CN109801276A - A kind of method and device calculating ambition ratio - Google Patents
A kind of method and device calculating ambition ratio Download PDFInfo
- Publication number
- CN109801276A CN109801276A CN201910030726.2A CN201910030726A CN109801276A CN 109801276 A CN109801276 A CN 109801276A CN 201910030726 A CN201910030726 A CN 201910030726A CN 109801276 A CN109801276 A CN 109801276A
- Authority
- CN
- China
- Prior art keywords
- lung
- area
- interest
- heart
- bounding box
- 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
Abstract
The present invention relates to technical field of image processing, especially a kind of method and device for the ambition ratio for calculating chest medical image.Wherein method includes the following steps: to input chest medical image；Obtain the bounding box of left lung bounding box, right side lung bounding box and heart area；Left lung area-of-interest, right side lung area-of-interest and heart area-of-interest are extracted in bounding box；The minimum bounding box of lung's area-of-interest is determined according to left lung area-of-interest and right side lung area-of-interest；Human body long axis direction and human body X direction are determined according to lung's area-of-interest minimum bounding box；Determine the minimum bounding box of heart area-of-interest；The ratio of the transverse diameter length of heart area minimum bounding box transverse diameter length and lung's minimum bounding box is calculated, the ratio of the area of heart area-of-interest area and lung's area-of-interest is calculated.Such technical solution can Accurate Segmentation linked groups, realize and automatic calculate ambition ratio, ambition area ratio.
Description
Technical field
The present invention relates to technical field of image processing, more particularly to a kind of method and device for calculating ambition ratio.
Background technique
Medical imaging is a kind of advanced technology for understanding human organ health status, such as X-ray equipment, DR equipment, CT
The imaging devices such as equipment, MR equipment have been widely applied in clinical diagnosis, and doctor is helped to observe the tissue for understanding inside of human body
Anatomical structure and lesion anatomical location information etc..
In clinic, breast imaging is usually used in assessing the disease of heart and lung, and wherein ambition ratio and ambition area ratio are to face
Bed radiologist judges the important diagnostic quantizating index of cardiomegaly, and the definition of ambition ratio is the dirty laterally maximum of picture centre
The ratio of diameter and lung's transverse direction maximum diameter.Ambition area ratio is the long-pending ratio with lung's area of picture centre visceral surface.
The calculation method of the ambition ratio of early stage is doctor according to clinic browsing image experience, is manuallyd locate in Chest Image
The maximum transverse diameter of heart area and the maximum transverse diameter of lung calculate Ejection fraction.Ambition area ratio is by delineating in the picture
Heart area and lung areas are selected, the area in picture centre dirty district domain and the area of lung areas are obtained.It the label and delineates
Process is easy the influence of the person's of being affected by clinical experience He other subjective factors, and cumbersome, leads to the work of clinician
Efficiency is lower.
It is existing to carry out technology of the ambition than calculating using image processing techniques, be mainly based upon the grayscale information of image into
The segmentation of row heart and lung areas or edges of regions are extracted, and Ejection fraction, such as lung based on OTSU and heart area are calculated
Threshold segmentation and Boundary Extraction, such methods for gray threshold selection and calculate rely on it is larger, and segmentation result hold
Interference vulnerable to other adjacent tissues.Since the imaging device noise situations of different vendor are there are larger difference, and it is imaged
Quality and clarity have differences, these will have a direct impact on the stability and accuracy of calculated result.
The calculation method of the above ambition ratio can not meet high-precision simultaneously, high stability, adapt to different vendor's imaging device
It is required with automatic.
Convolutional neural networks are one of neural networks, and network can voluntarily extract figure during handling digital picture
The features such as the topological relation of the texture of picture, shape and image, especially scale, rotate, translation and the distortion of other forms it is constant
Property has good robustness.Relative to existing computer image processing technology, for large-scale data and clinical imaging knot
Fruit has good robustness and adaptability.
Summary of the invention
Brief summary of the present invention is given below, in order to provide the basic reason about certain aspects of the invention
Solution.It should be appreciated that this summary is not an exhaustive overview of the invention.It is not intended to determine key of the invention
Or pith, nor determining intended limitation the scope of the present invention.Its purpose is only to provide certain concepts in simplified form,
Taking this as a prelude to a more detailed description discussed later.
