CN112950578A - Blood vessel identification and positioning method and device based on two-dimensional image enhancement - Google Patents
Blood vessel identification and positioning method and device based on two-dimensional image enhancement Download PDFInfo
- Publication number
- CN112950578A CN112950578A CN202110221254.6A CN202110221254A CN112950578A CN 112950578 A CN112950578 A CN 112950578A CN 202110221254 A CN202110221254 A CN 202110221254A CN 112950578 A CN112950578 A CN 112950578A
- Authority
- CN
- China
- Prior art keywords
- blood vessel
- image
- vessel image
- stage
- needle insertion
- 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.)
- Pending
Links
- 210000004204 blood vessel Anatomy 0.000 title claims abstract description 244
- 238000000034 method Methods 0.000 title claims abstract description 19
- 238000003780 insertion Methods 0.000 claims abstract description 87
- 230000037431 insertion Effects 0.000 claims abstract description 87
- 230000011218 segmentation Effects 0.000 claims abstract description 37
- 238000010241 blood sampling Methods 0.000 claims abstract description 34
- 238000012549 training Methods 0.000 claims abstract description 33
- 238000012545 processing Methods 0.000 claims abstract description 31
- 238000001914 filtration Methods 0.000 claims description 14
- 238000007781 pre-processing Methods 0.000 claims description 14
- 238000006243 chemical reaction Methods 0.000 claims description 10
- 238000005260 corrosion Methods 0.000 claims description 8
- 230000007797 corrosion Effects 0.000 claims description 8
- 238000003708 edge detection Methods 0.000 claims description 8
- 239000008280 blood Substances 0.000 claims description 7
- 210000004369 blood Anatomy 0.000 claims description 7
- 238000001514 detection method Methods 0.000 claims description 6
- 230000004807 localization Effects 0.000 claims 4
- 230000009286 beneficial effect Effects 0.000 description 6
- 239000002184 metal Substances 0.000 description 4
- 238000010586 diagram Methods 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 230000004907 flux Effects 0.000 description 2
- 230000001737 promoting effect Effects 0.000 description 1
- 210000004003 subcutaneous fat Anatomy 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
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
- G06T5/00—Image enhancement or restoration
- G06T5/40—Image enhancement or restoration using histogram techniques
-
- 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
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/13—Edge detection
-
- 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
- G06T2207/20—Special algorithmic details
- G06T2207/20024—Filtering details
- G06T2207/20032—Median filtering
-
- 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
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- 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
- G06T2207/20—Special algorithmic details
- G06T2207/20092—Interactive image processing based on input by user
- G06T2207/20104—Interactive definition of region of interest [ROI]
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Medical Informatics (AREA)
- Quality & Reliability (AREA)
- Radiology & Medical Imaging (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Image Processing (AREA)
- Image Analysis (AREA)
Abstract
The invention provides a blood vessel identification and positioning method and a device based on two-dimensional image enhancement, wherein the method comprises the steps of projecting an original blood vessel image to a blood sampling target area to form a stage blood vessel image; acquiring a blood vessel image at a first stage and training based on a projection region part in the blood vessel image to acquire a projection region ROI semantic segmentation model; inputting an original blood vessel image into a projection region ROI semantic segmentation model to extract an ROI region, and forming a two-stage blood vessel image based on the extracted ROI region; performing enhancement processing based on the two-stage blood vessel image to obtain an optimized blood vessel image, wherein the optimized blood vessel image contains the target blood vessel of the needle insertion; and acquiring a preferred needle inserting point of a needle inserting target blood vessel and a two-dimensional position parameter of the preferred needle inserting point in a one-stage blood vessel image. The blood vessel intelligent identification device can intelligently identify blood vessels and accurately position needle insertion points of blood sampling operation, so that the blood sampling operation is more effective and accurate, the psychological burden of patients is reduced, the pain of the whole process is relieved, and the blood sampling experience of the patients is improved.
Description
Technical Field
The invention belongs to the technical field of blood vessel identification and positioning, and particularly relates to a blood vessel identification and positioning method and device based on two-dimensional image enhancement.
