the reference medical image, and subtracting the first partial confidence map and the second partial confidence map to obtain an eight-neighborhood confidence coefficient change map; carrying out binarization processing on the eight-neighborhood confidence coefficient change map, setting the pixel point greater than 0 to be 1, and carrying out curve fitting on the pixel point with the pixel value of 1 to obtain N fitted straight lines, wherein N is greater than or equal to 1;
the partial confidence map of the corresponding image extracted in the above step can be directly and correspondingly determined according to the pixel points and the optimal sliding window size in step 2, that is, the confidence of the pixel points at the corresponding positions is extracted by comparing the image with the confidence map, so as to form the partial confidence map.
In this embodiment, the difference between the first partial confidence map and the second partial confidence map is the difference between the confidences of the corresponding pixels in the two maps, so as to obtain a confidence change value.
It should be noted that, in the above steps, the eight-neighborhood confidence degree change map of the eight-neighborhood image is used as a reference factor, because the region of interest may be a tumor or a non-tumor, when the region of interest is a non-tumor, the confidence degree of the neighborhood image information is usually the same as that of the region of interest, but if the region of interest is a tumor, the neighborhood image information is affected by the region of interest, that is, the neighborhood image information changes, so that whether the region of interest is a tumor type can be laterally detected by using the change of the neighborhood information.
In this embodiment, the obtaining process of the fitted N straight lines is:
1) respectively performing straight line fitting on pixel points with pixel values of 1 in the eight-neighborhood confidence coefficient change image to obtain a first straight line;
2) respectively calculating the distance from each pixel point to the first straight line, obtaining the distance mean value of all the pixel points, when the distance mean value is larger than or equal to the fitting degree threshold value, sequentially extracting the pixel points with the largest distance until the fitting degree of the remaining pixel points is smaller than the fitting degree threshold value, counting the number of the extracted pixel points, when the number is larger than the set number, performing straight line fitting on the extracted pixel points to obtain a second straight line, calculating to obtain the corresponding distance mean value, judging whether the fitting degree threshold value is larger than or equal to the fitting degree threshold value, and repeating until the pixel points of all the fitted straight lines meet the fitting degree threshold value, and obtaining N straight lines.
The set number is 3; threshold of fitness
The setting is 5, which can be determined according to actual conditions.
Step 4, judging whether the N straight lines have intersection points in the region of interest, if the intersection points exist and the confidence coefficient of the intersection points is greater than a set threshold value, the intersection points are pathological change points, and the medical image to be detected is a pathological image; otherwise, if no lesion point exists, the medical image to be detected is a normal image.
In this embodiment, the lesion point is screened according to the gradient of the confidence coefficient change value of the pixel point included in the fitted straight line, if the gradient from the intersection point in the region of interest to the outside is in a descending trend, the intersection point is retained, otherwise, the intersection point is deleted. Wherein, the gradient of the intersection point in the region of interest which is downward is from the region of interest to the outside, and whether the gradient of the confidence coefficient change value of the formed intersection point is downward or not is judged.
Further, in order to be more accurate, in this embodiment, the method further includes correcting the first confidence corresponding to each pixel point in the region of interest according to the intersection:
1) calculating Euclidean distance between the intersection point and each pixel point to be set as
，
Is shown as
The number of the intersection points is equal to or greater than the number of the intersection points,
is shown as
Each pixel point;
2) confidence correction value of each intersection
Comprises the following steps:
the corrected first confidence of each pixel point is
Wherein
Is the first confidence of each pixel point in the first confidence map,
the total number of the intersection points.
And then comparing the confidence of the corrected pixel points with a set threshold value to determine whether the intersection point is a lesion point. Meanwhile, in the above steps, before comparing the confidence of the intersection point with the set threshold, normalization processing is performed on the confidence of the pixel points, that is, the pixel values of all the pixel points in the region of interest after correction are normalized, and after normalization, the pixel points with the normalization value of more than or equal to 0.5 are extracted as the lesion pixel points.
Based on the determined lesion pixel points, the tumor position can be located, specifically, a category image is obtained according to the determined lesion points and the corrected first confidence map, connected domain analysis is carried out on the category image to obtain a plurality of connected domains, the connected domains with the number of the pixel points smaller than the set number are screened, and the central points of the surrounding frames of the rest connected domains are used as locating points.
The setting data in this embodiment is 5; the category image is obtained by assigning pixel points corresponding to the pathological change points to the pathological change categories.
According to the scheme, the confidence maps before and after replacement are obtained through replacement operation, the confidence change of the neighborhood of the target region is additionally analyzed, the tendency based on the neighborhood confidence change assists the confidence of the pixel point of the target region to correct to a certain degree, the judgment accuracy is improved, the probability of misjudgment and misjudgment is reduced, and the CT image detection and positioning of the cerebral aneurysm can be accurate.
