CN105496453B - A kind of ox ovarian follicle ultrasonic monitoring system and its monitoring method - Google Patents
A kind of ox ovarian follicle ultrasonic monitoring system and its monitoring method Download PDFInfo
- Publication number
- CN105496453B CN105496453B CN201510774242.0A CN201510774242A CN105496453B CN 105496453 B CN105496453 B CN 105496453B CN 201510774242 A CN201510774242 A CN 201510774242A CN 105496453 B CN105496453 B CN 105496453B
- Authority
- CN
- China
- Prior art keywords
- ovarian follicle
- image
- unit
- volume
- denoising
- 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
Abstract
The present invention provides a kind of ox ovarian follicle ultrasonic monitoring system and its monitoring method, including with lower unit:It reads and shows ultrasonoscopy unit, image cropping unit, image denoising unit, ovarian follicle detection unit, surface extraction unit, three-dimensional reconstruction unit, volume computing unit, volume sensing unit and point set rendering unit.The platform for handling ox ovarian follicle ultrasonoscopy and calculating the present invention provides one, program can carry out ultrasonoscopy denoising, detection ovarian follicle, extraction ovarian follicle surface, ovarian follicle three-dimensional reconstruction, the calculating of ovarian follicle volume and monitoring function.So as to monitor ox follicular development situation under different state of developments, and disease can be diagnosed and tracking and monitoring treatment, have many advantages, such as it is quick, efficiently, accuracy rate height, intuitive it is strong.
Description
Technical field
The present invention relates to a kind of ultrasonic monitoring system and its monitoring method, especially a kind of ox ovarian follicle ultrasonic monitoring system
And its monitoring method.
Background technology
Ultrasonic monitoring is the image subject that diagnose and treat is carried out using the physical characteristic of ultrasonic wave, is known as ultrasound doctor
It learns.Its clinical application range is extensive, has been known as indispensable diagnostic method in current clinical medicine at present.
It is reported at present in terms of the foundation to the development of the Monitoring Ovarian Function of ox, dominant follicle and ovarian cycle still few
See, it is especially more rare using advanced Modern ultrasound diagnostic technology as research means person.And different cultivars, different geographical
Ox physiological status be not quite similar, in order to disclose ovarian activity and hormonal change rule of the ox within the oestrous cycle, from
And it is the healthy and breeding for protecting ox, the economic loss of production unit is reduced, its breeding potential is improved and is given birth to for research ox
The animal reproductions such as reproductive physiology, estrus synchronization, embryo transfer biotechnology provide real time data, it is necessary to using Ultrasonic Diagnosis into
One step research.
The content of the invention
For the above problem of the prior art, it is a primary object of the present invention to provide a kind of ox ovarian follicle ultrasonic monitoring system
System and its monitoring method can read display ox ovarian follicle ultrasonoscopy, and further ultrasonoscopy handled and calculated,
So as to monitor ox follicular development situation under different state of developments, and disease can be diagnosed and tracking and monitoring treatment, have
Fast, efficiently, the advantages that accuracy rate is high, intuitive is strong.
In order to achieve the above object, the present invention adopts the following technical scheme that:A kind of ox ovarian follicle ultrasonic monitoring system, including
With lower unit:
It reads and shows ultrasonoscopy unit, for reading ultrasound image slices and the display under specified directory, and three
It ties up and supplies user from any angle with certain proportion Zoom display in window;
Image cropping unit, for carrying out cutting spatially to data set, user can choose given zone as needed
Image in domain;
For setting special parameter, denoising is carried out to the image set of loading for image denoising unit;
Ovarian follicle detection unit, on the premise of denoising, the ovarian follicle in image being detected and being shown;
Surface extraction unit, for ovarian follicle surface to be extracted and shown;
Three-dimensional reconstruction unit, for carrying out triangle fitting to surface point and showing;
Volume computing unit, for calculating ovarian follicle volume;
Volume sensing unit, for the data set continuously measured to be compared and monitored;
Point set rendering unit, for being shown to already present point set.
A kind of ox ovarian follicle ultrasonic monitoring method, includes the following steps:(1)It reads and shows ultrasonoscopy,(2)Image is cut out
It cuts,(3)Image denoising,(4)Ovarian follicle detects,(5)Surface extraction,(6)Three-dimensional reconstruction,(7)Volume calculates,(8)Volume detects,
(9)Point set renders.