A primary object of the present invention is, providing a kind of can adapt to different clinical instrumentation imaging results accurately
Method and device of the ambition than calculating；
According to an aspect of the invention, there is provided an ambition, than computing device, which includes: input unit, for defeated
Enter normotopia of chest image；Acquiring unit, for determining the bounding box in left lung region, right side lung areas and heart area；
Extraction unit, for extracting the area-of-interest of segmentation left lung, right side lung and heart；First determination unit, for true
Determine the transverse diameter length of lung's area-of-interest；Second determination unit, for determining the transverse diameter length of heart area area-of-interest；Meter
Unit is calculated, for calculating Ejection fraction and cardial event.
According to an aspect of the invention, there is provided an ambition is than calculation method, this method comprises: input normotopia of chest
Image；Determine the bounding box in left lung region, right side lung areas and heart area；Extract segmentation left lung, right lung
The area-of-interest in portion and heart；Determine the transverse diameter length of lung's area-of-interest；Determine the transverse diameter of heart area area-of-interest
Length；Calculate Ejection fraction and cardial event.
In addition, the embodiments of the present invention also provide the computer programs for realizing above-mentioned calculation method.
In addition, the embodiments of the present invention also provide the computer program product of at least computer-readable medium form,
Upper record has the computer program code for realizing the above method.
By the detailed description below in conjunction with attached drawing to highly preferred embodiment of the present invention, these and other of the invention is excellent
Point will be apparent from.
Detailed description of the invention
Below with reference to the accompanying drawings illustrate embodiments of the invention, the invention will be more easily understood it is above and its
Its objects, features and advantages.Component in attached drawing is intended merely to show the principle of the present invention.In the accompanying drawings, identical or similar
Technical characteristic or component will be indicated using same or similar appended drawing reference.
Fig. 1 is to show structure of the automatic ambition according to an embodiment of the invention than the exemplary means of computing device
Block diagram；
Fig. 2 is to show flow chart of the automatic ambition according to an embodiment of the invention than calculation method example process；
Fig. 3 is the bounding box information of heart area according to an embodiment of the present invention, left lung region and right side lung areas
Schematic diagram；
Fig. 4 is that left lung area-of-interest according to an embodiment of the present invention, right side lung area-of-interest and lung are interested
The schematic diagram of region minimum bounding box；
Fig. 5 is the schematic diagram of the minimum bounding box of heart area-of-interest and area-of-interest according to an embodiment of the present invention.
Specific embodiment
The implementation that the present invention will now be explained with reference to the accompanying drawings.It is described in an attached drawing of the invention or a kind of embodiment
Elements and features can be combined with elements and features shown in one or more other drawings or embodiments.It should infuse
Unrelated to the invention, portion known to persons of ordinary skill in the art is omitted in attached drawing and explanation for purposes of clarity in meaning
The expression and description of part and processing.
Fig. 1 is the block diagram for showing exemplary configuration according to an embodiment of the invention.
Computing device as shown in Figure 1 includes input unit 110, acquiring unit 120, the determining list of extraction unit 130, first
First 140, second determination unit 150, computing unit 160.
Fig. 2 is the method flow diagram of the ambition ratio of illustrative detection subject according to an embodiment of the present invention.In step
S210, input unit 100 input the image of subject.Input unit 110 will judge whether the image of input is normotopia of chest
Image, if it is the image of normotopia of chest, then alignment is normalized, and enters step S220.The normotopia of chest figure of input
Picture.
Here, if it is determined that the image of input is not the image of normotopia of chest, the e.g. image at other positions of human body,
So the device not will do it subsequent detection and calculation processing.
Then, in step S220, acquiring unit 120 obtains left side according to the scheduled location algorithm of neural network, from image
The Rectangular Bounding Volume of lung areas, right side lung areas and heart area.
Here, scheduled neural network location algorithm can be Faster-RCNN, Yolo, SSD of existing deep learning
Scheduling algorithm can position the bounding box for the multiple tissues for needing to position in image, as shown in figure 3, left lung using the algorithm
The bounding box 330 of the bounding box 320 in region, the bounding box 330 of right side lung areas and heart area.Shown bounding box surrounds
Including interested rectangle, which can be bigger than area-of-interest.
Here, the neural network location algorithm is preferably Faster-RCNN, and deep neural network is preferably
ResNet101.The parameter model of the location algorithm can by training obtain, as collect clinic in normotopia of chest image and
Clinical expert marks the bounding box of left and right lung areas and heart area, just using marked bounding box location information and chest
Bit image data are iterated training to neural network as input, obtain the model of location algorithm.
In embodiment, by inventor's test of many times, it is believed that the flag data of 1000 normotopia of chest images can be preferable
Meet actual demand, more training datas, obtained location algorithm model robustness is higher.