It should be noted that: the precedence order of the above embodiments of the present invention is only for description, and does not represent the merits of the embodiments. And specific embodiments thereof have been described above. Other embodiments are within the scope of the following claims. In some cases, the actions or steps recited in the claims may be performed in a different order than in the embodiments and still achieve desirable results. In addition, the processes depicted in the accompanying figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing may also be possible or may be advantageous.
The embodiments in the present specification are described in a progressive manner, and the same and similar parts among the embodiments are referred to each other, and each embodiment focuses on the differences from the other embodiments.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents, improvements and the like that fall within the spirit and principle of the present invention are intended to be included therein.
Claims (8)
1. A medical image cerebral aneurysm detection method based on artificial intelligence is characterized by comprising the following steps:
acquiring a medical image to be detected, and preprocessing the medical image to be detected to obtain a first confidence map;
extracting an interested region of a medical image to be detected, and obtaining an eight-neighborhood image of the interested region according to the interested region; combining the eight neighborhoods to obtain eight-channel neighborhood images, and inputting the eight-channel neighborhood images into a network prediction model to obtain corresponding reference images; replacing the pixel points of the interested region in the medical image to be detected with the pixel points of the reference image to obtain a reference medical image, and preprocessing the reference medical image to obtain a second confidence map;
extracting a first partial confidence map corresponding to an eight-neighborhood image of a medical image to be detected and a second partial confidence map corresponding to an eight-neighborhood image of a reference medical image, and subtracting the first partial confidence map and the second partial confidence map to obtain an eight-neighborhood confidence change map; carrying out binarization processing on the eight-neighborhood confidence coefficient change map, setting the pixel point greater than 0 to be 1, and carrying out curve fitting on the pixel point with the pixel value of 1 to obtain N fitted straight lines, wherein N is greater than or equal to 1;
judging whether the N straight lines have intersection points in the region of interest, if the intersection points exist and the confidence coefficient of the intersection points is greater than a set threshold value, the intersection points are pathological change points, and the medical image to be detected is a pathological image; otherwise, if no lesion point exists, the medical image to be detected is a normal image.
2. The method for detecting cerebral aneurysm based on artificial intelligence medical image according to claim 1, wherein the method for extracting the region of interest of the medical image to be detected is as follows:
selecting the first k pixels with high confidence coefficients in the first confidence coefficient image to form a first pixel point set;
calculating the Euclidean distance between any two pixels in the first pixel set, and when the Euclidean distance is smaller than a preset threshold value, combining the two pixels into a single pixel, wherein the pixel value of the single pixel is the pixel value of any one of the two pixels; supplementing the pixels which do not form the set with the first pixel set, and continuing merging judgment until the number of elements in the first pixel set is k, so as to obtain a new first pixel set;
construction of
And a sliding window with the size is adopted to extract the surrounding information of each pixel point of the new first pixel point set, and the information is taken as the confidence coefficient average value of the extracted areaEvaluation index
If the index is evaluated
Then to
For optimal template size, otherwise, order
，
Until an optimal sliding window size is obtained, wherein
Is shown in
The evaluation index obtained after the information is extracted by the size template,
is an evaluation index threshold value;
and extracting the region of the medical image to be detected according to the optimal sliding window size to obtain the region of interest with the same size as the optimal sliding window.
3. The method for detecting cerebral aneurysm based on artificial intelligence medical image of claim 2, wherein the training process of the network prediction model is as follows:
acquiring a plurality of real pathological medical images and non-pathological medical images, randomly selecting the central point of each medical image, acquiring a central image and eight neighborhood images according to the optimal sliding window size, taking the central image as annotation data, and taking the eight neighborhood images as input training samples of a network prediction model to train the network prediction model.
4. The method for detecting cerebral aneurysm based on artificial intelligence medical image of claim 1, wherein the obtaining procedure of the fitted N straight lines is as follows:
respectively performing straight line fitting on pixel points with pixel values of 1 in the eight-neighborhood confidence coefficient change image to obtain a first straight line;
respectively calculating the distance from each pixel point to the first straight line, obtaining the distance mean value of all the pixel points, when the distance mean value is larger than or equal to the fitting degree threshold value, sequentially extracting the pixel points with the largest distance until the fitting degree of the remaining pixel points is smaller than the fitting degree threshold value, counting the number of the extracted pixel points, when the number is larger than the set number, performing straight line fitting on the extracted pixel points to obtain a second straight line, calculating to obtain the corresponding distance mean value, judging whether the fitting degree threshold value is larger than or equal to the fitting degree threshold value, and repeating until the pixel points of all the fitted straight lines meet the fitting degree threshold value, and obtaining N straight lines.