As it is further preferably, in step(2)In image cropping, the single picture of selected digital image concentration, system provides
Interface allows user to choose ROI region with mouse;According to the coordinate for the rectangle that user chooses, new region is calculated compared with original picture
In X and the offset of Y-direction;According to offset, the origin position that every image is new in image set is calculated;The new images collection that will be obtained
Storage, cutting finish.
As it is further preferably, in step(3)In image denoising, traversal reads each frame of image set, by what is obtained
Data are stored in a three-dimensional matrice;Gray processing, normalization are carried out to image;Use improved anisotropy parameter algorithm pair
Image carries out denoising, improves calculating speed using Parallel Algorithm if necessary;Image after denoising is stored, in next step
Processing.
As it is further preferably, in step(4)In ovarian follicle detection, detection part uses the thought of Threshold segmentation, directly
Threshold process is carried out to three-dimensional matrice, and by image binaryzation, follicles regions pixel value is taken 1, other regions take 0, specific to walk
It is rapid as follows:The single picture in image set is taken, global threshold is obtained using maximum variance between clusters principle;Knot based on the first step
Fruit, to image set binaryzation;01 obtained matrix is stored, is handled for next step.
As it is further preferably, in step(5)In surface extraction, a 3x3x3 neighborhood under definition is three-dimensional, i.e. matrix
In 26 neighbor pixels around each pixel:Travel through each be not 0 pixel, calculate its neighborhood and S;If
S is 0, and it is isolated noise to illustrate the point, then is set to 0, if S is 26, it is the point inside follicles regions to illustrate him, then
It is set to 0;New matrix is obtained, is preserved for handling in next step.
As it is further preferably, in step(6)In three-dimensional reconstruction, point set that surface extraction obtains be discrete point, it is necessary to
It is fitted, can just be redeveloped into three-dimensional effect, step is as follows:Using Delaunay Tetrahedron subdivision algorithms, to input
Point set carry out Tetrahedron subdivision, obtain a sequence of point sets, adjacent two points describe a line in tetrahedron;Time
Each side of previous step is gone through, if the length of side meets apart from filter condition(Here maximum length is taken to be no more than unit 1), then protect
It stays, otherwise delete the side 2 points;Last point set is stored, the triangle point set in tetrahedron is rendered using OpenGL, is rebuild
It finishes.
As it is further preferably, in step(7)During volume calculates, cutd open since tetrahedron can be carried out to three-dimensional point set
Point, therefore ovarian follicle volume can also be calculated using the algorithm, step is as follows:The result point set that ovarian follicle is taken to be obtained after detecting, makes
Subdivision is carried out with Tetrahedron subdivision algorithm, which is accelerated than relatively time-consuming using parallel algorithm if necessary;
The single tetrahedral length of side is taken to calculate single tetrahedral volume V for unit 1;Four obtained in statistic procedure 1
The number of face body is N;V is multiplied with N, you can obtain the volume of ovarian follicle.
As it is further preferably, in step(8)In volume detection, using this algorithm, during to the difference of same ovarian follicle
The ultrasonic picture of phase is handled, and obtained result is stored, uses the volume data of line chart draw calculation.
Beneficial effects of the present invention are as follows:The processing of ox ovarian follicle ultrasonoscopy and calculating are put down the present invention provides one
Platform, program can carry out ultrasonoscopy denoising, detection ovarian follicle, extraction ovarian follicle surface, ovarian follicle three-dimensional reconstruction, the calculating of ovarian follicle volume
And monitoring function.So as to monitor ox follicular development situation under different state of developments, and disease can be diagnosed and with
Track monitoring treatment has many advantages, such as that quick, efficient, accuracy rate is high, intuitive is strong.
Specific embodiment
The realization, functional characteristics and advantageous effect of the object of the invention, do further below in conjunction with specific embodiment
It is bright.
Specific embodiment is described in further detail technical solution of the present invention below, so that the technology of this field
Personnel can be better understood from the present invention and can be practiced, but illustrated embodiment is not as a limitation of the invention.
Monitor Preparatory work of experiment:B ultrasound instrument, laptop, linear array transrectal probe are connected, by tested ox with six columns
Column Baoding, staff scribble couplant on probe, carefully slowly bring probe into rectum, first with hand strap pop one's head in across
Rectal wall searches out uterus to be checked, ovary, then by Ultrasonic-B probe close to uterus and ovary.And it needs during the inspection process constantly
The angle of conversion probe and ovary, to detect the ovarian follicle and corpus luteum of different position in ovary.It is selected when on the screen of host
During typical image, freeze frame transmits image, and measures required data with the electronics measurement ruler of B ultrasound.
