CN109615624B - Blood flow velocity waveform automatic identification method based on ultrasonic image - Google Patents
Blood flow velocity waveform automatic identification method based on ultrasonic image Download PDFInfo
- Publication number
- CN109615624B CN109615624B CN201811480361.5A CN201811480361A CN109615624B CN 109615624 B CN109615624 B CN 109615624B CN 201811480361 A CN201811480361 A CN 201811480361A CN 109615624 B CN109615624 B CN 109615624B
- Authority
- CN
- China
- Prior art keywords
- image
- blood flow
- flow velocity
- waveform
- velocity waveform
- 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
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0012—Biomedical image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10132—Ultrasound image
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30004—Biomedical image processing
- G06T2207/30101—Blood vessel; Artery; Vein; Vascular
- G06T2207/30104—Vascular flow; Blood flow; Perfusion
Abstract
An automatic blood flow velocity waveform identification method based on ultrasonic images belongs to the field of image processing algorithms. The method comprises the steps of carrying out graying and binarization preprocessing on an image based on a clinical ultrasonic image containing a blood flow velocity waveform obtained by collection or shooting; transversely dividing the image by adopting a dichotomy method based on the binary image, and dividing a region with a high gray value below the blood flow velocity waveform; traversing upwards from the last row of the segmented region, and identifying the boundary of the waveform, namely the blood flow velocity waveform when the region has the maximum gray gradient and continues upwards without highlight gray in a certain length region; and finally, traversing and removing the interference of the numbers, frames and the like of the non-waveforms from the middle positions of the waveforms to the two ends, and outputting the digitized waveforms. The invention realizes the automatic identification and digital output of the blood flow velocity waveform in the image, and lays a foundation for the analysis work of the blood flow velocity waveform when the waveform source data of the ultrasonic equipment can not be acquired.
Description
Technical Field
The invention belongs to the field of image processing algorithms, and relates to an automatic blood flow velocity waveform identification method based on an ultrasonic image.
Background
Ultrasound is a conventional blood flow velocity detection method in clinic, and has wide application value in diagnosis of human diseases. The blood flow velocity waveform detected by the ultrasonic equipment contains rich information related to diseases, and the blood flow waveform has important reference significance no matter in clinical real-time observation or subsequent analysis of the diseases. The amplitude, shape of the blood flow waveform can be observed and analyzed manually in real time, but in addition to these obvious features, the waveform may contain information related to the disease that cannot be observed directly. In the subsequent analysis of the disease, this information can be exploited by means of digital signal processing. However, due to data protection of the ultrasound equipment manufacturer, the source data of the blood flow waveform is often not directly available. Researchers can only obtain an ultrasonic picture of a blood flow velocity waveform by shooting usually, then manually wipe off redundant information by software such as Photoshop and the like, keep an area of interest, and then extract a waveform by using waveform digitization software such as Getdata and the like by using a manual framing method for analysis. This results in a lot of tedious, complicated and inefficient repetitive work, which results in waste of time and human resources when the sample size is large, and is not suitable for clinical use. In order to meet the requirement of blood flow waveform analysis, the invention develops an automatic identification method of blood flow velocity waveform based on ultrasonic images, and the ultrasonic images containing the blood flow velocity waveform are subjected to waveform identification and digitized output, so that a foundation is laid for the subsequent analysis work of the blood flow velocity waveform, and the method has certain value for the early work of disease diagnosis.
Disclosure of Invention
The invention provides an automatic identification method of blood flow velocity waveform based on an ultrasonic image, which realizes automatic identification and digital output of the blood flow velocity waveform in the ultrasonic image. The blood flow velocity waveform automatic identification method based on the ultrasonic image comprises the following steps: based on the characteristics of high gray value of the blood flow waveform region, after graying and binarization, the image is subjected to dichotomy transverse segmentation. And when the overall gray value of the image area after the multiple segmentation is relatively high and the duty ratio of the white area reaches a certain threshold value, the area with the high gray value below the blood flow velocity waveform is considered to be segmented. And then, the whole body is traversed upwards, and when the maximum gray gradient is experienced and no highlight gray is existed in a certain length region which is continuously upwards, the boundary of the waveform, namely the blood flow velocity waveform, can be identified. The waveform can be output after the interference of a digital frame and the like is eliminated.