In step S230, extraction unit 230 is according to partitioning algorithm model, respectively in left lung bounding box, right side lung
In bounding box and heart area bounding box, divide left lung area-of-interest, right lung area-of-interest and heart area-of-interest.
Here, predetermined partitioning algorithm can be the existing neural network partitioning algorithm based on deep learning, utilize the calculation
Method, can tissue area-of-interest to need to divide in the bounding box in segmented image, wherein left lung area-of-interest is such as
In Fig. 4 shown in 420, right side lung area-of-interest is as shown in 410 in Fig. 4, and heart area-of-interest is as shown in 550 in Fig. 5.Institute
The corresponding location of pixels of tissue in area-of-interest covering image shown.
Here, the neural network partitioning algorithm is preferably the partitioning algorithm based on UNet.The mould of the partitioning algorithm
Type can be obtained by training, such as collect normotopia of chest image sheet and handmarking's left lung region of interest in clinic
Domain, right side lung area-of-interest and heart area-of-interest, just using marked region of interest location information domain and chest
Bit image data are trained deep neural network, obtain the model of partitioning algorithm.
In embodiment, by inventor's test of many times, it is believed that the flag data of 1000 normotopia of chest images can be preferable
Meet actual demand, more training datas, obtained partitioning algorithm model robustness is higher.
By positioning the left lung bounding box of acquisition in step S220, left lung packet is intercepted from normotopia of chest image
The image in box is enclosed, left lung area-of-interest is divided by left lung neural network partitioning algorithm model, it is emerging to obtain sense
The confidence level in interesting region, and left lung area-of-interest is passed through into coordinate conversion map into original normotopia of chest image, such as
In Fig. 4 shown in 420.
By positioning the right side lung bounding box of acquisition in step S220, the interception right side lung packet from normotopia of chest image
The image in box is enclosed, right side lung area-of-interest is divided by right side lung neural network partitioning algorithm model, it is emerging to obtain sense
The confidence level in interesting region, and right side lung area-of-interest is passed through into the image where coordinate conversion map to original bounding box
In, as shown in 410 in Fig. 4.
By positioning the heart area bounding box of acquisition in step S220, heart area packet is intercepted from normotopia of chest image
The image in box is enclosed, heart area-of-interest is divided by heart area neural network partitioning algorithm model, obtains region of interest
The confidence level in domain, and by heart area area-of-interest by the image where coordinate conversion map to original bounding box, such as
In figure shown in 550.
Here, left lung area-of-interest, right side lung area-of-interest obtained in step S230 and heart sense
Interest region is selected if there is overlapping, lap position according to the corresponding confidence level in lap position when each region segmentation, choosing
The high classification of confidence level is selected as the overlapping region position generic.
In step S240, the first determination unit 140 determines the minimum bounding box of lung's area-of-interest, determines human body axis
Direction and human body transverse diameter direction, determine the transverse diameter length of lung's area-of-interest.
Here, lung's area-of-interest, by the left lung area-of-interest and step S230 that are extracted in step S230
The right side lung area-of-interest of extraction is composed.
Lung's area-of-interest minimum bounding box preferred orientations bounding box (OBB) described here, OBB bounding box determine method
It is as follows: according to the location coordinate information all put in lung's area-of-interest, by principal component analysis (PCA) obtain characteristic value and
Feature vector calculates the central point of lung's area-of-interest whole coordinate points, calculates lung's area-of-interest all the points in feature
Projector distance under vector has determined that the side length of minimum bounding box.
The corresponding feature vector of characteristic value that the principal component analysis obtains is the axis direction vector of human body, as 440 in Fig. 4
It is shown.The smallest feature vector direction of characteristic value is X direction, as shown in 450 in Fig. 4.
Here according to lung's area-of-interest minimum bounding box determine that lung is interested in the side length of human body X direction
The transverse diameter length in region, as shown in 460 in Fig. 4.
In step S250, the second determination unit 150 determines the minimum bounding box of heart area-of-interest, determines heart sense
The transverse diameter length in interest region.
Here the axis direction of the minimum bounding box of heart area-of-interest is using the human body axis direction determined in step S240
With human body X direction, wherein as indicated at 440 in Fig. 4, X direction is as shown in 450 in Fig. 4 for human body axis direction.
Here the determination method of the minimum bounding box of heart area-of-interest is as follows, calculates the center of heart area-of-interest
Point calculates two side lengths that all projector distance of the point respectively in two axis directions of bounding box have confirmed that minimum bounding box.