Ox ovarian follicle ultrasonic monitoring system of the embodiment of the present invention, including with lower unit:
It reads and shows ultrasonoscopy unit, for reading ultrasound image slices and the display under specified directory, and three
It ties up and supplies user from any angle with certain proportion Zoom display in window;
Image cropping unit, for carrying out cutting spatially to data set, user can choose given zone as needed
Image in domain;
For setting special parameter, denoising is carried out to the image set of loading for image denoising unit;
Ovarian follicle detection unit, on the premise of denoising, the ovarian follicle in image being detected and being shown;
Surface extraction unit, for ovarian follicle surface to be extracted and shown;
Three-dimensional reconstruction unit, for carrying out triangle fitting to surface point and showing;
Volume computing unit, for calculating ovarian follicle volume;
Volume sensing unit, for the data set continuously measured to be compared and monitored;
Point set rendering unit, for being shown to already present point set.
Ox ovarian follicle ultrasonic monitoring method of the embodiment of the present invention, specifically comprises the following steps:
(1)It reads and shows ultrasonoscopy,
(2)Image cropping:For the cross-sectional image size of system input for 800x600 sizes, when actual treatment, need not be so
Big size, therefore image cropping is needed, specific steps:
1. the single picture that selected digital image is concentrated, system, which provides interface, allows user to choose ROI region with mouse;
2. according to the coordinate for the rectangle that user chooses, offset in X and Y-direction of the new region compared with original picture is calculated
Amount;
3. according to offset, the origin position that every image is new in image set is calculated;
4. obtained new images collection is stored, cutting finishes.
(3)Image denoising:Core is that denoising, tool are carried out to image using improved anisotropy parameter algorithm
Body step is as follows:
1. traversal reads each frame of image set, obtained data are stored in a three-dimensional matrice;
2. pair image carries out gray processing, normalization;
3. carrying out denoising to image using improved anisotropy parameter algorithm, carried if necessary using Parallel Algorithm
High calculating speed;
Image after denoising is stored, is handled for next step.
(4)Ovarian follicle detects:
Detection part uses the thought of Threshold segmentation, directly carries out threshold process to three-dimensional matrice, and by image binaryzation,
Follicles regions pixel value is taken 1, other regions take 0, are described in detail below:
1. choosing the single picture in image set, global threshold is obtained using maximum variance between clusters principle;
It is 2. based on the first step as a result, to image set binaryzation;
3. 01 obtained matrix is stored, handled for next step.
(5)Surface extraction:
It is that detection part obtains the result is that possible follicles regions, matrix data describes a solid area, rear
Continuous when rendering, we only need surface region, extraction process:
A 3x3x3 neighborhood under 1. definition is three-dimensional, i.e., 26 neighbor pixels around each pixel in matrix:
2. travel through each be not 0 pixel, calculate its neighborhood and S;
3. if S is 0, it is isolated noise to illustrate the point, then is set to 0, if S is 26, it is follicles regions to illustrate him
Internal point, then be also set to 0;
4. obtaining new matrix, preserve for handling in next step.
(6)Three-dimensional reconstruction:
The point set that surface extraction obtains be discrete point, it is necessary to be fitted to it, can just be redeveloped into three-dimensional effect, step
It is as follows:
1. using Delaunay Tetrahedron subdivision algorithms, Tetrahedron subdivision is carried out to the point set of input, obtains a point set
Sequence, adjacent two points describe a line in tetrahedron;
2. each side of previous step is traveled through, if the length of side meets apart from filter condition(Here maximum length is taken to be no more than
Unit 1), then retain, otherwise delete the side 2 points;
Last point set is stored, the triangle point set in tetrahedron is rendered using OpenGL, reconstruction finishes.
(7)Volume calculates:
Since Tetrahedron subdivision can be carried out to three-dimensional point set, therefore ovarian follicle volume can also be calculated using the algorithm,
Step is as follows:
1. the result point set that ovarian follicle is taken to be obtained after detecting carries out subdivision using Tetrahedron subdivision algorithm, which compares consumption
When, accelerated if necessary using parallel algorithm;
2. the single tetrahedral length of side is taken to calculate single tetrahedral volume V for unit 1;
3. the tetrahedral number obtained in statistic procedure 1 is N;
V is multiplied with N, you can obtain the volume of ovarian follicle.
(8)Volume detects:
Using this algorithm, the ultrasonic picture of the different times of same ovarian follicle is handled, obtained result is stored,
Use the volume data of line chart draw calculation.
(9)Point set renders.