Background
At present, the current clinical blood sampling situation basically depends on the experience of medical staff, and blood vessels are identified and needle insertion points are determined through observation of blood sampled staff.
However, medical staff cannot guarantee that every blood sampling work is accurate and effective, or due to the fact that the experience of the medical staff is insufficient, the subcutaneous fat of a person to be sampled is too thick, or due to the fact that the person to be sampled is old in age, the blood vessel is weak and the like, the success rate of manual blood sampling and the patient experience are prone to suffering from problems.
Disclosure of Invention
Therefore, the technical problem to be solved by the present invention is to provide a blood vessel identification and positioning method and device based on two-dimensional image enhancement, which can intelligently identify blood vessels and accurately position needle insertion points for blood sampling operation, so that the blood sampling operation is more effective and accurate, the psychological burden of patients can be reduced, the pain of the whole process can be reduced, and the blood sampling experience of patients can be further improved.
In order to solve the above problems, the present invention provides a method for identifying and positioning blood vessels based on two-dimensional image enhancement, comprising:
acquiring an image of a blood sampling target area to form an original blood vessel image, and projecting the original blood vessel image to the blood sampling target area to form a first-stage blood vessel image;
acquiring the blood vessel image at the first stage and training based on a projection region part in the blood vessel image to acquire a projection region ROI semantic segmentation model;
inputting the original blood vessel image into the projection region ROI semantic segmentation model to extract an ROI region, and forming a two-stage blood vessel image based on the extracted ROI region;
performing enhancement processing on the two-stage blood vessel image to obtain an optimized blood vessel image, wherein the optimized blood vessel image contains a needle insertion target blood vessel;
and acquiring a preferred needle inserting point of the needle inserting target blood vessel and a two-dimensional position parameter of the preferred needle inserting point in the one-stage blood vessel image.
Preferably, acquiring the blood vessel image of the stage and training and acquiring a projection region ROI semantic segmentation model based on a projection region part therein is implemented as follows:
and marking a projection region part in the blood vessel image at the first stage as a target region ROI and making the projection region ROI as a training set, and performing model training by using a deplaybV 3+ semantic segmentation model of MobilenetV2 to obtain the projection region ROI semantic segmentation model.
Preferably, the obtaining of the optimized blood vessel image by performing the enhancement processing based on the two-stage blood vessel image specifically includes:
and sequentially carrying out gray level conversion, preprocessing operation and binarization operation on the two-stage blood vessel image to obtain the optimized blood vessel image.
Preferably, the first and second electrodes are formed of a metal,
the preprocessing operation comprises histogram equalization processing and primary median filtering which are performed in sequence and limit the contrast; the binarization operation comprises edge detection operation, corrosion expansion, secondary median filtering and minimum external ellipse fitting operation which are carried out in sequence.
Preferably, the first and second electrodes are formed of a metal,
acquiring a preferred needle insertion point of the needle insertion target blood vessel and two-dimensional position parameters of the preferred needle insertion point in the stage blood vessel image specifically include:
performing thinning operation on the needle insertion target blood vessel, extracting a skeleton part of the needle insertion target blood vessel, acquiring a single-pixel contour of the needle insertion target blood vessel, further acquiring a central line of the single-pixel contour, and determining the trend of the needle insertion target blood vessel; and performing linear detection on the blood vessel of the needle insertion target to obtain two end points of a fitting straight line of the blood vessel of the needle insertion target, and taking the first end point in the needle insertion movement direction of the blood taking needle in the two end points of the fitting straight line as the preferable needle insertion point.