In order to achieve the purpose, the invention is realized by the following technical scheme:
an automatic identification method of blood flow velocity waveform based on ultrasonic image, comprising the following steps:
a1, graying the ultrasonic image to obtain a grayscale image;
a2, performing binarization processing on the gray-scale image to obtain a binarized image;
a3, performing multiple transverse segmentation on the binary image by adopting a dichotomy to obtain a region containing a high gray value below a blood flow velocity waveform;
step A4, based on the region divided by the algorithm step A3, in the whole binary image, traversing all the columns of the whole image from the last line of the region, namely the segLowRow line, stopping traversing when the region experiences the maximum gray gradient and continues upwards and has no highlight gray in a certain length region, and identifying the boundary of a waveform, namely a blood flow velocity waveform;
a5, eliminating the interference of numbers, symbols, frames and the like on the left and right sides of the blood flow velocity waveform;
step A6, outputting the digitized waveform.
As a further technical solution of the present invention, the method for transversely dividing a binarized image by binary method in step a3 includes the steps of:
step B1, dividing the image at 1/2 of image line number each time, comparing the total gray scale value of the upper half part and the lower half part of the image after each division and the duty ratio of the respective white area; the white area is a high gray value area contained below the blood flow velocity waveform;
and step B2, discarding the upper half image with relatively low total gray value, dividing the rest lower half image from the middle line again until the duty ratio of the white area in the divided image is more than or equal to the set threshold value, preferably 0.4, stopping dividing, and obtaining the area containing the high gray value below the blood flow velocity waveform.
As a further technical solution of the present invention, the characteristics in step a4, including the judgment basis for stopping traversal, are: since there may be a black gap below the blood flow velocity waveform and not all are white regions of high gray scale values, it cannot be assumed that a transition of gray scale values is a boundary of the waveform during the upward pass. One condition that needs to be added is that a maximum gray scale gradient (255) is encountered and continues up for a length of time without white areas. The length is set to 10% of the total number of lines of the original image, as a result of experience with multiple images.
As a further aspect of the present invention, the method for canceling interference such as numbers, symbols, and frames on both left and right sides of a blood flow velocity waveform in step a5 includes the steps of:
step C1, traversing from the middle row of the image, namely the middle position of the waveform, to two ends respectively after the waveform boundary is identified, and judging whether a number or a symbol is encountered;
step C2, judging whether a rectangular frame is encountered;
and step C3, ending traversal when encountering interference such as numbers, symbols or frames and the like, and outputting a waveform.
Compared with the prior art, the invention has the beneficial effects that:
the invention provides an automatic blood flow velocity waveform identification method based on an ultrasonic image, which simplifies the extraction process of the blood flow velocity waveform in a clinical ultrasonic image, and scientific researchers can realize automatic waveform identification and output by using the method, thereby laying a foundation for the analysis work of the blood flow velocity waveform and saving time.
Drawings
FIG. 1 is a flow chart of an embodiment of a waveform identification algorithm in an ultrasound image of the present invention; images 1-3 are sequentially arranged from left to right;
FIG. 2 is a graph of three different ultrasound images of a blood flow velocity waveform selected in accordance with the present invention;
FIG. 3 shows the variation trend of the duty ratios of the white areas of the three images along with the dividing times;
FIG. 4 is a flowchart of an algorithm for bisection segmentation of images in the present invention;
FIG. 5 shows the blood flow velocity waveform identification results of the present invention for three ultrasound images.
Detailed Description
The present invention will be described in detail below with reference to specific embodiments and the accompanying drawings. However, the present invention is not limited to the following examples.
Example 1
A1, selecting three different ultrasonic images containing blood flow velocity waveforms, and graying the ultrasonic images to obtain a grayscale image;
a2, performing binarization processing on the gray-scale image to obtain a binarized image;
because the gray value at the blood flow velocity waveform in most ultrasonic images is higher, the binarization threshold is set to be 100 based on the experimental experience of three ultrasonic images.