Here the transverse diameter length of heart area-of-interest is side length of the minimum bounding box in transverse diameter direction, such as 540 institute in Fig. 5
Show.
In step S260, computing unit 160 calculates the area-of-interest of heart area-of-interest transverse diameter length and lung
The ratio of transverse diameter length is ambition ratio, calculates the area ratio of heart area-of-interest area and lung's area-of-interest as the heart
Chest area ratio.
The present invention can detect heart area, left lung region and right side lung region in normotopia of chest image automatically
Domain can calculate ambition ratio and ambition area ratio automatically, without manual intervention, significantly speed up the diagnosis process of doctor,
And precision is high, strong robustness, stability is strong.
The present invention also proposes a kind of program product of instruction code for being stored with machine-readable.Described instruction code is by machine
When device reads and executes, above-mentioned calculation method according to an embodiment of the present invention can be performed.
Correspondingly, it is also wrapped for carrying the storage medium of the program product of the above-mentioned instruction code for being stored with machine-readable
It includes in disclosure of the invention.The storage medium includes but is not limited to floppy disk, CD, magneto-optic disk, storage card, memory stick etc..
In addition, method of the invention be not limited to specifications described in time sequencing execute, can also according to it
His time sequencing, concurrently or independently execute.Therefore, the execution sequence of method described in this specification is not to this hair
Bright technical scope is construed as limiting.
Although being had been disclosed above by the description to specific embodiments of the present invention to the present invention, it answers
The understanding, above-mentioned all embodiments and example are exemplary, and not restrictive.Those skilled in the art can be in institute
Design is to various modifications of the invention, improvement or equivalent in attached spirit and scope of the claims.These modification, improve or
Person's equivalent should also be as being to be considered as included in protection scope of the present invention.
Claims (8)
1. device is compared in a kind of calculating ambition, comprising:
Input unit, for inputting chest image；
Acquiring unit obtains left lung region, right side lung areas according to scheduled location algorithm from the chest image
With the encirclement box position of heart area；
Extraction unit extracts left lung, right side lung and heart according to scheduled extraction algorithm from the chest image
Region of interest field mark；
First determination unit determines the transverse diameter length of lung；
Second determination unit determines the transverse diameter length of heart；
Computing unit, it is interested according to the heart according to the cardiac broad diameter length and lung's transverse diameter length computation ambition ratio
Region and lung's area-of-interest calculate ambition area ratio.
2. the apparatus according to claim 1, wherein scheduled left lung, right side lung and the heart of the acquiring unit
Dirty location algorithm is neural network model, and the parameter of the positioning neural network model is by multiple chest images and having marked
Left lung, right side lung and the heart of the chest image of note surround box position training and obtain, comprising:
The multiple chest image is input to initial positioning Classification Neural model, obtains the left side of each chest image
The bounding box location information of lung, right side lung and heart area；
According to the bounding box location information of the left lung of each chest image, right side lung and heart area and mark
Left lung, right side lung and the heart area bounding box location information of each chest image of note carry out reverse train, generate
The preset location algorithm neural network model.
3. the apparatus according to claim 1, wherein scheduled left lung, right side lung and the heart of the extraction unit
The extraction algorithm in dirty district domain is neural network model, and the parameter for extracting neural network model is by multiple chest images
And the area-of-interest position mark training of the left lung, right side lung and heart of marked chest image obtains, comprising:
The multiple chest image is input to initial positioning Classification Neural model, obtains the left side of each chest image
The region of interest location information domain of lung, right side lung and heart area；
According to the region of interest location information domain of the left lung of each chest image, right side lung and heart area and
Left lung, right side lung and the heart region of interest location information domain of marked each chest image carry out reverse train,
Generate the preset extraction algorithm neural network model.
4. the apparatus according to claim 1, wherein the transverse diameter length of lung passes through with lower section in first determination unit
Formula obtains:
Obtain lung's area-of-interest of left lung area-of-interest and right side lung area-of-interest composition；
Determine the minimum bounding box of lung's area-of-interest；
Human body plotted and human body X direction are determined according to the directional information of lung's minimum bounding box；
The transverse diameter length of lung is determined according to edge lengths of the lung's minimum bounding box in human body X direction.
5. according to the method described in claim 4, wherein, the minimum bounding box of lung's area-of-interest is direction encirclement
Box.