The invention also provides a blood vessel identification and positioning device based on two-dimensional image enhancement, which comprises:
the blood sampling device comprises an acquisition unit, a processing unit and a processing unit, wherein the acquisition unit is used for acquiring an image of a blood sampling target area to form an original blood vessel image and projecting the original blood vessel image to the blood sampling target area to form a first-stage blood vessel image;
the model training unit is used for acquiring the blood vessel image at the first stage and training a projection region part based on the blood vessel image to acquire a projection region ROI semantic segmentation model;
the ROI region acquisition unit is used for inputting the original blood vessel image into the projection region ROI semantic segmentation model to extract an ROI region and forming a two-stage blood vessel image based on the extracted ROI region;
the positioning unit is used for performing enhancement processing on the basis of the two-stage blood vessel image to obtain an optimized blood vessel image, and the optimized blood vessel image contains a needle insertion target blood vessel;
and acquiring a preferred needle inserting point of the needle inserting target blood vessel and a two-dimensional position parameter of the preferred needle inserting point in the one-stage blood vessel image.
Preferably, acquiring the blood vessel image of the stage and training and acquiring a projection region ROI semantic segmentation model based on a projection region part therein is implemented as follows:
and marking a projection region part in the blood vessel image at the first stage as a target region ROI and making the projection region ROI as a training set, and performing model training by using a deplaybV 3+ semantic segmentation model of MobilenetV2 to obtain the projection region ROI semantic segmentation model.
Preferably, the obtaining of the optimized blood vessel image by performing the enhancement processing based on the two-stage blood vessel image specifically includes:
and sequentially carrying out gray level conversion, preprocessing operation and binarization operation on the two-stage blood vessel image to obtain the optimized blood vessel image.
Preferably, the first and second electrodes are formed of a metal,
the preprocessing operation comprises histogram equalization processing and primary median filtering which are performed in sequence and limit the contrast; the binarization operation comprises edge detection operation, corrosion expansion, secondary median filtering and minimum external ellipse fitting operation which are carried out in sequence.
Preferably, the first and second electrodes are formed of a metal,
acquiring a preferred needle insertion point of the needle insertion target blood vessel and two-dimensional position parameters of the preferred needle insertion point in the stage blood vessel image specifically include:
performing thinning operation on the needle insertion target blood vessel, extracting a skeleton part of the needle insertion target blood vessel, acquiring a single-pixel contour of the needle insertion target blood vessel, further acquiring a central line of the single-pixel contour, and determining the trend of the needle insertion target blood vessel; and performing linear detection on the blood vessel of the needle insertion target to obtain two end points of a fitting straight line of the blood vessel of the needle insertion target, and taking the first end point in the needle insertion movement direction of the blood taking needle in the two end points of the fitting straight line as the preferable needle insertion point.
The invention provides a blood vessel identification and positioning method and a device based on two-dimensional image enhancement, which are used for acquiring a one-stage blood vessel image formed by projection, extracting an ROI (region of interest) region by adopting a projection region ROI semantic segmentation model obtained by training so as to acquire a two-stage blood vessel image, and the two-stage blood vessel image is enhanced to obtain an optimized blood vessel image, and finally two-dimensional position parameters of the needle inserting target blood vessel and the corresponding preferred needle inserting point are obtained, thereby realizing the accurate identification and positioning of the target blood vessel of the needle insertion and the preferred needle insertion point, changing the working mode of identifying the blood vessel and searching the needle insertion point completely depending on the personal experience of medical personnel in the prior art, being beneficial to the realization of intelligent automatic blood sampling by adopting the technical proposal of the invention, and then can promote blood sampling operation success rate and alleviate the painful, the aspect of promoting blood sampling experience and feeling etc. brings beneficial effect.
Drawings
FIG. 1 is a schematic diagram illustrating a two-dimensional image enhancement based blood vessel identification and location method according to an embodiment of the present invention;
FIG. 2 is a schematic diagram illustrating steps of a method for identifying and locating a blood vessel based on an image according to another embodiment of the present invention;
fig. 3 is a schematic structural diagram of a blood vessel identification and positioning device based on two-dimensional image enhancement according to an embodiment of the present invention.