Claims (8)
- A kind of 1. ox ovarian follicle ultrasonic monitoring method, it is characterised in that:Include the following steps:(1)It reads and shows ultrasonoscopy, (2)Image cropping,(3)Image denoising,(4)Ovarian follicle detects,(5)Surface extraction,(6)Three-dimensional reconstruction,(7)Volume calculates,(8)Body Product detection,(9)Point set renders;In step(5)In surface extraction, a 3x3x3 neighborhood under definition is three-dimensional, i.e., in matrix 26 around each pixel A neighbor pixel:Travel through each be not 0 pixel, calculate its neighborhood and S;If S is 0, illustrate that the point is isolated Noise, then be set to 0, if S is 26, it is the point inside follicles regions to illustrate him, then is also set to 0;It obtains new Matrix is preserved for handling in next step;The ox ovarian follicle ultrasonic monitoring method uses ox ovarian follicle ultrasonic monitoring system monitoring, the ox ovarian follicle ultrasonic monitoring System is included with lower unit:It reads and shows ultrasonoscopy unit, for reading ultrasound image slices and the display under specified directory, and in three-dimensional window Supply user from any angle with certain proportion Zoom display in mouthful;Image cropping unit, for carrying out cutting spatially to data set, user chooses the figure in specific region as needed Picture;For setting special parameter, denoising is carried out to the image set of loading for image denoising unit;Ovarian follicle detection unit, on the premise of denoising, the ovarian follicle in image being detected and being shown;Surface extraction unit, for ovarian follicle surface to be extracted and shown;Three-dimensional reconstruction unit, for carrying out triangle fitting to surface point and showing;Volume computing unit, for calculating ovarian follicle volume;Volume sensing unit, for the data set continuously measured to be compared and monitored;Point set rendering unit, for being shown to already present point set.
- 2. ox ovarian follicle ultrasonic monitoring method according to claim 1, it is characterised in that:In step(2)In image cropping, The single picture that selected digital image is concentrated, system, which provides interface, allows user to choose ROI region with mouse;The rectangle chosen according to user Coordinate, calculate new region compared with original picture in X and the offset of Y-direction;According to offset, every figure in image set is calculated As new origin position;Obtained new images collection is stored, cutting finishes.
- 3. ox ovarian follicle ultrasonic monitoring method according to claim 1, it is characterised in that:In step(3)In image denoising, Traversal reads each frame of image set, and obtained data are stored in a three-dimensional matrice;Gray processing, normalizing are carried out to image Change;Denoising is carried out to image using anisotropy parameter algorithm, the image after denoising is stored, is handled for next step.
- 4. ox ovarian follicle ultrasonic monitoring method according to claim 3, it is characterised in that:Parallelization is used during denoising Algorithm improves calculating speed.
- 5. ox ovarian follicle ultrasonic monitoring method according to claim 1, it is characterised in that:In step(4)In ovarian follicle detection, Detection part uses the thought of Threshold segmentation, directly carries out threshold process to three-dimensional matrice, and by image binaryzation, by ovarian follicle area Domain pixel value takes 1, other regions take 0, are as follows:The single picture in image set is taken, uses maximum variance between clusters Principle obtains global threshold;It is based on the first step as a result, to image set binaryzation;01 obtained matrix is stored, in next step Processing.
- 6. ox ovarian follicle ultrasonic monitoring method according to claim 1, it is characterised in that:In step(6)In three-dimensional reconstruction, The point set that surface extraction obtains be discrete point, it is necessary to be fitted to it, can just be redeveloped into three-dimensional effect, step is as follows:Make With Delaunay Tetrahedron subdivision algorithms, Tetrahedron subdivision is carried out to the point set of input, obtains a sequence of point sets, it is adjacent Two points describe a line in tetrahedron;Each side of previous step is traveled through, if the length of side meets apart from filter condition, Retain, otherwise delete the side 2 points;Last point set is stored, the triangle point set in tetrahedron is rendered using OpenGL, weight It builds and finishes.
- 7. ox ovarian follicle ultrasonic monitoring method according to claim 1, it is characterised in that:In step(7)During volume calculates, Step is as follows:The result point set that ovarian follicle is taken to be obtained after detecting carries out subdivision using Tetrahedron subdivision algorithm;It takes single tetrahedral The length of side is unit 1, calculates single tetrahedral volume V;The tetrahedral number obtained in statistics first step is N;By V and N is multiplied, you can obtains the volume of ovarian follicle.