6. the apparatus according to claim 1, wherein the second determination unit cardiac transverse diameter length is in the following manner
It obtains:
Minimum bounding box is obtained according to the human body plotted, the human body X direction and the heart area-of-interest；
The transverse diameter length of heart area is obtained in the side length of human body X direction according to the minimum bounding box.
7. device according to claim 4, wherein the minimum bounding box of the heart area is oriented bounding box.
8. the apparatus according to claim 1, wherein the chest image is obtained by x-ray equipment, CT equipment or MRI machine
It takes.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910030726.2A CN109801276B (en) | 2019-01-14 | 2019-01-14 | Method and device for calculating heart-chest ratio |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910030726.2A CN109801276B (en) | 2019-01-14 | 2019-01-14 | Method and device for calculating heart-chest ratio |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109801276A true CN109801276A (en) | 2019-05-24 |
| CN109801276B CN109801276B (en) | 2023-05-19 |
Family
ID=66558723
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910030726.2A Active CN109801276B (en) | 2019-01-14 | 2019-01-14 | Method and device for calculating heart-chest ratio |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109801276B (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110517243A (en) * | 2019-08-23 | 2019-11-29 | 强联智创（北京）科技有限公司 | A kind of localization method and system based on DSA image |
| CN110610487A (en) * | 2019-08-29 | 2019-12-24 | 上海杏脉信息科技有限公司 | Method and device for measuring increase of heart shadow |
| CN111242956A (en) * | 2020-01-09 | 2020-06-05 | 西北工业大学 | U-Net-based ultrasonic fetal heart and fetal lung deep learning joint segmentation method |
| CN112651942A (en) * | 2020-12-28 | 2021-04-13 | 三星电子（中国）研发中心 | Layout detection method and device |
| WO2022050713A1 (en) * | 2020-09-07 | 2022-03-10 | 주식회사 뷰노 | Method for reading chest image |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20060285751A1 (en) * | 2005-06-15 | 2006-12-21 | Dawei Wu | Method, apparatus and storage medium for detecting cardio, thoracic and diaphragm borders |
| CN1943512A (en) * | 2006-11-08 | 2007-04-11 | 沈阳东软医疗系统有限公司 | Device and method for determining interesting zone |
| US20100189320A1 (en) * | 2007-06-19 | 2010-07-29 | Agfa Healthcare N.V. | Method of Segmenting Anatomic Entities in 3D Digital Medical Images |
| CN104732520A (en) * | 2015-01-31 | 2015-06-24 | 西安华海盈泰医疗信息技术有限公司 | Cardio-thoracic ratio measuring algorithm and system for chest digital image |
| CN105447870A (en) * | 2015-12-01 | 2016-03-30 | 沈阳东软医疗系统有限公司 | Extraction method and apparatus for body tissues in medical image |
| CN106991697A (en) * | 2017-04-17 | 2017-07-28 | 心医国际数字医疗系统(大连)有限公司 | The ambition ratio measuring method of chest digitized video |
| CN107403183A (en) * | 2017-07-21 | 2017-11-28 | 桂林电子科技大学 | The intelligent scissor method that conformity goal is detected and image segmentation is integrated |
| CN107665497A (en) * | 2016-07-29 | 2018-02-06 | 上海联影医疗科技有限公司 | In a kind of medical image calculate ambition than method |
| CN108717700A (en) * | 2018-04-09 | 2018-10-30 | 杭州依图医疗技术有限公司 | A kind of method and device of detection tubercle length electrical path length |
-
2019
- 2019-01-14 CN CN201910030726.2A patent/CN109801276B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20060285751A1 (en) * | 2005-06-15 | 2006-12-21 | Dawei Wu | Method, apparatus and storage medium for detecting cardio, thoracic and diaphragm borders |
| CN1943512A (en) * | 2006-11-08 | 2007-04-11 | 沈阳东软医疗系统有限公司 | Device and method for determining interesting zone |
| US20100189320A1 (en) * | 2007-06-19 | 2010-07-29 | Agfa Healthcare N.V. | Method of Segmenting Anatomic Entities in 3D Digital Medical Images |
| CN104732520A (en) * | 2015-01-31 | 2015-06-24 | 西安华海盈泰医疗信息技术有限公司 | Cardio-thoracic ratio measuring algorithm and system for chest digital image |
| CN105447870A (en) * | 2015-12-01 | 2016-03-30 | 沈阳东软医疗系统有限公司 | Extraction method and apparatus for body tissues in medical image |
| CN107665497A (en) * | 2016-07-29 | 2018-02-06 | 上海联影医疗科技有限公司 | In a kind of medical image calculate ambition than method |