Detailed Description
With reference to fig. 1 to fig. 3, according to an embodiment of the present invention, there is provided a blood vessel identification and location method based on two-dimensional image enhancement, including:
acquiring an image of a blood sampling target area to form an original blood vessel image, and projecting the original blood vessel image to the blood sampling target area to form a first-stage blood vessel image;
acquiring the blood vessel image at the first stage, and training based on a projection region part in the blood vessel image to acquire a projection region ROI (region of interest) semantic segmentation model;
inputting the original blood vessel image into the projection region ROI semantic segmentation model to extract an ROI region, and forming a two-stage blood vessel image based on the extracted ROI region;
performing enhancement processing on the two-stage blood vessel image to obtain an optimized blood vessel image, wherein the optimized blood vessel image contains a needle insertion target blood vessel;
and acquiring a preferred needle inserting point of the needle inserting target blood vessel and a two-dimensional position parameter of the preferred needle inserting point in the one-stage blood vessel image.
According to the technical scheme, a one-stage blood vessel image formed by projection is obtained, a projection region ROI semantic segmentation model obtained by training is adopted to extract an ROI region so as to obtain a two-stage blood vessel image, the two-stage blood vessel image is enhanced to obtain an optimized blood vessel image, and two-dimensional position parameters of a needle inserting target blood vessel and a corresponding preferred needle inserting point are finally obtained, so that accurate identification and positioning of the needle inserting target blood vessel and the preferred needle inserting point are realized, the working mode that blood vessels are identified and the needle inserting point is searched by completely depending on personal experience of medical workers in the prior art is changed, the technical scheme provided by the invention can be beneficial to realization of intelligent automatic blood sampling, and further beneficial effects can be brought in the aspects of improving the success rate of blood sampling operation, relieving pain of patients, improving the experience of blood sampling and the like.
In some embodiments, acquiring the one-stage blood vessel image and training to obtain a projection region ROI semantic segmentation model based on a projection region part therein is implemented as follows: and marking a projection region part in the blood vessel image at the first stage as a target region ROI and making the projection region ROI as a training set, and performing model training by using a deplaybV 3+ semantic segmentation model of MobilenetV2 to obtain the projection region ROI semantic segmentation model. The projection region ROI semantic segmentation model obtained through training can further improve the identification precision of blood vessels in the image.
In some embodiments, the performing enhancement processing based on the two-stage blood vessel image to obtain an optimized blood vessel image specifically includes: and sequentially carrying out gray level conversion, preprocessing operation and binarization operation on the two-stage blood vessel image to obtain the optimized blood vessel image.
Specifically, the gray conversion is to perform gray processing on the two-stage blood vessel image, and convert the color three-channel image into a gray single-channel image, so that the processing speed of image data can be increased.
The preprocessing operation comprises histogram equalization processing and primary median filtering which are performed in sequence for limiting contrast, and the step mainly comprises the steps of preprocessing the gray level image subjected to gray level conversion, eliminating noise caused by randomness of photon flux and performing denoising processing on the image.
The binarization operation comprises edge detection operation, corrosion expansion, secondary median filtering and minimum external ellipse fitting operation which are carried out in sequence. Wherein the edge detection is capable of identifying and extracting a vessel portion in the projection region; the corrosion expansion and the secondary median filtering can eliminate noise points in the extracted blood vessel contour image in the blood vessel part; the operation object of the minimum circumscribed ellipse fitting operation is the blood vessel contour, the average length, the average width, the area, the relative position and the factors of the blood vessel contour can be quantitatively analyzed according to the fitting result through the operation, the data of the four dimensions are normalized, the optimal index is calculated according to the weight of each factor, and the blood vessel which is most suitable for needle insertion is found from the blood vessel contour according to the optimal index and serves as the target blood vessel for needle insertion.
In some embodiments, the acquiring a preferred needle insertion point of the needle insertion target blood vessel and the two-dimensional position parameter of the preferred needle insertion point in the one-stage blood vessel image specifically include:
performing thinning operation on the needle insertion target blood vessel, extracting a skeleton part of the needle insertion target blood vessel, acquiring a single-pixel contour of the needle insertion target blood vessel, further acquiring a central line of the single-pixel contour, and determining the trend of the needle insertion target blood vessel, so that a preferred needle insertion point and a needle insertion track can be ensured to be always in a blood vessel region; and performing linear detection on the blood vessel of the needle insertion target to obtain two end points of a fitting straight line of the blood vessel of the needle insertion target, and taking the first end point in the needle insertion movement direction of the blood taking needle in the two end points of the fitting straight line as the preferable needle insertion point.