- 8. ox ovarian follicle ultrasonic monitoring method according to claim 1, it is characterised in that:In step(8)In volume detection, Use the volume data of line chart draw calculation.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510774242.0A CN105496453B (en) | 2015-11-13 | 2015-11-13 | A kind of ox ovarian follicle ultrasonic monitoring system and its monitoring method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510774242.0A CN105496453B (en) | 2015-11-13 | 2015-11-13 | A kind of ox ovarian follicle ultrasonic monitoring system and its monitoring method |
Publications (2)
Publication Number | Publication Date |
---|---|
CN105496453A CN105496453A (en) | 2016-04-20 |
CN105496453B true CN105496453B (en) | 2018-05-25 |
Family
ID=55705023
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201510774242.0A Active CN105496453B (en) | 2015-11-13 | 2015-11-13 | A kind of ox ovarian follicle ultrasonic monitoring system and its monitoring method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN105496453B (en) |
Families Citing this family (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110246135B (en) * | 2019-07-22 | 2022-01-18 | 新名医(北京)科技有限公司 | Follicle monitoring method, device, system and storage medium |
CN112617891B (en) * | 2020-12-22 | 2023-02-10 | 居天智慧(深圳)有限公司 | Reproductive follicle detector |
WO2022140960A1 (en) * | 2020-12-28 | 2022-07-07 | 深圳迈瑞生物医疗电子股份有限公司 | Follicle tracking method and system |
Family Cites Families (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2009285315A (en) * | 2008-05-30 | 2009-12-10 | Panasonic Corp | Ultrasonic diagnostic apparatus |
CN104299238A (en) * | 2014-10-22 | 2015-01-21 | 北京航空航天大学 | Organ tissue contour extraction method based on medical image |
CN104574263A (en) * | 2015-01-28 | 2015-04-29 | 湖北科技学院 | Quick three-dimensional ultrasonic reconstruction and display method on basis of GPU (graphics processing unit) |
-
2015
- 2015-11-13 CN CN201510774242.0A patent/CN105496453B/en active Active
Also Published As
Publication number | Publication date |
---|---|
CN105496453A (en) | 2016-04-20 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110325119B (en) | Ovarian follicle count and size determination | |
EP2974665B1 (en) | Apparatus and method for supporting computer aided diagnosis (cad) based on probe speed | |
US10368833B2 (en) | Method and system for fetal visualization by computing and displaying an ultrasound measurement and graphical model | |
US11622744B2 (en) | System and method for automated ovarian follicular monitoring | |
CN105496453B (en) | A kind of ox ovarian follicle ultrasonic monitoring system and its monitoring method | |
US11389133B2 (en) | Method and apparatus for follicular quantification in 3D ultrasound images | |
CN113781439B (en) | Ultrasonic video focus segmentation method and device | |
US20130072797A1 (en) | 3d ultrasound apparatus and method for operating the same | |
Hu et al. | Automated placenta segmentation with a convolutional neural network weighted by acoustic shadow detection | |
CA3143181A1 (en) | Adaptive image processing method and system in assisted reproductive technologies | |
CN103142211B (en) | Heart magnetic signal processing method based on extreme value circle | |
Potočnik et al. | Computerized detection and recognition of follicles in ovarian ultrasound images: a review | |
Liu et al. | Automated classification and measurement of fetal ultrasound images with attention feature pyramid network | |
Liu et al. | Doppler ultrasound imaging combined with fetal heart detection in predicting fetal distress in pregnancy-induced hypertension under the guidance of artificial intelligence algorithm | |
WO2020215485A1 (en) | Fetal growth parameter measurement method, system, and ultrasound device | |
Zhao et al. | Transfsm: Fetal anatomy segmentation and biometric measurement in ultrasound images using a hybrid transformer | |
Faghih et al. | Automated ovarian follicular monitoring: A novel real-time approach | |
Yang et al. | Segmentation of the common carotid artery with active shape models from 3D ultrasound images | |
TWI722297B (en) | Internal edge detection system and method thereof for processing medical images | |
Hiremath et al. | Recognition of follicles in ultrasound images of ovaries using geometric features | |
WO2022140960A1 (en) | Follicle tracking method and system | |
CN111481233B (en) | Thickness measuring method for transparent layer of fetal cervical item | |
JP2017111129A (en) | Contour extraction device, contour extraction method and program | |
CN110916724A (en) | B-ultrasonic image fetal head circumference detection method based on closed loop shortest path | |
Gautam et al. | Review on ultrasound-image analysis of fetus using Deep Learning |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |