CN119048502B - A method and system for assisting in recognizing nursing images in reproductive medicine - Google Patents
A method and system for assisting in recognizing nursing images in reproductive medicine Download PDFInfo
- Publication number
- CN119048502B CN119048502B CN202411525789.2A CN202411525789A CN119048502B CN 119048502 B CN119048502 B CN 119048502B CN 202411525789 A CN202411525789 A CN 202411525789A CN 119048502 B CN119048502 B CN 119048502B
- Authority
- CN
- China
- Prior art keywords
- follicle
- suspected
- follicular
- image
- frame
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0012—Biomedical image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/50—Image enhancement or restoration using two or more images, e.g. averaging or subtraction
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/187—Segmentation; Edge detection involving region growing; involving region merging; involving connected component labelling
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/60—Analysis of geometric attributes
- G06T7/62—Analysis of geometric attributes of area, perimeter, diameter or volume
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H30/00—ICT specially adapted for the handling or processing of medical images
- G16H30/40—ICT specially adapted for the handling or processing of medical images for processing medical images, e.g. editing
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Radiology & Medical Imaging (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Epidemiology (AREA)
- Public Health (AREA)
- Primary Health Care (AREA)
- Geometry (AREA)
- Quality & Reliability (AREA)
- Image Analysis (AREA)
Abstract
The invention relates to the technical field of image data processing, in particular to a nursing image auxiliary identification method and system for a reproductive medicine department. The method comprises the steps of obtaining connected domains of follicular images, obtaining the possibility of suspected follicular regions of each connected domain according to the edge morphological characteristics of each connected domain, screening out the suspected follicular regions of each frame of follicular images, analyzing the position distribution and morphological characteristics of the suspected follicular regions among different frames of follicular images to obtain multiple groups of matched suspected follicular regions, obtaining the same follicular possibility of each group of matched suspected follicular regions, screening out the same follicular marking regions on different frames of follicular images, and improving the analysis accuracy of follicular development conditions by accurately identifying the corresponding same follicular regions in different frames of follicular images.
Description
Technical Field
The invention relates to the technical field of image data processing, in particular to a nursing image auxiliary identification method and system for a reproductive medicine department.
Background
When the follicle is mature in a female body during auxiliary reproduction, the method is particularly important, the change condition of the follicle is accurately identified, and the success rate of auxiliary reproduction can be improved, so that a vaginal ultrasound technology is generally adopted to obtain an follicular image in reproductive medical science and care, and the follicle is further identified.
In the prior art, the follicle identification is carried out on the acquired image by adopting an edge detection method, namely a closed edge curve is found, and a fixed characteristic template is used for positioning the follicle area, but as a plurality of follicles exist in the image and the morphological characteristics of the follicles in different development periods are different, only a certain fixed template is used for identification, erroneous judgment and missed judgment can possibly occur, the determination of the development follicle area is inaccurate, and the follicle identification effect is poor.
Disclosure of Invention
In order to solve the technical problems of inaccurate determination of a development follicular region and poor follicular recognition effect, the invention aims to provide a nursing image auxiliary recognition method and system for a reproductive medicine department, and the adopted technical scheme is as follows:
the invention provides a nursing image auxiliary identification method for a reproductive medicine department, which comprises the following steps:
acquiring multiple frames of follicular images of a pregnancy-prepared patient according to a time sequence;
for any frame of follicular image, obtaining a connected domain of the follicular image according to the gray distribution of pixel points on the follicular image; obtaining the possibility of suspected follicle areas of each connected domain according to the edge morphological characteristics of each connected domain; screening out the suspected follicle area of each frame of follicle image according to the possibility of the suspected follicle area;
Obtaining a plurality of groups of matched suspected follicle areas according to the position distribution and morphological characteristics of the suspected follicle areas among different frames of follicle images; obtaining the same follicle possibility of each group of matched suspected follicle areas according to the position change characteristics and the morphological change characteristics of each group of matched suspected follicle areas;
And screening out the same follicle marking area on different frames of follicle images according to the same follicle possibility.
Further, the method for acquiring the likelihood of the suspected follicle region comprises the following steps:
for each connected domain, acquiring a difference value of 8 chain code values between each edge pixel point and a previous adjacent edge pixel point as a chain code difference value;
And obtaining the probability of the suspected follicle region of each connected domain according to the absolute value of the average chain code difference value between all the edge pixel points and the previous adjacent edge pixel point and the fluctuation characteristic of the chain code difference value, wherein the absolute value of the average chain code difference value and the fluctuation characteristic of the chain code difference value are in negative correlation with the probability of the suspected follicle region.
Further, the method for acquiring the suspected follicle region comprises the following steps:
and for any frame of follicular image, if the probability of the suspected follicular region of the connected domain is greater than or equal to a preset probability threshold, taking the corresponding connected domain as the suspected follicular region.