5. The method of claim 1, wherein the lesion points are selected according to a gradient of confidence level change values of pixel points included in the fitted straight line, and if the gradient of the intersection point in the region of interest is decreasing, the intersection point is retained, otherwise, the intersection point is deleted.
6. The method according to claim 1 or 5, further comprising modifying the first confidence level of each pixel point in the region of interest according to the intersection:
1) calculating Euclidean distance between the intersection point and each pixel point to be set as
，
Is shown as
The number of the intersection points is equal to or greater than the number of the intersection points,
is shown as
Each pixel point;
2) confidence correction value of each intersection
Comprises the following steps:
the corrected first confidence of each pixel point is
Wherein
Is the first confidence of each pixel point in the first confidence map,
the total number of the intersection points.
7. The method of claim 6, further comprising normalizing the confidence levels of the pixels before comparing the confidence level of the intersection with a predetermined threshold.
8. The method for detecting cerebral aneurysm based on artificial intelligence medical imaging of claim 7, wherein a category image is obtained according to the determined lesion point and the corrected first confidence map, a connected domain analysis is performed on the category image to obtain a plurality of connected domains, the connected domains with the number of pixels smaller than the set number are screened out, and the central point of the bounding box of each remaining connected domain is used as a positioning point.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210279446.7A CN114359288B (en) | 2022-03-22 | 2022-03-22 | Medical image cerebral aneurysm detection and positioning method based on artificial intelligence |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210279446.7A CN114359288B (en) | 2022-03-22 | 2022-03-22 | Medical image cerebral aneurysm detection and positioning method based on artificial intelligence |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114359288A true CN114359288A (en) | 2022-04-15 |
| CN114359288B CN114359288B (en) | 2022-06-07 |
Family
ID=81094474
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210279446.7A Expired - Fee Related CN114359288B (en) | 2022-03-22 | 2022-03-22 | Medical image cerebral aneurysm detection and positioning method based on artificial intelligence |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114359288B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114549532A (en) * | 2022-04-27 | 2022-05-27 | 珠海市人民医院 | Cerebral ischemia auxiliary analysis method and system based on medical image processing |
| CN115359051A (en) * | 2022-10-19 | 2022-11-18 | 江苏诺阳家居科技有限公司 | Aneurysm identification method based on pattern identification |
| CN117611926B (en) * | 2024-01-22 | 2024-04-23 | 重庆医科大学绍兴柯桥医学检验技术研究中心 | Medical image recognition method and system based on AI model |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20070025592A1 (en) * | 2005-07-27 | 2007-02-01 | Kabushiki Kaisha Toshiba | Target-region detection apparatus, method and program |
| US20120070052A1 (en) * | 2009-02-27 | 2012-03-22 | Commissariat A L'energie Atomique Et Aux Energies Alternatives | Methods for segmenting images and detecting specific structures |
| CN107507201A (en) * | 2017-09-22 | 2017-12-22 | 深圳天琴医疗科技有限公司 | A kind of medical image cutting method and device |
| CN108596893A (en) * | 2018-04-24 | 2018-09-28 | 东北大学 | a kind of image processing method and system |
| CN110796656A (en) * | 2019-11-01 | 2020-02-14 | 上海联影智能医疗科技有限公司 | Image detection method, image detection device, computer equipment and storage medium |
| CN111428812A (en) * | 2020-04-14 | 2020-07-17 | 沈阳先进医疗设备技术孵化中心有限公司 | Construction method and device of medical image data |
| CN112508783A (en) * | 2020-11-19 | 2021-03-16 | 西安全志科技有限公司 | Image processing method based on directional interpolation, computer device and computer readable storage medium |
| CN113011398A (en) * | 2021-04-28 | 2021-06-22 | 北京邮电大学 | Target change detection method and device for multi-temporal remote sensing image |
-
2022
- 2022-03-22 CN CN202210279446.7A patent/CN114359288B/en not_active Expired - Fee Related
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20070025592A1 (en) * | 2005-07-27 | 2007-02-01 | Kabushiki Kaisha Toshiba | Target-region detection apparatus, method and program |
| US20120070052A1 (en) * | 2009-02-27 | 2012-03-22 | Commissariat A L'energie Atomique Et Aux Energies Alternatives | Methods for segmenting images and detecting specific structures |