According to an embodiment of the present invention, there is also provided a blood vessel identification and location device based on two-dimensional image enhancement, including:
the blood sampling device comprises an acquisition unit, a processing unit and a processing unit, wherein the acquisition unit is used for acquiring an image of a blood sampling target area to form an original blood vessel image and projecting the original blood vessel image to the blood sampling target area to form a first-stage blood vessel image;
the model training unit is used for acquiring the blood vessel image at the first stage and training a projection region part based on the blood vessel image to acquire a projection region ROI (region of interest) semantic segmentation model;
the ROI region acquisition unit is used for inputting the original blood vessel image into the projection region ROI semantic segmentation model to extract an ROI region and forming a two-stage blood vessel image based on the extracted ROI region;
the positioning unit is used for performing enhancement processing on the basis of the two-stage blood vessel image to obtain an optimized blood vessel image, and the optimized blood vessel image contains a needle insertion target blood vessel;
and acquiring a preferred needle inserting point of the needle inserting target blood vessel and a two-dimensional position parameter of the preferred needle inserting point in the one-stage blood vessel image.
According to the technical scheme, a one-stage blood vessel image formed by projection is obtained, a projection region ROI semantic segmentation model obtained by training is adopted to extract an ROI region so as to obtain a two-stage blood vessel image, the two-stage blood vessel image is enhanced to obtain an optimized blood vessel image, and two-dimensional position parameters of a needle inserting target blood vessel and a corresponding preferred needle inserting point are finally obtained, so that accurate identification and positioning of the needle inserting target blood vessel and the preferred needle inserting point are realized, the working mode that blood vessels are identified and the needle inserting point is searched by completely depending on personal experience of medical workers in the prior art is changed, the technical scheme provided by the invention can be beneficial to realization of intelligent automatic blood sampling, and further beneficial effects can be brought in the aspects of improving the success rate of blood sampling operation, relieving pain of patients, improving the experience of blood sampling and the like.
In some embodiments, acquiring the one-stage blood vessel image and training to obtain a projection region ROI semantic segmentation model based on a projection region part therein is implemented as follows: and marking a projection region part in the blood vessel image at the first stage as a target region ROI and making the projection region ROI as a training set, and performing model training by using a deplaybV 3+ semantic segmentation model of MobilenetV2 to obtain the projection region ROI semantic segmentation model. The projection region ROI semantic segmentation model obtained through training can further improve the identification precision of blood vessels in the image.
In some embodiments, the performing enhancement processing based on the two-stage blood vessel image to obtain an optimized blood vessel image specifically includes: and sequentially carrying out gray level conversion, preprocessing operation and binarization operation on the two-stage blood vessel image to obtain the optimized blood vessel image.
Specifically, the gray conversion is to perform gray processing on the two-stage blood vessel image, and convert the color three-channel image into a gray single-channel image, so that the processing speed of image data can be increased.
The preprocessing operation comprises histogram equalization processing and primary median filtering which are performed in sequence for limiting contrast, and the step mainly comprises the steps of preprocessing the gray level image subjected to gray level conversion, eliminating noise caused by randomness of photon flux and performing denoising processing on the image.
The binarization operation comprises edge detection operation, corrosion expansion, secondary median filtering and minimum external ellipse fitting operation which are carried out in sequence. Wherein the edge detection is capable of identifying and extracting a vessel portion in the projection region; the corrosion expansion and the secondary median filtering can eliminate noise points in the extracted blood vessel contour image in the blood vessel part; the operation object of the minimum circumscribed ellipse fitting operation is the blood vessel contour, the average length, the average width, the area, the relative position and the factors of the blood vessel contour can be quantitatively analyzed according to the fitting result through the operation, the data of the four dimensions are normalized, the optimal index is calculated according to the weight of each factor, and the blood vessel which is most suitable for needle insertion is found from the blood vessel contour according to the optimal index and serves as the target blood vessel for needle insertion.
In some embodiments, the acquiring a preferred needle insertion point of the needle insertion target blood vessel and the two-dimensional position parameter of the preferred needle insertion point in the one-stage blood vessel image specifically include:
performing thinning operation on the needle insertion target blood vessel, extracting a skeleton part of the needle insertion target blood vessel, acquiring a single-pixel contour of the needle insertion target blood vessel, further acquiring a central line of the single-pixel contour, and determining the trend of the needle insertion target blood vessel, so that a preferred needle insertion point and a needle insertion track can be ensured to be always in a blood vessel region; and performing linear detection on the blood vessel of the needle insertion target to obtain two end points of a fitting straight line of the blood vessel of the needle insertion target, and taking the first end point in the needle insertion movement direction of the blood taking needle in the two end points of the fitting straight line as the preferable needle insertion point.
It is readily understood by a person skilled in the art that the advantageous ways described above can be freely combined, superimposed without conflict.
The present invention is not limited to the above preferred embodiments, and any modifications, equivalent substitutions and improvements made within the spirit and principle of the present invention should be included in the protection scope of the present invention. The above is only a preferred embodiment of the present invention, and it should be noted that, for those skilled in the art, several improvements and modifications can be made without departing from the technical principle of the present invention, and these improvements and modifications should also be regarded as the protection scope of the present invention.
Claims (10)
1. A blood vessel identification and positioning method based on two-dimensional image enhancement is characterized by comprising the following steps:
acquiring an image of a blood sampling target area to form an original blood vessel image, and projecting the original blood vessel image to the blood sampling target area to form a first-stage blood vessel image;
acquiring the blood vessel image at the first stage and training based on a projection region part in the blood vessel image to acquire a projection region ROI semantic segmentation model;
inputting the original blood vessel image into the projection region ROI semantic segmentation model to extract an ROI region, and forming a two-stage blood vessel image based on the extracted ROI region;
performing enhancement processing on the two-stage blood vessel image to obtain an optimized blood vessel image, wherein the optimized blood vessel image contains a needle insertion target blood vessel;
and acquiring a preferred needle inserting point of the needle inserting target blood vessel and a two-dimensional position parameter of the preferred needle inserting point in the one-stage blood vessel image.
2. The method for identifying and positioning blood vessels based on two-dimensional image enhancement as claimed in claim 1, wherein the obtaining of the blood vessel image of the first stage and the training of the projected region part based on the obtained projected region ROI semantic segmentation model are realized by the following method:
and marking a projection region part in the blood vessel image at the first stage as a target region ROI and making the projection region ROI as a training set, and performing model training by using a deplaybV 3+ semantic segmentation model of MobilenetV2 to obtain the projection region ROI semantic segmentation model.
3. The method for identifying and positioning blood vessels based on two-dimensional image enhancement according to claim 1, wherein the enhancement processing based on the two-stage blood vessel images to obtain the optimized blood vessel images specifically comprises:
and sequentially carrying out gray level conversion, preprocessing operation and binarization operation on the two-stage blood vessel image to obtain the optimized blood vessel image.
4. The two-dimensional image enhancement based blood vessel identification and localization method according to claim 3,
the preprocessing operation comprises histogram equalization processing and primary median filtering which are performed in sequence and limit the contrast; the binarization operation comprises edge detection operation, corrosion expansion, secondary median filtering and minimum external ellipse fitting operation which are carried out in sequence.
5. The two-dimensional image enhancement based blood vessel identification and localization method according to claim 1,
acquiring a preferred needle insertion point of the needle insertion target blood vessel and two-dimensional position parameters of the preferred needle insertion point in the stage blood vessel image specifically include:
performing thinning operation on the needle insertion target blood vessel, extracting a skeleton part of the needle insertion target blood vessel, acquiring a single-pixel contour of the needle insertion target blood vessel, further acquiring a central line of the single-pixel contour, and determining the trend of the needle insertion target blood vessel; and performing linear detection on the blood vessel of the needle insertion target to obtain two end points of a fitting straight line of the blood vessel of the needle insertion target, and taking the first end point in the needle insertion movement direction of the blood taking needle in the two end points of the fitting straight line as the preferable needle insertion point.
6. A blood vessel identification and positioning device based on two-dimensional image enhancement is characterized by comprising:
the blood sampling device comprises an acquisition unit, a processing unit and a processing unit, wherein the acquisition unit is used for acquiring an image of a blood sampling target area to form an original blood vessel image and projecting the original blood vessel image to the blood sampling target area to form a first-stage blood vessel image;
the model training unit is used for acquiring the blood vessel image at the first stage and training a projection region part based on the blood vessel image to acquire a projection region ROI semantic segmentation model;
the ROI region acquisition unit is used for inputting the original blood vessel image into the projection region ROI semantic segmentation model to extract an ROI region and forming a two-stage blood vessel image based on the extracted ROI region;
the positioning unit is used for performing enhancement processing on the basis of the two-stage blood vessel image to obtain an optimized blood vessel image, and the optimized blood vessel image contains a needle insertion target blood vessel;
and acquiring a preferred needle inserting point of the needle inserting target blood vessel and a two-dimensional position parameter of the preferred needle inserting point in the one-stage blood vessel image.
7. The apparatus for identifying and positioning blood vessels based on two-dimensional image enhancement as claimed in claim 6, wherein the obtaining of the blood vessel image of the stage and the training of obtaining the ROI semantic segmentation model of the projection region based on the projection region part thereof are realized by the following steps:
and marking a projection region part in the blood vessel image at the first stage as a target region ROI and making the projection region ROI as a training set, and performing model training by using a deplaybV 3+ semantic segmentation model of MobilenetV2 to obtain the projection region ROI semantic segmentation model.
8. The apparatus for identifying and locating a blood vessel based on two-dimensional image enhancement according to claim 6, wherein the enhancement processing based on the two-stage blood vessel image to obtain the optimized blood vessel image specifically comprises:
and sequentially carrying out gray level conversion, preprocessing operation and binarization operation on the two-stage blood vessel image to obtain the optimized blood vessel image.
9. The two-dimensional image enhancement based blood vessel identification and localization device according to claim 8,
the preprocessing operation comprises histogram equalization processing and primary median filtering which are performed in sequence and limit the contrast; the binarization operation comprises edge detection operation, corrosion expansion, secondary median filtering and minimum external ellipse fitting operation which are carried out in sequence.
10. The two-dimensional image enhancement based blood vessel identification and localization device according to claim 6,
acquiring a preferred needle insertion point of the needle insertion target blood vessel and two-dimensional position parameters of the preferred needle insertion point in the stage blood vessel image specifically include:
performing thinning operation on the needle insertion target blood vessel, extracting a skeleton part of the needle insertion target blood vessel, acquiring a single-pixel contour of the needle insertion target blood vessel, further acquiring a central line of the single-pixel contour, and determining the trend of the needle insertion target blood vessel; and performing linear detection on the blood vessel of the needle insertion target to obtain two end points of a fitting straight line of the blood vessel of the needle insertion target, and taking the first end point in the needle insertion movement direction of the blood taking needle in the two end points of the fitting straight line as the preferable needle insertion point.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110221254.6A CN112950578A (en) | 2021-02-26 | 2021-02-26 | Blood vessel identification and positioning method and device based on two-dimensional image enhancement |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110221254.6A CN112950578A (en) | 2021-02-26 | 2021-02-26 | Blood vessel identification and positioning method and device based on two-dimensional image enhancement |
Publications (1)
Publication Number | Publication Date |
---|---|
CN112950578A true CN112950578A (en) | 2021-06-11 |
Family
ID=76246692
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110221254.6A Pending CN112950578A (en) | 2021-02-26 | 2021-02-26 | Blood vessel identification and positioning method and device based on two-dimensional image enhancement |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN112950578A (en) |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20110301500A1 (en) * | 2008-10-29 | 2011-12-08 | Tim Maguire | Automated vessel puncture device using three-dimensional(3d) near infrared (nir) imaging and a robotically driven needle |
CN107397534A (en) * | 2017-08-28 | 2017-11-28 | 深圳麦科田软件技术有限公司 | A kind of integrated vein blood vessel identifying system and method |
CN111887864A (en) * | 2020-06-05 | 2020-11-06 | 哈工大机器人(中山)无人装备与人工智能研究院 | End effector for blood collection robotic device |
CN112043922A (en) * | 2020-09-25 | 2020-12-08 | 伏羲九针智能科技(北京)有限公司 | Automatic puncture device for venous blood vessels |
CN112102227A (en) * | 2020-06-05 | 2020-12-18 | 哈尔滨工业大学 | Blood vessel optimal puncture point selection method based on blood vessel size characteristics |
CN112288731A (en) * | 2020-11-05 | 2021-01-29 | 杭州深睿博联科技有限公司 | Blood vessel stenosis analysis method and device |
-
2021
- 2021-02-26 CN CN202110221254.6A patent/CN112950578A/en active Pending
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20110301500A1 (en) * | 2008-10-29 | 2011-12-08 | Tim Maguire | Automated vessel puncture device using three-dimensional(3d) near infrared (nir) imaging and a robotically driven needle |
CN107397534A (en) * | 2017-08-28 | 2017-11-28 | 深圳麦科田软件技术有限公司 | A kind of integrated vein blood vessel identifying system and method |
CN111887864A (en) * | 2020-06-05 | 2020-11-06 | 哈工大机器人(中山)无人装备与人工智能研究院 | End effector for blood collection robotic device |
CN112102227A (en) * | 2020-06-05 | 2020-12-18 | 哈尔滨工业大学 | Blood vessel optimal puncture point selection method based on blood vessel size characteristics |
CN112043922A (en) * | 2020-09-25 | 2020-12-08 | 伏羲九针智能科技(北京)有限公司 | Automatic puncture device for venous blood vessels |
CN112288731A (en) * | 2020-11-05 | 2021-01-29 | 杭州深睿博联科技有限公司 | Blood vessel stenosis analysis method and device |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN100586371C (en) | Image processing process based on magnetic resonance three-dimensional renogram | |
CN110956635A (en) | Lung segment segmentation method, device, equipment and storage medium | |
CN111667467B (en) | Clustering algorithm-based lower limb vascular calcification index multi-parameter accumulation calculation method | |
CN113017625B (en) | Control method and device of blood sampling robot | |
CN110930374A (en) | Acupoint positioning method based on double-depth camera | |
JP2014508021A (en) | Medical device for examining the neck | |
CN113420826B (en) | Liver focus image processing system and image processing method | |
CN108229584A (en) | A kind of Multimodal medical image recognition methods and device based on deep learning | |
CN114757960B (en) | Tooth segmentation and reconstruction method based on CBCT image and storage medium | |
CN108491770A (en) | A kind of data processing method based on fracture image | |
CN110338759B (en) | Facial pain expression data acquisition method | |
CN112927187A (en) | Method for automatically identifying and positioning focal cortical dysplasia epileptic focus | |
CN115862819B (en) | Medical image management method based on image processing | |
CN110648304A (en) | Intelligent auxiliary diagnosis method for handheld hard endoscope | |
CN113017566B (en) | Image-based blood vessel identification and positioning method and device | |
CN109785325A (en) | A method of the Multimodal medical image based on deep learning | |
CN112183518B (en) | Automatic vein target determination method, device and equipment | |
CN112950578A (en) | Blood vessel identification and positioning method and device based on two-dimensional image enhancement | |
CN115439445A (en) | Hepatic blood vessel and liver tumor recognition system | |
CN114176602B (en) | Method for simultaneously positioning electrocardiograph P wave, QRS wave and T wave based on deep learning multi-target detection | |
Cerrolaza et al. | Fully-automatic glottis segmentation with active shape models. | |
Zhang et al. | A Comparison of Applying Image Processing and Deep Learning in Acne Region Extraction | |
CN113592821A (en) | Femoral medullary cavity type detection method and system | |
CN109993754B (en) | Method and system for skull segmentation from images | |
CN113570545A (en) | Visual identification pain grading assessment method |
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 |