A3, performing multiple transverse segmentation on the binary image by adopting a dichotomy to obtain a region with a high gray value below a blood flow velocity waveform;
for most ultrasound images, when a white region with a high gray value is segmented below the blood flow velocity waveform, the increase rate of the duty ratio of the white region is reduced. Successive dichotomy segmentation was performed on three different ultrasound images with the duty ratios of the respective white regions as shown in fig. 3. As can be seen from the figure, the duty ratio of the three images after being divided for many times is more than 0.6. However, since the number of division times required to reach 0.6 is large, in order to save calculation time, the duty threshold may be set to 0.4 in actual division.
Step A4, based on the region divided by the algorithm, traversing all the columns of the whole binary image from the last line of the region, namely the segLowRow line, when the image experiences the maximum gray gradient and continues to the upper part without highlight gray in a region with a certain length, stopping traversing, and identifying the boundary of the waveform, namely the blood flow velocity waveform;
a5, eliminating the interference of numbers, symbols, frames and the like on the left and right sides of the blood flow velocity waveform;
step A6, outputting the digitized waveform.
As a further technical solution of the present invention, the method for transversely dividing a binarized image by binary method in step a3 includes the steps of:
b1, dividing the image at 1/2 of image line number, comparing the total gray scale value of the upper half part and the lower half part of the image after each division and the duty ratio of the respective white area;
and step B2, abandoning the half part image with lower total gray value, and dividing the rest part from the middle line again until the duty ratio of the white area in the divided image is more than or equal to the set threshold value 0.4, stopping dividing, and obtaining the area with high gray value below the blood flow velocity waveform.
Fig. 4 shows a flowchart of an algorithm for dividing an image by bisection, where rows is the total number of lines of a binarized image, segUpRow and segLowRow are two parameters used to record the highest line and the lowest line of the image divided each time, ratio _ up and ratio _ low are white area duty ratios of the upper and lower two partial images after division each time, and ratio is used to determine whether to end division, and its initial value is 0, if ratio is greater than 0.4, segUpRow and segLowRow at this time are recorded, and division is ended, otherwise, division is continued.
As a further technical solution of the present invention, the characteristics in step a4, including the judgment basis for stopping traversal, are: since there may be a black gap below the blood flow velocity waveform and not all are white regions of high gray scale values, it cannot be assumed that a transition of gray scale values is a boundary of the waveform during the upward pass. One condition that needs to be added is that a maximum gray scale gradient (255) is encountered and continues up for a length of time without white areas. The length is set to 10% of the total number of lines of the original image, as a result of experience with multiple images.
As a further aspect of the present invention, the method for canceling interference such as numbers, symbols, and frames on both left and right sides of a blood flow velocity waveform in step a5 includes the steps of:
step C1, traversing from the middle row of the image, namely the middle position of the waveform, to two ends respectively after the waveform boundary is identified, and judging whether a number or a symbol is encountered;
step C2, judging whether a rectangular frame is encountered;
and step C3, ending the traversal when the interference of numbers, symbols or frames is encountered.
Numbers, symbols and the like are generally positioned at the edge of an ultrasonic image and generally cannot be completely reserved after binarization, so that a break is formed at the start or the end of a waveform, and the removal of the numbers, symbols and the like can be judged by judging whether the waveform is broken or not; the border has a higher slope due to the existence of the vertical line, which is represented as a larger jump in the image and presents an almost unchanged waveform (the horizontal line of the border) after the jump, so that whether the border is met can be judged based on the size of the jump and whether the horizontal line is changed after the jump, and the set jump threshold is 30.