| CN107507201A (en) * | 2017-09-22 | 2017-12-22 | 深圳天琴医疗科技有限公司 | A kind of medical image cutting method and device |
| CN108596893A (en) * | 2018-04-24 | 2018-09-28 | 东北大学 | a kind of image processing method and system |
| CN110796656A (en) * | 2019-11-01 | 2020-02-14 | 上海联影智能医疗科技有限公司 | Image detection method, image detection device, computer equipment and storage medium |
| CN111428812A (en) * | 2020-04-14 | 2020-07-17 | 沈阳先进医疗设备技术孵化中心有限公司 | Construction method and device of medical image data |
| CN112508783A (en) * | 2020-11-19 | 2021-03-16 | 西安全志科技有限公司 | Image processing method based on directional interpolation, computer device and computer readable storage medium |
| CN113011398A (en) * | 2021-04-28 | 2021-06-22 | 北京邮电大学 | Target change detection method and device for multi-temporal remote sensing image |
Non-Patent Citations (6)
| Title |
|---|
| ZHEN JIE HUANG ET AL: "Medical Image Segmentation of Blood Vessels Based on Clifford Algebra and Voronoi Diagram", 《JOURNAL OF SOFTWARE》 * |
| ZHEN JIE HUANG ET AL: "Medical Image Segmentation of Blood Vessels Based on Clifford Algebra and Voronoi Diagram", 《JOURNAL OF SOFTWARE》, vol. 13, no. 6, 30 June 2018 (2018-06-30), pages 361 - 373 * |
| 李茂桐 等: "颅内动脉瘤破裂风险评估的研究进展", 《医学综述》 * |
| 李茂桐 等: "颅内动脉瘤破裂风险评估的研究进展", 《医学综述》, vol. 25, no. 18, 30 September 2019 (2019-09-30), pages 3662 - 3666 * |
| 耿辰 等: "基于深度学习的时间飞跃法磁共振血管成像对脑动脉瘤自动检测方法研究", 《上海医学》 * |
| 耿辰 等: "基于深度学习的时间飞跃法磁共振血管成像对脑动脉瘤自动检测方法研究", 《上海医学》, vol. 43, no. 5, 31 December 2020 (2020-12-31), pages 274 - 279 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114549532A (en) * | 2022-04-27 | 2022-05-27 | 珠海市人民医院 | Cerebral ischemia auxiliary analysis method and system based on medical image processing |
| CN115359051A (en) * | 2022-10-19 | 2022-11-18 | 江苏诺阳家居科技有限公司 | Aneurysm identification method based on pattern identification |
| CN115359051B (en) * | 2022-10-19 | 2023-12-15 | 江苏诺阳家居科技有限公司 | Aneurysm identification method based on pattern identification |
| CN117611926B (en) * | 2024-01-22 | 2024-04-23 | 重庆医科大学绍兴柯桥医学检验技术研究中心 | Medical image recognition method and system based on AI model |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114359288B (en) | 2022-06-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN115018828B (en) | Defect detection method for electronic component | |
| CN113689428B (en) | Mechanical part stress corrosion detection method and system based on image processing | |
| CN111626190B (en) | Water level monitoring method for scale recognition based on clustering partition | |
| CN114359288B (en) | Medical image cerebral aneurysm detection and positioning method based on artificial intelligence | |
| CN106446896B (en) | Character segmentation method and device and electronic equipment | |
| CN115330800B (en) | Automatic segmentation method of radiotherapy target area based on image processing | |
| CN108537751B (en) | Thyroid ultrasound image automatic segmentation method based on radial basis function neural network | |
| CN110276763B (en) | Retina blood vessel segmentation map generation method based on credibility and deep learning | |
| CN106372624B (en) | Face recognition method and system | |
| CN112215790A (en) | KI67 index analysis method based on deep learning | |
| CN114820625B (en) | Automobile top block defect detection method | |
| CN114897896B (en) | Building wood defect detection method based on gray level transformation | |
| CN115797352B (en) | Tongue picture image processing system for traditional Chinese medicine health-care physique detection | |
| CN115994907B (en) | Intelligent processing system and method for comprehensive information of food detection mechanism | |
| CN112819821A (en) | Cell nucleus image detection method | |
| CN115439654A (en) | Method and system for finely dividing weakly supervised farmland plots under dynamic constraint | |
| CN109147932B (en) | Cancer cell HER2 gene amplification analysis method and system | |
| CN112258536B (en) | Integrated positioning and segmentation method for calluses and cerebellum earthworm parts | |
| CN114638800A (en) | Improved Faster-RCNN-based head shadow mark point positioning method | |
| CN116993736B (en) | Intelligent identification system for osteoporosis area | |
| CN112258532B (en) | Positioning and segmentation method for callus in ultrasonic image | |
| Gunawan et al. | Fuzzy Region Merging Using Fuzzy Similarity Measurement on Image Segmentation | |
| CN114066887B (en) | Rice chalkiness area detection method, device, equipment and storage medium | |
| CN111127400A (en) | Method and device for detecting breast lesions | |
| CN114972272A (en) | Grad-CAM-based segmentation method for new coronary pneumonia lesions |
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 | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20220607 |