| CN106991697A (en) * | 2017-04-17 | 2017-07-28 | 心医国际数字医疗系统(大连)有限公司 | The ambition ratio measuring method of chest digitized video |
| CN107403183A (en) * | 2017-07-21 | 2017-11-28 | 桂林电子科技大学 | The intelligent scissor method that conformity goal is detected and image segmentation is integrated |
| CN108717700A (en) * | 2018-04-09 | 2018-10-30 | 杭州依图医疗技术有限公司 | A kind of method and device of detection tubercle length electrical path length |
Non-Patent Citations (2)
| Title |
|---|
| DE LA CRUZ R R G,ET AL: "iXray: A Machine Learning-Based Digital Radiograph Pattern Recognition System for Lung Pathology Detection", 《MECHATRONICS AND MACHINE VISION》 * |
| 刘炎等: "一种X光胸片心胸比率自动计算方法", 《中国生物医学工程学报》 * |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110517243A (en) * | 2019-08-23 | 2019-11-29 | 强联智创（北京）科技有限公司 | A kind of localization method and system based on DSA image |
| CN110517243B (en) * | 2019-08-23 | 2022-03-25 | 强联智创（北京）科技有限公司 | Positioning method and system based on DSA image |
| CN110610487A (en) * | 2019-08-29 | 2019-12-24 | 上海杏脉信息科技有限公司 | Method and device for measuring increase of heart shadow |
| CN111242956A (en) * | 2020-01-09 | 2020-06-05 | 西北工业大学 | U-Net-based ultrasonic fetal heart and fetal lung deep learning joint segmentation method |
| WO2022050713A1 (en) * | 2020-09-07 | 2022-03-10 | 주식회사 뷰노 | Method for reading chest image |
| KR20220032405A (en) * | 2020-09-07 | 2022-03-15 | 주식회사 뷰노 | Method to read chest image |
| KR102507451B1 (en) * | 2020-09-07 | 2023-03-08 | 주식회사 뷰노 | Method to read chest image |
| CN112651942A (en) * | 2020-12-28 | 2021-04-13 | 三星电子（中国）研发中心 | Layout detection method and device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109801276B (en) | 2023-05-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109801276A (en) | A kind of method and device calculating ambition ratio | |
| US9697639B2 (en) | Three-dimensional model data generation device, method and program | |
| CN109754387B (en) | Intelligent detection and positioning method for whole-body bone imaging radioactive concentration focus | |
| US6785409B1 (en) | Segmentation method and apparatus for medical images using diffusion propagation, pixel classification, and mathematical morphology | |
| Zook et al. | Statistical analysis of fractal-based brain tumor detection algorithms | |
| US10318839B2 (en) | Method for automatic detection of anatomical landmarks in volumetric data | |
| ES2914387T3 (en) | immediate study | |
| US7583829B2 (en) | Method and apparatus for embolism analysis | |
| US7382907B2 (en) | Segmenting occluded anatomical structures in medical images | |
| CN107230206A (en) | A kind of 3D Lung neoplasm dividing methods of the super voxel sequence lung images based on multi-modal data | |
| CN112885453A (en) | Method and system for identifying pathological changes in subsequent medical images | |
| Bano et al. | AutoFB: Automating fetal biometry estimation from standard ultrasound planes | |
| US9275452B2 (en) | Method and system for automatically determining compliance of cross sectional imaging scans with a predetermined protocol | |
| JP2013090799A (en) | Image processing device, method and program | |
| Wang et al. | Automatic segmentation of spinal canals in CT images via iterative topology refinement | |
| Gao et al. | Accurate lung segmentation for X-ray CT images | |
| CN103208114A (en) | Stomach adipose tissue extraction method based on interactive segmentation | |
| CN111091605B (en) | Rib visualization method, identification method and computer-readable storage medium | |
| Schwing et al. | Reliable extraction of the mid-sagittal plane in 3D brain MRI via hierarchical landmark detection | |
| CN110148113A (en) | A kind of lesion target area information labeling method based on tomoscan diagram data | |
| CN115330732A (en) | Method and device for determining pancreatic cancer | |
| WO2023133929A1 (en) | Ultrasound-based human tissue symmetry detection and analysis method | |
| Feng et al. | Automatic fetal weight estimation using 3D ultrasonography | |
| Rachmawati et al. | Bone Scan Image Segmentation based on Active Shape Model for Cancer Metastasis Detection | |
| WO2019245506A2 (en) | A method and an algorithm to conduct a safe biopsy on lung airways |
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 |