Further, the method for acquiring the matching suspected follicle region comprises the following steps:
obtaining centroid coordinates of each suspected follicle region of each frame of follicle image;
Constructing a neighborhood range taking the centroid coordinates of each suspected follicle region of a previous frame of follicle image as the center in each frame of follicle image, and if the number of pixels in the suspected follicle region in the corresponding neighborhood range is greater than that of pixels in the suspected follicle region corresponding to the previous frame of follicle image, taking the suspected follicle region corresponding to the neighborhood range as a candidate suspected follicle region;
And if the similar suspected follicle areas exist in the follicle images of different frames, forming a group of matched suspected follicle areas.
Further, the method for acquiring the same follicle likelihood comprises the following steps:
For any group of matched suspected follicle areas, obtaining the ratio of the number of pixel points in the matched suspected follicle areas between each frame of follicle image and the previous frame of follicle image, and taking the ratio as the morphological increasing rate of the corresponding matched suspected follicle areas of each frame of follicle image;
obtaining a difference value of centroid coordinates in a matched suspected follicle area between each frame of follicle image and a previous frame of follicle image as a displacement vector of the matched suspected follicle area corresponding to each frame of follicle image;
and obtaining the same follicle possibility of the corresponding group of matched suspected follicle areas according to the difference of the morphological increase rate and the difference of the displacement vector of the corresponding matched suspected follicle areas between the adjacent frame follicle images, wherein the difference of the morphological increase rate and the difference of the displacement vector are in negative correlation with the same follicle possibility.
Further, the method for acquiring the same follicle-marking region comprises the following steps:
And if the same follicle probability of a group of matched suspected follicle areas is greater than or equal to a preset same follicle threshold value, taking the corresponding matched suspected follicle areas on different frame follicle images as the same follicle marking area.
Further, the method further comprises the following steps of:
and obtaining the excellent degree of each identical follicle marking area according to the morphological characteristics of each identical follicle marking area on the last frame of follicle image and the possibility of corresponding to the identical follicle, and screening out the optimal follicle area.
Further, the method for obtaining the goodness comprises the following steps:
And fusing the number of pixels in each same follicle-marking area and the possibility of the same follicle for the last frame of follicle image to obtain the good degree corresponding to the same follicle-marking area, wherein the number of pixels and the possibility of the same follicle are positively correlated with the good degree.
Further, the method for acquiring the optimal follicle region comprises the following steps:
And selecting the best numerical value in the good degree of all the same follicle-marking areas, and taking the corresponding same follicle-marking area as the optimal follicle area.
The invention also provides a nursing image auxiliary identification system for the reproductive medicine department, which comprises a memory, a processor and a computer program stored in the memory and capable of running on the processor, wherein the processor realizes the steps of any nursing image auxiliary identification method for the reproductive medicine department when executing the computer program.
The invention has the following beneficial effects:
The method comprises the steps of obtaining a connected domain of a follicular image according to gray level distribution of pixel points on the follicular image, obtaining the possibility of a suspected follicular region of each connected domain according to edge morphological characteristics of each connected domain due to continuous and smooth circular boundaries, screening out the suspected follicular region of each frame of follicular image according to the possibility of the suspected follicular region, determining the region most likely to be a follicular, reducing the range of subsequent processing, obtaining a plurality of groups of matched suspected follicular regions according to position distribution and morphological characteristics of the suspected follicular region between different frames of follicular images, obtaining the similarity of the same follicular region of each group of matched suspected follicular region according to the position change characteristics and the morphological change characteristics of each group of matched suspected follicular region, quantifying the similarity of the same follicular region of each group of matched follicular region, screening out the suspicious follicular region according to the similarity, and analyzing the same follicular region according to the difference of the similarity of the same follicular region, wherein the change of the same follicular region is continuous and regular, and the invention is beneficial to the improvement of the accuracy of the development of the same follicular region.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions and advantages of the prior art, the following description will briefly explain the drawings used in the embodiments or the description of the prior art, and it is obvious that the drawings in the following description are only some embodiments of the invention, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flowchart of a method for assisted identification of a nursing image for a reproductive medical science, according to an embodiment of the present invention;
FIG. 2 is a schematic diagram of an image of a follicle according to an embodiment of the present invention;
FIG. 3 is a flowchart of a method for acquiring a matching suspected follicle area according to an embodiment of the present invention;
fig. 4 is a flowchart of a method for obtaining a same follicle likelihood according to an embodiment of the present invention.
Detailed Description
In order to further describe the technical means and effects adopted by the invention to achieve the preset aim, the following detailed description refers to the specific implementation, structure, characteristics and effects of the auxiliary identification method and system for nursing images in reproductive medicine department according to the invention, which are provided by the invention, with reference to the accompanying drawings and the preferred embodiments. In the following description, different "one embodiment" or "another embodiment" means that the embodiments are not necessarily the same. Furthermore, the particular features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.
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 invention provides a specific scheme of a nursing image auxiliary identification method and a system for a reproductive medical science department, which are specifically described below with reference to the accompanying drawings.
Referring to fig. 1, a flowchart of a method for assisted identification of a nursing image of a reproductive medical science, according to an embodiment of the present invention, specifically includes:
step S1, acquiring multi-frame follicular images of a pregnancy-prepared patient according to a time sequence.
In order to assist a sterile couple in completing conception by using a medical assistance method, the follicular growth of a pregnant patient needs to be monitored. Firstly, planning the frequency of carrying out yin-ultrasonic detection on a patient according to the menstrual cycle stage of the patient and the clinical manifestation of the patient, generally putting a B-ultrasonic probe into the vagina of the pregnant patient to observe all pelvic structures after the 8 th day of the menstrual cycle of the pregnant patient, ensuring that the positions of the probes are the same every time in order to reduce subsequent errors, obtaining the yin-ultrasonic detection of the same positions for a plurality of times, analyzing the follicular condition and the development degree, and acquiring multi-frame follicular images of the pregnant patient according to the time sequence.
In the embodiment of the present invention, the number of frames for obtaining the follicular image is 3 or more, and the analysis is performed until the follicular is mature, wherein the same method is used for the analysis for obtaining the multiple frames of follicular images subsequently.
In other embodiments of the present invention, the size of the time interval may be specifically set according to specific situations, and is not limited and described herein.
Step S2, for any frame of follicular image, acquiring connected domains of the follicular image according to gray level distribution of pixel points on the follicular image, acquiring suspected follicular region possibility of each connected domain according to edge morphological characteristics of each connected domain, and screening out suspected follicular regions of each frame of follicular image according to the suspected follicular region possibility.
The interior of the follicular region is an anechoic region, the gray value difference between the follicular region and the surrounding low or high echo region is obvious, and the follicular region and the surrounding low or high echo region are in a quasi-circular bubble shape, as shown in fig. 2, a schematic diagram of a follicular image is shown, and a part marked by a numerical sequence number is displayed as a follicular, so that for any frame of follicular image, a connected region of the follicular image is obtained according to the gray distribution of pixel points on the follicular image, the interference of noise on analysis is reduced, and the subsequent independent processing and analysis of each region are facilitated.
In one embodiment of the present invention, the method for acquiring the connected domain includes performing edge detection on each frame of follicle image to obtain an edge curve in each frame of follicle image, and taking a region surrounded by the closed edge curve as the connected domain. It should be noted that specific edge detection is a technical means well known to those skilled in the art, and will not be described herein.
Because the follicular region is in a shape similar to a circular bubble, the circular boundary is continuous and smooth, the direction change between the adjacent edge pixel points is small, and the change is relatively uniform, so that the possibility of the suspected follicular region of each connected domain is obtained according to the edge morphological characteristics of each connected domain.
Preferably, in one embodiment of the present invention, the method for acquiring the likelihood of the suspected follicle region includes:
The 8-chain code of the edge pixel points can represent the boundary trend of the connected domains, and for each connected domain, the difference value of the 8-chain code value between each edge pixel point and the previous adjacent edge pixel point is obtained and used as the chain code difference value;
And obtaining the probability of the suspected follicle region of each connected domain according to the absolute value of the average chain code difference value between all the edge pixel points and the previous adjacent edge pixel point and the fluctuation characteristic of the chain code difference value, wherein the absolute value of the average chain code difference value and the fluctuation characteristic of the chain code difference value are in negative correlation with the probability of the suspected follicle region.
It should be noted that, in an embodiment of the present invention, the average value of the chain code differences between all the edge pixel points and the previous adjacent edge pixel point, that is, the average chain code difference value, is calculated to represent the overall change level of the edge of the connected domain, the smaller the absolute value of the average chain code difference value, the closer the chain code value between the adjacent edge pixel points, the smaller the direction change between the adjacent edge pixel points, the smoother the edge, the more likely the follicle region, and the greater the likelihood of the suspected follicle region, and the negative correlation relationship is presented.
In one embodiment of the invention, the fluctuation feature is represented by calculating variance, the larger the variance is, the larger the fluctuation feature is, the smaller the fluctuation feature is, the more uniform the change of the chain code difference value is, the more likely the change is a follicle area, the greater the possibility of suspected follicles is, the negative correlation relation is represented, in other embodiments of the invention, the fluctuation feature can be represented by adopting methods such as extremely poor and standard deviation, and the specific means are technical means well known to those skilled in the art and are not repeated herein.
In one embodiment of the present invention, taking a connected domain as an example, the formula of the likelihood of a suspected follicular region is expressed as:
;
Wherein, A likelihood of a suspected follicular region representing a connected domain; representing average chain code difference values between all edge pixel points on the connected domain and the previous adjacent edge pixel points; representing absolute values of average chain code differences between all edge pixel points on the connected domain and the previous adjacent edge pixel points; representing the variance of the difference value of the chain codes between all edge pixel points on the connected domain and the previous adjacent edge pixel point, namely the fluctuation characteristic; an exponential function based on a natural constant is represented.
In the formula of the probability of the suspected follicular region, the natural constant-based exponential function is used for theNegative correlation mapping is carried out, the smaller the absolute value of the average chain code difference value is, the smaller the fluctuation characteristic is, the more uniform the edge region change between adjacent edge pixel points is, the smoother the edge change of the communicating region is, the more likely is the follicular region, and the greater the likelihood of the suspected follicular region is, conversely, the larger the absolute value of the average chain code difference value is, the greater the fluctuation characteristic is, the more unstable the edge region change between the adjacent edge pixel points is, the more likely is bending or turning of the edge change of the communicating region, and the less likely is the suspected follicular region.
In other embodiments of the present invention, the likelihood of the suspected follicular region of the connected domain may be reflected by analyzing the circularity of the connected domain, and the greater the circularity, the more likely it is the suspected follicular region, and the greater the likelihood of the suspected follicular region, the specific means are technical means known to those skilled in the art, and will not be described herein.
By analyzing the likelihood of the suspected follicular region of each connected domain, the connected domain with higher likelihood is considered as the region likely to be a follicular region, the follicular can be effectively identified and positioned, and the accuracy and reliability of the follicular region can be improved. The suspicious follicle region of each frame of follicle image is screened out according to the possibility of the suspicious follicle region.
Preferably, in one embodiment of the present invention, the method for acquiring a suspected follicle region includes:
and for any frame of follicular image, if the probability of the suspected follicular region of the connected domain is greater than or equal to a preset probability threshold, taking the corresponding connected domain as the suspected follicular region.
It should be noted that, in one embodiment of the present invention, the size of the preset likelihood threshold is 0.7, and in other embodiments of the present invention, the size of the preset likelihood threshold may be specifically set according to specific situations, which is not limited and described herein.
And step S3, obtaining a plurality of groups of matched suspected follicle areas according to the position distribution and the morphological characteristics of the suspected follicle areas among different frame follicle images, and obtaining the same follicle possibility of each group of matched suspected follicle areas according to the position change characteristics and the morphological change characteristics of each group of matched suspected follicle areas.
The development of the follicle is a continuous process, from the original follicle to the preovulation follicle, the follicle in each development stage has specific morphological characteristics and position distribution, but the displacement and morphological change of the follicle are stable in a short time, so that the positions and morphological characteristics of suspected follicle areas in different frame follicle images can be observed to match corresponding follicle areas according to the continuity and change rule of the follicle areas. And obtaining a plurality of groups of matched suspected follicle areas according to the position distribution and morphological characteristics of the suspected follicle areas among different frames of follicle images.
Preferably, in an embodiment of the present invention, referring to fig. 3, a flowchart of a method for acquiring a matching suspected follicle region is shown, including:
step S301, obtaining the centroid coordinates of each suspected follicle area of each frame of follicle image.
To ensure consistency and comparability of analysis between different frames of follicular images, the same operation is used to construct a coordinate system for the follicular images with the left side edge of each frame of follicular image as the vertical axis of the coordinate system and the lower side edge perpendicular to the vertical axis as the horizontal axis of the coordinate system. In one embodiment of the present invention, the centroid coordinate is obtained by calculating the mean value of the position coordinates of all the pixel points in the suspected follicle area as the centroid coordinate.
Step S302, constructing a neighborhood range taking the centroid coordinates of each suspected follicle region of a previous frame of follicle image as the center in each frame of follicle image, if the number of pixels in the suspected follicle region in the corresponding neighborhood range is larger than that of pixels in the suspected follicle region corresponding to the previous frame of follicle image, taking the suspected follicle region corresponding to the neighborhood range as a candidate suspected follicle region, and selecting the candidate suspected follicle region with the minimum relative distance between the centroid coordinates and the center position of the neighborhood range as the similar suspected follicle region of the suspected follicle region corresponding to the previous frame of follicle image.
The follicles in the follicle images of different frames are continuously developed, in the short-time development process of the follicles, the changes of the displacement and the form are relatively stable, and the displacement rate is smaller, so that the number of pixel points is necessarily larger than that of the pixel points of the suspicious follicle region corresponding to the follicle image of the previous frame, and the relative distance is relatively smaller, so that the follicles of the same type are more likely to be in development change.
In one embodiment of the present invention, the method for obtaining the relative distance is to calculate the euclidean distance, and in other embodiments of the present invention, the method for calculating the relative distance may also be used to obtain the relative distance by using the existing distance calculation method such as the manhattan distance, and the specific means are technical means known to those skilled in the art, which are not described herein.
In one embodiment of the present invention, the neighborhood range is obtained by taking the centroid coordinate as the center and constructing the radius asThe circle range of (2) is used as a neighborhood range, wherein the radius can be specifically set by an implementer according to specific conditions, and if no suspected follicle area exists in the circle range, no matching exists; in other embodiments of the present invention, the size of the neighborhood range may be specifically set according to specific situations, which is not limited and described herein.
Step S303, if the similar suspected follicle areas exist in the follicle images of different frames in each suspected follicle area, a group of matched suspected follicle areas is formed.
Due to the factors of operation, environmental influence and easy interference of the B ultrasonic, similar round bubble interference items possibly exist in the follicular development process, and the fact that the interference items randomly appear and continuous variation trend does not exist is considered, similar suspected follicular areas exist in different frame follicular images in each suspected follicular area is selected for matching analysis.
The same follicle may have certain displacement and growth law in different frame images, the follicle gradually moves towards the ovary wall, the displacement distance is not large in a short time, the follicle gradually grows and becomes large along with time, the growth rate and the displacement rate of the follicle are relatively stable in a short time, the follicle changes continuously along with time, and the change trend of the follicle is reflected by matching the position change characteristics and the form change characteristics among the suspected follicle areas, so that the same follicle possibility of each group of matched suspected follicle areas is obtained according to the position change characteristics and the form change characteristics of each group of matched suspected follicle areas.
Preferably, in one embodiment of the present invention, referring to fig. 4, a flowchart of a method for acquiring the same follicle likelihood is shown, including:
step S401, for any group of matched suspected follicle areas, obtaining the ratio of the number of pixels in the matched suspected follicle areas between each frame of follicle image and the previous frame of follicle image, and taking the ratio as the morphological increasing rate of the matched suspected follicle areas corresponding to each frame of follicle image.
The number of the pixel points in the matched suspected follicle area can reflect the shape area of the matched suspected follicle area, the larger the number of the pixel points is, the larger the area is, the shape is relatively larger, the larger the ratio of the number of the pixel points in the matched suspected follicle area between each frame of follicle image and the previous frame of follicle image is, the shape change condition of the matched suspected follicle area can be reflected, and the larger the ratio is, the smaller the number of the pixel points in the matched suspected follicle area of the corresponding previous frame is, and the shape increase rate is larger.
In one embodiment of the invention, the equation for the morphology enhancement rate can be expressed as: Wherein, the method comprises the steps of, wherein, Represent the firstThe frame follicle image is correspondingly matched with the morphological increasing rate of the suspected follicle area; Represent the first The frame follicle image is correspondingly matched with the number of pixel points in the suspected follicle area; Represent the first The frame follicle image is correspondingly matched with the number of pixel points in the suspected follicle area; the number of frames of the follicle image is represented.
Step S402, obtaining a difference value of centroid coordinates in a matched suspected follicle area between each frame of follicle image and a previous frame of follicle image as a displacement vector of the matched suspected follicle area corresponding to each frame of follicle image.
By calculating the difference value of the centroid coordinates, the displacement of the follicle in the image can be accurately quantified, and the displacement vector can be obtained to reflect the growth speed and direction of the follicle, and the larger the difference value of the centroid coordinates is, the farther the distance between the centroid coordinates is, the faster the follicle grows.
It should be noted that, in one embodiment of the present invention, the formula of the displacement vector is expressed as: wherein, the method comprises the steps of, Represent the firstThe frame follicle image is correspondingly matched with the displacement vector of the suspected follicle area; Represent the first The frame follicle image is correspondingly matched with the centroid abscissa of the suspected follicle region; Represent the first The frame follicle image is correspondingly matched with the centroid abscissa of the suspected follicle region; Represent the first The frame follicle image is correspondingly matched with the centroid ordinate of the suspected follicle area; Represent the first The frame follicle image corresponds to the centroid ordinate of the matching suspected follicle region.
Step S403, according to the difference of the morphological increasing rate and the difference of the displacement vector of the corresponding matched suspected follicle areas between the adjacent frame follicle images, the same follicle possibility of the corresponding group matched suspected follicle areas is obtained, and the difference of the morphological increasing rate and the difference of the displacement vector are in negative correlation with the same follicle possibility.
The difference of the morphological increase rate can reflect the morphological change condition of the matched suspected follicle area between the adjacent frame follicle images, and the larger the difference of the morphological increase rate is, the less similar the morphological change is, the worse the stability is, the less the probability of the same follicle is, and the negative correlation relation is formed.
It should be noted that, since the displacement vector is a vector and includes a size and a direction, in order to analyze the displacement change condition of the matching suspected follicle area between the adjacent frame follicle images, in one embodiment of the present invention, the difference of the mode lengths of the displacement vectors of the matching suspected follicle area between the adjacent frame follicle images and the included angle of the displacement vector are analyzed to represent the difference of the displacement vectors, the larger the difference of the mode lengths of the displacement vectors is, the larger the included angle of the displacement vectors is, the less the displacement change of the matching suspected follicle area between the adjacent frame follicle images is, and the same follicle probability is smaller, i.e. the difference of the displacement vectors and the same follicle probability are in a negative correlation relationship.
In one embodiment of the invention, the formula for the likelihood of the same follicle for any set of matching suspected follicular regions is expressed as:
;
Wherein, Representing a set of identical follicle likelihoods that match a suspected follicle region; Represent the first The frame follicle image is correspondingly matched with the morphological increasing rate of the suspected follicle area; Represent the first The frame follicle image is correspondingly matched with the morphological increasing rate of the suspected follicle area; Represent the first The frame follicle image is correspondingly matched with the displacement vector of the suspected follicle area; Represent the first The frame follicle image is correspondingly matched with the displacement vector of the suspected follicle area; Represent the first Frame follicular image and the firstIncluded angles of displacement vectors of the corresponding matching suspected follicle areas between the frame follicle images; Represent the first Frame follicular image and the firstAn included angle cosine value of the displacement vector of the corresponding matching suspected follicle region between the frame follicle images; Represent the first The frame follicle image is correspondingly matched with the modular length of the displacement vector of the suspected follicle area; Represent the first The frame follicle image is correspondingly matched with the modular length of the displacement vector of the suspected follicle area; the number of frames of the follicle image is represented.
In the same formula of follicular possibilities,Analyzing the overall change condition among all the follicle images of adjacent frames by solving an average value, and mapping the included angle by a cosine value function, wherein the larger the included angle is, the smaller the cosine value of the included angle is, and the smaller the probability of the same follicle is; Representing calculation number Frame follicular image and the firstThe absolute value of the difference value of the modular length of the displacement vectors of the corresponding matched suspected follicle areas between the frame follicle images is smaller, the distance of the displacement vectors is closer, and the probability of the same follicle is higher; Representing calculation number Frame follicular image and the firstThe absolute value of the difference value of the morphological increase rate of the corresponding matched suspected follicle area between the frame follicle images is larger, the morphological increase rate is not consistent, and the probability of the same follicle is smaller.
And S4, screening out the same follicle marking area on different frames of follicle images according to the same follicle possibility.
By analyzing the same follicle probability of each group of matched suspected follicle regions, the connected regions with higher probability are regarded as regions with the same follicle change, so that the follicles can be effectively identified and positioned, and the accuracy and reliability of the follicle regions can be improved. The same follicular marker region on different frames of follicular images is selected according to the same follicular probability.
Preferably, in one embodiment of the present invention, the method for acquiring the same follicle-marker region includes:
And if the same follicle probability of a group of matched suspected follicle areas is greater than or equal to a preset same follicle threshold value, taking the corresponding matched suspected follicle areas on different frame follicle images as the same follicle marking area.
It should be noted that, in one embodiment of the present invention, the size of the preset follicle threshold is 0.75, and in other embodiments of the present invention, the size of the preset follicle threshold is a technical means well known to those skilled in the art, which is not limited and described herein in detail.
It should be noted that, after the same follicular marking area is obtained, the method further includes obtaining the quality degree of each same follicular marking area according to the morphological characteristics of each same follicular marking area on the last frame of follicular image and the possibility of corresponding to the same follicular marking area, and screening out the optimal follicular area.
The good follicles usually have round and full shapes and are stable in development, so that the larger the morphological characteristics are and the more the changes among different frames are consistent, the larger the possibility of the same follicle is, and the higher the good degree is, so that the good degree of each follicle is obtained according to the morphological characteristics of each same follicle marking area on the last frame of follicle image and the possibility of the same follicle corresponding to the same follicle, and the optimal follicle area is screened out.
Preferably, in one embodiment of the present invention, the method for obtaining the goodness includes:
And fusing the number of pixels in each same follicle-marking area and the possibility of the same follicle for the last frame of follicle image to obtain the good degree corresponding to the same follicle-marking area, wherein the number of pixels and the possibility of the same follicle are positively correlated with the good degree.
In some embodiments of the present invention, the number of pixels in the follicular region and the probability of the same follicle may be fused by multiplication or addition, and the greater the number of pixels, the greater the probability of the same follicle and the greater the degree of follicular merit, and specific means are technical means well known to those skilled in the art, and will not be described herein.
In one embodiment of the present invention, the formula for the goodness of the last frame of the follicular image is expressed as:
;
Wherein, Represent the firstThe degree of well-being of the same follicular marking region; Representing the last frame of follicular image Is the first of (2)The number of pixel points in the same follicle marking area; Represent the first The same follicular region likelihood of a same follicular marker region.
In the formula of the fine degree, the larger the number of pixels is, the larger the possibility of the same follicular region is, the more stable the follicular development is, and the larger the fine degree is.
Preferably, in one embodiment of the present invention, the method for acquiring an optimal follicle region includes:
And selecting the best numerical value in the good degree of all the same follicle-marking areas, and taking the corresponding same follicle-marking area as the optimal follicle area.
In another embodiment of the present invention, the same follicle-marking area and the optimal follicle area can more intuitively reflect the growth and development states of the follicle, and after the same follicle-marking area and the optimal follicle area on different frame follicle images are obtained, the follicle can be more accurately identified and monitored, including: and carrying out local histogram homogenization treatment on the same follicular marking area on each frame of follicular image to obtain an enhanced follicular image, improving the brightness of the follicular marking area and the information display intensity of the follicular marking area while enabling the outline and the internal structure of the follicular marking area to be more obvious in the image, further combining the position of the optimal follicular marking area, and more clearly knowing the growth and development conditions of follicles between different frames of images and improving the recognition and monitoring effects of follicles.
In summary, the method comprises the steps of obtaining connected domains of follicular images, obtaining the possibility of suspected follicular regions of each connected domain according to the edge morphological characteristics of each connected domain, screening the suspected follicular regions of each frame of follicular image, analyzing the position distribution and morphological characteristics of the suspected follicular regions among different frames of follicular images to obtain multiple groups of matched suspected follicular regions, obtaining the same possibility of follicle of each group of matched suspected follicular regions, screening the same follicle marking regions on different frames of follicular images, and improving the analysis accuracy of follicular development conditions by accurately identifying the corresponding same follicle regions in different frames of follicular images.
The invention also provides a reproduction medical nursing image auxiliary identification system, which comprises a memory, a processor and a computer program stored in the memory and capable of running on the processor, wherein the processor realizes any one of the steps of the reproduction medical nursing image auxiliary identification method when executing the computer program.
It should be noted that the sequence of the embodiments of the present invention is only for description, and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.
In this specification, each embodiment is described in a progressive manner, and identical and similar parts of each embodiment are all referred to each other, and each embodiment mainly describes differences from other embodiments.
Claims (8)
1. A reproductive medical science nursing image auxiliary identification method, characterized in that the method comprises the following steps:
acquiring multiple frames of follicular images of a pregnancy-prepared patient according to a time sequence;
for any frame of follicular image, obtaining a connected domain of the follicular image according to the gray distribution of pixel points on the follicular image; obtaining the possibility of suspected follicle areas of each connected domain according to the edge morphological characteristics of each connected domain; screening out the suspected follicle area of each frame of follicle image according to the possibility of the suspected follicle area;
Obtaining a plurality of groups of matched suspected follicle areas according to the position distribution and morphological characteristics of the suspected follicle areas among different frames of follicle images; obtaining the same follicle possibility of each group of matched suspected follicle areas according to the position change characteristics and the morphological change characteristics of each group of matched suspected follicle areas;
Screening out the same follicle mark area on different frames of follicle images according to the same follicle possibility;
the method for acquiring the matched suspected follicle area comprises the following steps:
obtaining centroid coordinates of each suspected follicle region of each frame of follicle image;
Constructing a neighborhood range taking the centroid coordinates of each suspected follicle region of a previous frame of follicle image as the center in each frame of follicle image, and if the number of pixels in the suspected follicle region in the corresponding neighborhood range is greater than that of pixels in the suspected follicle region corresponding to the previous frame of follicle image, taking the suspected follicle region corresponding to the neighborhood range as a candidate suspected follicle region;
If the similar suspected follicle areas exist in the follicle images of different frames, a group of matched suspected follicle areas are formed;
The method for acquiring the possibility of the same follicle comprises the following steps:
For any group of matched suspected follicle areas, obtaining the ratio of the number of pixel points in the matched suspected follicle areas between each frame of follicle image and the previous frame of follicle image, and taking the ratio as the morphological increasing rate of the corresponding matched suspected follicle areas of each frame of follicle image;
obtaining a difference value of centroid coordinates in a matched suspected follicle area between each frame of follicle image and a previous frame of follicle image as a displacement vector of the matched suspected follicle area corresponding to each frame of follicle image;
and obtaining the same follicle possibility of the corresponding group of matched suspected follicle areas according to the difference of the morphological increase rate and the difference of the displacement vector of the corresponding matched suspected follicle areas between the adjacent frame follicle images, wherein the difference of the morphological increase rate and the difference of the displacement vector are in negative correlation with the same follicle possibility.
2. The assisted identification method of nursing images of reproductive medicine department according to claim 1, wherein the method for acquiring the possibility of the suspected follicular region comprises:
for each connected domain, acquiring a difference value of 8 chain code values between each edge pixel point and a previous adjacent edge pixel point as a chain code difference value;
And obtaining the probability of the suspected follicle region of each connected domain according to the absolute value of the average chain code difference value between all the edge pixel points and the previous adjacent edge pixel point and the fluctuation characteristic of the chain code difference value, wherein the absolute value of the average chain code difference value and the fluctuation characteristic of the chain code difference value are in negative correlation with the probability of the suspected follicle region.
3. The assisted identification method of nursing images of reproductive medicine department according to claim 1, wherein the method for acquiring suspected follicular region comprises:
and for any frame of follicular image, if the probability of the suspected follicular region of the connected domain is greater than or equal to a preset probability threshold, taking the corresponding connected domain as the suspected follicular region.
4. The assisted identification method of nursing images in reproductive medicine department according to claim 1, wherein the method for acquiring the same follicle marker region comprises the following steps:
And if the same follicle probability of a group of matched suspected follicle areas is greater than or equal to a preset same follicle threshold value, taking the corresponding matched suspected follicle areas on different frame follicle images as the same follicle marking area.
5. The assisted identification method of nursing images of reproductive medicine department according to claim 1, wherein after acquiring the same follicular marking region, further comprising:
and obtaining the excellent degree of each identical follicle marking area according to the morphological characteristics of each identical follicle marking area on the last frame of follicle image and the possibility of corresponding to the identical follicle, and screening out the optimal follicle area.
6. The assisted identification method of nursing images for reproductive medicine department according to claim 5, wherein the acquisition method of the goodness degree comprises the following steps:
And fusing the number of pixels in each same follicle-marking area and the possibility of the same follicle for the last frame of follicle image to obtain the good degree corresponding to the same follicle-marking area, wherein the number of pixels and the possibility of the same follicle are positively correlated with the good degree.
7. The assisted identification method of nursing images for reproductive medicine department according to claim 5, wherein the acquisition method of the optimal follicle area comprises the following steps:
And selecting the best numerical value in the good degree of all the same follicle-marking areas, and taking the corresponding same follicle-marking area as the optimal follicle area.
8. A reproductive medical nursing image auxiliary identification system, the system comprising a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the computer program, implements the steps of a reproductive medical nursing image auxiliary identification method according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202411525789.2A CN119048502B (en) | 2024-10-30 | 2024-10-30 | A method and system for assisting in recognizing nursing images in reproductive medicine |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202411525789.2A CN119048502B (en) | 2024-10-30 | 2024-10-30 | A method and system for assisting in recognizing nursing images in reproductive medicine |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN119048502A CN119048502A (en) | 2024-11-29 |
| CN119048502B true CN119048502B (en) | 2024-12-27 |
Family
ID=93587244
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202411525789.2A Active CN119048502B (en) | 2024-10-30 | 2024-10-30 | A method and system for assisting in recognizing nursing images in reproductive medicine |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN119048502B (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104966045A (en) * | 2015-04-02 | 2015-10-07 | 北京天睿空间科技有限公司 | Video-based airplane entry-departure parking lot automatic detection method |
| CN107253485A (en) * | 2017-05-16 | 2017-10-17 | 北京交通大学 | Foreign matter invades detection method and foreign matter intrusion detection means |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111192251B (en) * | 2019-12-30 | 2023-03-28 | 上海交通大学医学院附属国际和平妇幼保健院 | Follicle ultrasonic processing method and system based on level set image segmentation |
| CN114041166A (en) * | 2020-12-28 | 2022-02-11 | 深圳迈瑞生物医疗电子股份有限公司 | Method and system for tracking oocytes |
| CN118230367A (en) * | 2022-12-21 | 2024-06-21 | 北京眼神智能科技有限公司 | Palm vein recognition method, palm vein recognition device, storage medium and palm vein recognition equipment |
-
2024
- 2024-10-30 CN CN202411525789.2A patent/CN119048502B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104966045A (en) * | 2015-04-02 | 2015-10-07 | 北京天睿空间科技有限公司 | Video-based airplane entry-departure parking lot automatic detection method |
| CN107253485A (en) * | 2017-05-16 | 2017-10-17 | 北京交通大学 | Foreign matter invades detection method and foreign matter intrusion detection means |
Also Published As
| Publication number | Publication date |
|---|---|
| CN119048502A (en) | 2024-11-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109166133B (en) | Soft tissue organ image segmentation method based on key point detection and deep learning | |
| CN110338840B (en) | Three-dimensional imaging data display processing method and three-dimensional ultrasonic imaging method and system | |
| US8699766B2 (en) | Method and apparatus for extracting and measuring object of interest from an image | |
| CN110929728B (en) | Image region-of-interest dividing method, image segmentation method and device | |
| US20250025133A1 (en) | Ultrasonic imaging method and device | |
| WO2022140960A1 (en) | Follicle tracking method and system | |
| WO2020215485A1 (en) | Fetal growth parameter measurement method, system, and ultrasound device | |
| CN116912255B (en) | Follicular region segmentation method for ovarian tissue analysis | |
| Wang et al. | A method of ultrasonic image recognition for thyroid papillary carcinoma based on deep convolution neural network. | |
| US12159407B1 (en) | Optimization and enhancement method for medical health physical examination big data | |
| CN118379313B (en) | A medical image segmentation method for intracranial arteriosclerotic plaques in patients with atrial fibrillation | |
| CN119090901B (en) | An efficient detection method, system and device for craniopharyngioma lesions | |
| Faghih et al. | Automated ovarian follicular monitoring: A novel real-time approach | |
| Gao et al. | Deep endpoints focusing network under geometric constraints for end-to-end biometric measurement in fetal ultrasound images | |
| CN111986137B (en) | Biological organ lesion detection method, apparatus, device, and readable storage medium | |
| CN119048502B (en) | A method and system for assisting in recognizing nursing images in reproductive medicine | |
| Wulaning Ayu et al. | Pixel Classification Based on Local Gray Level Rectangle Window Sampling for Amniotic Fluid Segmentation. | |
| CN112750099A (en) | Follicle measurement method, ultrasound apparatus, and computer-readable storage medium | |
| CN117808813B (en) | Respiratory system lung nodule tumor cell image detection method and system and storage medium | |
| CN110063753B (en) | Method and device for measuring head circumference of fetus in ultrasonic image | |
| US20230225700A1 (en) | Cranial ultrasonic standard plane imaging and automatic detection and display method for abnormal regions | |
| CN113409275B (en) | Method for determining thickness of transparent layer behind fetal neck based on ultrasonic image and related device | |
| CN112561912B (en) | Medical image lymph node detection method based on priori knowledge | |
| CN112070089B (en) | A method and system for intelligent diagnosis of diffuse thyroid disease based on ultrasound images | |
| JP2024546482A (en) | Eye tracking method, device, equipment and storage medium used in anterior segment OCTA |
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 |