The blood flow velocity waveform automatic identification method based on the ultrasonic image comprises the following steps: based on the characteristics of high gray value of the blood flow waveform region, after graying and binarization, the image is subjected to dichotomy transverse segmentation. And when the overall gray value of the image area after the multiple segmentation is relatively high and the duty ratio of the white area reaches a certain threshold value, the area with the high gray value below the blood flow velocity waveform is considered to be segmented. And then, the whole body is traversed upwards, and when the maximum gray gradient is experienced and no highlight gray is existed in a certain length region which is continuously upwards, the boundary of the waveform, namely the blood flow velocity waveform, can be identified. The waveform can be output after the interference of a digital frame and the like is eliminated. The blood flow velocity waveform identification results for the three ultrasound images are shown in fig. 5.
Claims (3)
1. An automatic blood flow velocity waveform identification method based on ultrasonic images is characterized by comprising the following steps:
a1, graying the ultrasonic image to obtain a grayscale image;
a2, performing binarization processing on the gray-scale image to obtain a binarized image;
a3, performing multiple transverse segmentation on the binary image by adopting a dichotomy to obtain a region containing a high gray value below a blood flow velocity waveform;
step A4, based on the region divided by the algorithm step A3, in the whole binary image, traversing all the columns of the whole image from the last line of the region, namely the segLowRow line, stopping traversing when the region experiences the maximum gray gradient and continues upwards and has no highlight gray in a certain length region, and identifying the boundary of a waveform, namely a blood flow velocity waveform;
a5, eliminating the interference of numbers, symbols and frames on the left and right sides of the blood flow velocity waveform;
step A6, outputting a digitized waveform;
the method for transversely dividing the binarized image by the binary method in the step a3 includes the following steps:
step B1, dividing the image at 1/2 of image line number each time, comparing the total gray scale value of the upper half part and the lower half part of the image after each division and the duty ratio of the respective white area; the white area is a high gray value area contained below the blood flow velocity waveform;
and step B2, discarding the upper half image with relatively low total gray value, dividing the rest lower half image from the middle line again until the duty ratio of the white area in the divided image is more than or equal to the set threshold value, preferably 0.4, stopping dividing, and obtaining the area containing the high gray value below the blood flow velocity waveform.
2. The method for automatically recognizing a blood flow velocity waveform based on an ultrasound image as claimed in claim 1, wherein the step a4 includes the step of stopping the traversal according to the following steps: because a black gap may exist below the blood flow velocity waveform and all the gaps are not white areas with high gray values, the jump of the gray values cannot be determined as the boundary of the waveform during the upward traversal; one condition that needs to be added is that a maximum gray scale gradient (255) is encountered and the process continues up to a length where there are no white areas, the length being set to 10% of the total number of lines in the original image.
3. The method for automatically identifying a blood flow velocity waveform based on an ultrasound image as claimed in claim 1, wherein the method for eliminating the interference of numbers, symbols and borders existing on the left and right sides of the blood flow velocity waveform in step a5 comprises the following steps:
step C1, traversing from the middle row of the image, namely the middle position of the waveform, to two ends respectively after the waveform boundary is identified, and judging whether a number or a symbol is encountered;
step C2, judging whether a rectangular frame is encountered;
and step C3, ending the traversal when encountering the interference of the numbers, the symbols or the frames, and outputting the waveform.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811480361.5A CN109615624B (en) | 2018-12-05 | 2018-12-05 | Blood flow velocity waveform automatic identification method based on ultrasonic image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811480361.5A CN109615624B (en) | 2018-12-05 | 2018-12-05 | Blood flow velocity waveform automatic identification method based on ultrasonic image |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109615624A CN109615624A (en) | 2019-04-12 |
| CN109615624B true CN109615624B (en) | 2022-03-22 |
Family
ID=66006451
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811480361.5A Active CN109615624B (en) | 2018-12-05 | 2018-12-05 | Blood flow velocity waveform automatic identification method based on ultrasonic image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109615624B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113487553A (en) * | 2021-06-30 | 2021-10-08 | 武汉楚精灵医疗科技有限公司 | Intestinal tract endoscope withdrawal speed smoothing method |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1650190A (en) * | 2002-04-30 | 2005-08-03 | 皇家飞利浦电子股份有限公司 | Synthetically focused ultrasonic diagnostic imaging system for tissue and flow imaging |
| CN105708495A (en) * | 2016-01-26 | 2016-06-29 | 飞依诺科技(苏州)有限公司 | Ultrasonic color blood flow imaging boundary processing method and system |
| CN106028947A (en) * | 2015-05-07 | 2016-10-12 | 深圳迈瑞生物医疗电子股份有限公司 | Blood flow imaging method and system |
| CN107341801A (en) * | 2017-07-14 | 2017-11-10 | 合肥工业大学 | A kind of blood flow measuring method based on doppler flow inaging sonograph |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8315812B2 (en) * | 2010-08-12 | 2012-11-20 | Heartflow, Inc. | Method and system for patient-specific modeling of blood flow |
-
2018
- 2018-12-05 CN CN201811480361.5A patent/CN109615624B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1650190A (en) * | 2002-04-30 | 2005-08-03 | 皇家飞利浦电子股份有限公司 | Synthetically focused ultrasonic diagnostic imaging system for tissue and flow imaging |
| CN106028947A (en) * | 2015-05-07 | 2016-10-12 | 深圳迈瑞生物医疗电子股份有限公司 | Blood flow imaging method and system |
| CN105708495A (en) * | 2016-01-26 | 2016-06-29 | 飞依诺科技(苏州)有限公司 | Ultrasonic color blood flow imaging boundary processing method and system |
| CN107341801A (en) * | 2017-07-14 | 2017-11-10 | 合肥工业大学 | A kind of blood flow measuring method based on doppler flow inaging sonograph |
Non-Patent Citations (2)
| Title |
|---|
| "Hemodynamic analysis of sequential graft from right coronary system to left coronary system";Wenxin Wang et al.;《BioMedical Engineering OnLine》;20161231;第546-556页 * |
| "应用三尖瓣血流速度图评估肺动脉高压的研究";孙丹丹 等;《中国医学影像学杂志》;20171231;第13-16页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109615624A (en) | 2019-04-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US7912265B2 (en) | Computer-implemented system and method for automated and highly accurate plaque analysis, reporting, and visualization | |
| JP5355275B2 (en) | Cell image analyzer | |
| CN110120042B (en) | Crop image pest and disease damage area extraction method based on SLIC super-pixel and automatic threshold segmentation | |
| EP1585058A3 (en) | Medical image processing apparatus and method of processing medical image | |
| JP2007048000A (en) | Fingerprint image identification system, fingerprint image identification method, and program | |
| CN107174232B (en) | Electrocardiogram waveform extraction method | |
| CN110866932A (en) | Multi-channel tongue edge detection device and method and storage medium | |
| WO2021147866A1 (en) | Ecg signal acquisition method and device, storage medium, and electronic device | |
| Kang et al. | Real-time image processing system for endoscopic applications | |
| CN102194121A (en) | Application of improved maximum between-class variance method in tongue crack recognition | |
| CN111768408A (en) | Gastrointestinal marker automatic identification method and gastrointestinal marker automatic identification system | |
| CN109615624B (en) | Blood flow velocity waveform automatic identification method based on ultrasonic image | |
| CN107220644B (en) | Electrocardiogram scanning image gradient correction method | |
| JP5939023B2 (en) | Computer program and image extraction apparatus | |
| CN113223041A (en) | Method, system and storage medium for automatically extracting target area in image | |
| CN117078671A (en) | Thyroid ultrasonic image intelligent analysis system | |
| CN111340780A (en) | Focus detection method based on three-dimensional ultrasonic image | |
| CN111105427B (en) | Lung image segmentation method and system based on connected region analysis | |
| CN109635646B (en) | Head image processing method, system, equipment and storage medium | |
| CN107027067B (en) | Method and system for acquiring subtitle information in MV video resources | |
| WO2019181072A1 (en) | Image processing method, computer program, and recording medium | |
| CN112634240B (en) | Thyroid ultrasound image interference interface automatic identification and removal method based on threshold segmentation | |
| CN111242968B (en) | Method and system for detecting tooth area in tongue sample | |
| CN114663443A (en) | 12-lead paper electrocardiogram digitization method and device | |
| CN112465817A (en) | Pavement crack detection method based on directional filter |
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 |