CN118071750B - Femoral artery puncture ultrasonic image real-time processing method, device and working method - Google Patents
Femoral artery puncture ultrasonic image real-time processing method, device and working method Download PDFInfo
- Publication number
- CN118071750B CN118071750B CN202410480109.3A CN202410480109A CN118071750B CN 118071750 B CN118071750 B CN 118071750B CN 202410480109 A CN202410480109 A CN 202410480109A CN 118071750 B CN118071750 B CN 118071750B
- Authority
- CN
- China
- Prior art keywords
- puncture
- femoral artery
- detection
- blood vessel
- plaque
- 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
- 210000001105 femoral artery Anatomy 0.000 title claims abstract description 54
- 238000000034 method Methods 0.000 title claims abstract description 32
- 238000003672 processing method Methods 0.000 title claims abstract description 8
- 238000001514 detection method Methods 0.000 claims abstract description 68
- 210000004204 blood vessel Anatomy 0.000 claims abstract description 39
- 238000012549 training Methods 0.000 claims abstract description 35
- 230000011218 segmentation Effects 0.000 claims abstract description 17
- 238000012545 processing Methods 0.000 claims abstract description 16
- 230000004927 fusion Effects 0.000 claims abstract description 13
- 230000002792 vascular Effects 0.000 claims abstract description 10
- 238000001914 filtration Methods 0.000 claims abstract description 4
- 239000000523 sample Substances 0.000 claims description 40
- 238000002604 ultrasonography Methods 0.000 claims description 13
- 210000003462 vein Anatomy 0.000 claims description 13
- 238000004422 calculation algorithm Methods 0.000 claims description 12
- 238000013519 translation Methods 0.000 claims description 12
- 210000001367 artery Anatomy 0.000 claims description 10
- 238000012360 testing method Methods 0.000 claims description 8
- 238000010276 construction Methods 0.000 claims description 5
- 230000006870 function Effects 0.000 claims description 4
- 238000010606 normalization Methods 0.000 claims description 4
- 210000000689 upper leg Anatomy 0.000 claims description 3
- 238000011161 development Methods 0.000 description 3
- 230000018109 developmental process Effects 0.000 description 3
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 239000004575 stone Substances 0.000 description 2
- 208000008589 Obesity Diseases 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 238000004140 cleaning Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 235000020824 obesity Nutrition 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 210000002321 radial artery Anatomy 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 238000001356 surgical procedure Methods 0.000 description 1
- 208000024891 symptom Diseases 0.000 description 1
Abstract
The femoral artery puncture ultrasonic image real-time processing method, the femoral artery puncture ultrasonic image real-time processing device and the femoral artery puncture ultrasonic image working method can be used for carrying out vascular cutting and plaque detection in real time through ultrasonic images in a femoral artery puncture operation, and are high in accuracy and high in real-time performance. The method comprises the following steps: (1) Constructing a shared backbone network so as to extract image features, adopting an end-to-end single-target detector, selecting resnet a pre-training model, and only reserving a small-target detection output module; (2) feature fusion: fusing high-level semantic features and shallow detail features, filtering out plaque detection frames which do not coincide with the vessel wall, carrying out tracking processing on each detection frame, carrying out short-distance compensation on missed detection by a single-target tracking (KCF) method through time sequence information before and after fusion on missed detection between continuous frames, and controlling single-target tracking within 2 frames; (3) Outputting a plaque detection result, and decoding and outputting a plaque external rectangular frame; (4) Outputting the blood vessel segmentation result, and decoding and outputting a surrounding area of the blood vessel outline.
Description
Technical Field
The invention relates to the technical field of medical image processing, in particular to a femoral artery puncture ultrasonic image real-time processing method, a femoral artery puncture ultrasonic image real-time processing device and a working method, which are mainly used for performing vascular cutting and plaque detection in real time through ultrasonic images in femoral artery puncture operation.
Background
In femoral artery puncture surgery under ultrasound guidance, an ultrasound doctor and a surgeon are generally required to cooperate, and the ultrasound doctor gives guidance of a puncture point in real time through an ultrasound device, thereby assisting the surgeon in performing rapid and accurate puncture. In this procedure, the sonographer is required to find the location of the femoral artery and to scan the femoral artery all the way to the bifurcation to identify plaque and assist the surgeon in puncturing after the puncture point is determined. The experience of the scanning physician itself, the angle of the probe, the accuracy of plaque determination, and the condition of the patient (obesity, etc.) all affect the determination. The scanning time is long, the coverage is wide, and the ultrasonic doctor is required to hold the probe for a long time during puncture. If emergency treatment and limited patient exploration conditions are met, when personnel are not fully equipped, the situation of limited clinical puncture can occur.
In light of the above, it is desirable to design a robot in place of an sonographer to automatically or semi-automatically perform femoral artery puncture with the assistance of the surgeon. It is important to find the arterial region in the vascular region in the real-time dynamic ultrasonic image, and find the puncture needle position to avoid the plaque in the blood vessel, i.e. the method of performing the blood vessel cutting and plaque detection in real time with the assistance of ultrasound, and this is also the foundation stone of the puncture robot with the assistance of ultrasound.
Disclosure of Invention
In order to overcome the defects of the prior art, the technical problem to be solved by the invention is to provide the real-time processing method of the femoral artery puncture ultrasonic image, which can be used for carrying out vascular cutting and plaque detection in real time through the ultrasonic image in the femoral artery puncture operation, and has high accuracy and strong real-time performance.
The technical scheme of the invention is as follows: the femoral artery puncture ultrasonic image real-time processing method comprises the following steps:
(1) Constructing a shared backbone network so as to extract image features, adopting an end-to-end single-target detector, selecting resnet a pre-training model, and only reserving a small-target detection output module;
(2) Feature fusion: combining high-level semantic features and shallow detail features, filtering out plaque detection frames which are not coincident with the vessel wall to reduce false detection, tracking each detection frame, and carrying out short-distance compensation on missed detection by means of single-target tracking KCF (kcF) through time sequence information before and after combination for possible missed detection among continuous frames, wherein single-target tracking is controlled within 2 frames;
(3) Outputting a plaque detection result, and decoding and outputting a plaque external rectangular frame;
(4) Outputting the blood vessel segmentation result, and decoding and outputting a surrounding area of the blood vessel outline.
The invention constructs a shared backbone network to extract image features, fuses high-level semantic features and shallow detail features, filters out plaque detection frames which do not coincide with the vessel wall to reduce false detection, carries out tracking processing on each detection frame, carries out short-distance compensation on missed detection by means of single-target tracking KCF through time sequence information before and after fusion, decodes and outputs plaque circumscribed rectangular frames, decodes and outputs surrounding areas of the outer contours of the vessels, thereby realizing real-time vessel cutting and plaque detection through ultrasonic images in femoral artery puncture operation, and having high accuracy and labor cost saving compared with manual operation.
Also provided is a femoral artery puncture ultrasonic image real-time processing device, which comprises:
the main network construction module is configured to construct a shared main network so as to extract image characteristics, adopts an end-to-end single-target detector, selects resnet a pre-training model, and only retains a small-target detection output module;
The feature fusion module is configured to fuse high-level semantic features with shallow detail features, filter out plaque detection frames which do not coincide with the vessel wall to reduce false detection, track each detection frame, and carry out short-distance compensation on missed detection by means of single-target tracking KCF (kcF) through time sequence information before and after fusion for possible missed detection among continuous frames, wherein the single-target tracking is controlled within 2 frames;
The first output module is configured to output a plaque detection result, and decode and output a plaque external rectangular frame;
And a second output module configured to output a blood vessel segmentation result, and decode and output a surrounding area of the blood vessel outer contour.
The working method of the femoral artery puncture ultrasonic image real-time processing device is also provided, and the working method comprises the following steps:
(I) The ultrasonic probe is placed at the root of the thigh to find the transverse cutting of the femoral artery; performing a blood vessel segmentation algorithm to determine a femoral artery transection profile; according to the abscissa of the blood vessel contour center and the abscissa of the image center, calculating the translation direction of the probe, executing the probe translation action, slowly translating the probe, and arranging the blood vessel abscissa center in the image abscissa center;
(II) when the blood vessel abscissa center is at the image abscissa center, performing a probe rotating operation, and rotating the probe by 90 degrees to obtain a femoral artery longitudinal cutting image; performing a translation probe operation, translating the probe along the direction of the probe, and finding out a vascular bifurcation to be used as an endpoint of the femoral artery;
(III) operating the ultrasound device into a doppler mode to determine that the vicinity of the puncture is on a femoral artery vessel and that there is no vein above; performing translation probe operation, searching a position suitable for puncture from the femoral artery end point forwards, and determining the end point reached by the puncture needle head; executing a puncture needle end point calculation algorithm and a blood vessel contour segmentation algorithm, and determining a puncture end point and a puncture path; executing a plaque detection algorithm to ensure that no plaque exists on the puncture path, and if no plaque exists, finishing the search of the puncture position;
(IV) if the puncture point meets the condition, executing the adjustment of the angle of the puncture needle and performing puncture according to the calculated puncture angle and depth; and (3) if the puncture point is not in accordance with the condition, executing the operation of translating the probe, continuing to translate the probe forward, and repeating the step (III).
Drawings
Fig. 1 is a flow chart of a method for real-time processing of ultrasound images of femoral artery puncture in accordance with the present invention.
Detailed Description
As shown in fig. 1, the real-time processing method of the femoral artery puncture ultrasonic image comprises the following steps:
(1) Constructing a shared backbone network so as to extract image features, adopting an end-to-end single-target detector, selecting resnet a pre-training model, and only reserving a small-target detection output module;
(2) Feature fusion: combining high-level semantic features and shallow-level detail features, filtering out plaque detection frames which do not coincide with the vessel wall to reduce false detection, tracking each detection frame, and carrying out short-distance compensation on the missed detection by means of a single-target tracking KCF (kernelized correlation filters, nucleated correlation filter) method through timing sequence information before and after combination of continuous frames, wherein the single-target tracking is controlled within 2 frames;
(3) Outputting a plaque detection result, and decoding and outputting a plaque external rectangular frame;
(4) Outputting the blood vessel segmentation result, and decoding and outputting a surrounding area of the blood vessel outline.
The invention constructs a shared backbone network to extract image features, fuses high-level semantic features and shallow detail features, filters out plaque detection frames which do not coincide with the vessel wall to reduce false detection, carries out tracking processing on each detection frame, carries out short-distance compensation on missed detection by means of single-target tracking KCF through time sequence information before and after fusion, decodes and outputs plaque circumscribed rectangular frames, decodes and outputs surrounding areas of the outer contours of the vessels, thereby realizing real-time vessel cutting and plaque detection through ultrasonic images in femoral artery puncture operation, and having high accuracy and labor cost saving compared with manual operation.
Preferably, the training data are femoral artery blood vessel ultrasonic images and segmentation and detection frames corresponding to the images, and the training data are divided into a training set and a testing set according to a ratio of 4:1; while background data not including any objects is included as a negative sample.
Preferably, the step (1) comprises the following sub-steps:
Initializing, loading resnet a pre-training model of the ultrasonic data, wherein the pre-training model is trained on a large amount of unlabeled ultrasonic data in a self-supervision mode by using a contrast learning method;
(1.2) initializing an optimizer SGD required by model training, configuring an initial learning rate and a learning rate reduction strategy of a network, and instantiating a corresponding loss function;
(1.3) loading training data, initializing a data enhancement strategy to increase the robustness of the model, and improving the convergence speed and accuracy of the model through a regularization strategy;
and (1.4) sending the data into a network, calculating loss, calculating gradient, updating model parameters, converging the models, and storing the best result model on a test set.
Preferably, in the step (1.3), the data enhancement policy includes: image horizontal flip, pan, rotate, contrast, brightness, saturation, scale, normalization.
Preferably, in the step (4), the arteriovenous identification distinguishes whether the vascular region is an artery or a vein; firstly, transferring an image to an HSV color space, and respectively identifying red and blue areas through threshold values; the number of occurrences of red and blue is counted for each pixel in a time window of 2 seconds, and if the ratio exceeds 80%, a vein is identified, and if the ratio is between 10% and 80%, an artery is identified.
It will be understood by those skilled in the art that all or part of the steps in implementing the above embodiment method may be implemented by a program to instruct related hardware, where the program may be stored in a computer readable storage medium, where the program when executed includes the steps of the above embodiment method, and the storage medium may be: ROM/RAM, magnetic disks, optical disks, memory cards, etc. Accordingly, the invention also includes a real-time femoral artery puncture ultrasound image processing device corresponding to the method of the invention, which is generally represented in the form of functional modules corresponding to the steps of the method. The device comprises:
the main network construction module is configured to construct a shared main network so as to extract image characteristics, adopts an end-to-end single-target detector, selects resnet a pre-training model, and only retains a small-target detection output module;
The feature fusion module is configured to fuse high-level semantic features with shallow detail features, filter out plaque detection frames which do not coincide with the vessel wall to reduce false detection, track each detection frame, and carry out short-distance compensation on missed detection by means of single-target tracking KCF (kcF) through time sequence information before and after fusion for possible missed detection among continuous frames, wherein the single-target tracking is controlled within 2 frames;
The first output module is configured to output a plaque detection result, and decode and output a plaque external rectangular frame;
And a second output module configured to output a blood vessel segmentation result, and decode and output a surrounding area of the blood vessel outer contour.
Preferably, the training data are femoral artery blood vessel ultrasonic images and segmentation and detection frames corresponding to the images, and the training data are divided into a training set and a testing set according to a ratio of 4:1; while background data not including any objects is included as a negative sample.
Preferably, the backbone network construction module performs the steps of:
Initializing, loading resnet a pre-training model of the ultrasonic data, wherein the pre-training model is trained on a large amount of unlabeled ultrasonic data in a self-supervision mode by using a contrast learning method;
(1.2) initializing an optimizer SGD required by model training, configuring an initial learning rate and a learning rate reduction strategy of a network, and instantiating a corresponding loss function;
(1.3) loading training data, initializing a data enhancement strategy to increase the robustness of the model, and improving the convergence speed and accuracy of the model through a regularization strategy;
(1.4) sending the data into a network, calculating loss, calculating gradient, updating model parameters, converging the models, and storing the best result model on a test set;
In the step (1.3), the data enhancement policy includes: image horizontal flip, pan, rotate, contrast, brightness, saturation, scale, normalization.
Preferably, in the second output module, the arteriovenous identification distinguishes whether the vascular region is an artery or a vein; firstly, transferring an image to an HSV color space, and respectively identifying red and blue areas through threshold values; the number of occurrences of red and blue is counted for each pixel in a time window of2 seconds, and if the ratio exceeds 80%, a vein is identified, and if the ratio is between 10% and 80%, an artery is identified.
The working method of the femoral artery puncture ultrasonic image real-time processing device is also provided, and the working method comprises the following steps:
(I) The ultrasonic probe is placed at the root of the thigh to find the transverse cutting of the femoral artery; performing a blood vessel segmentation algorithm to determine a femoral artery transection profile; according to the abscissa of the blood vessel contour center and the abscissa of the image center, calculating the translation direction of the probe, executing the probe translation action, slowly translating the probe, and arranging the blood vessel abscissa center in the image abscissa center;
(II) when the blood vessel abscissa center is at the image abscissa center, performing a probe rotating operation, and rotating the probe by 90 degrees to obtain a femoral artery longitudinal cutting image; performing a translation probe operation, translating the probe along the direction of the probe, and finding out a vascular bifurcation to be used as an endpoint of the femoral artery;
(III) operating the ultrasound device into a doppler mode to determine that the vicinity of the puncture is on a femoral artery vessel and that there is no vein above; performing translation probe operation, searching a position suitable for puncture from the femoral artery end point forwards, and determining the end point reached by the puncture needle head; executing a puncture needle end point calculation algorithm and a blood vessel contour segmentation algorithm, and determining a puncture end point and a puncture path; executing a plaque detection algorithm to ensure that no plaque exists on the puncture path, and if no plaque exists, finishing the search of the puncture position;
(IV) if the puncture point meets the condition, executing the adjustment of the angle of the puncture needle and performing puncture according to the calculated puncture angle and depth; and (3) if the puncture point is not in accordance with the condition, executing the operation of translating the probe, continuing to translate the probe forward, and repeating the step (III).
The invention has the following technical effects:
1. the method adopted in the ultrasonic scanning of the invention is that on the basis of manual operation, artificial intelligence artery and vein recognition, plaque recognition and finally puncture point selection are carried out, thus saving labor cost, particularly puncture time can be effectively saved in rescuing patients with severe symptoms, and human resources are liberated.
2. The invention is an innovation under the assistance of the current artificial real-time intelligence, the selection of blood vessels and the identification of plaques are carried out in static state, the real-time selection and the identification are carried out at home and abroad at present, in the operation of ultrasonic artificial leading, the invention replaces manual work by the operation of a machine, not only improves the accuracy, but also has the accuracy of the identification of the arteries and the veins up to 95.4%, the accuracy of the identification of the plaques up to 91.82%, and the accuracy is higher along with the updating of further data.
3. At present, 2788 pictures of blood vessels are marked, in plaque marking 3953, the accuracy of blood vessels and plaques is over 90%, the specificity is over 91%, the accuracy and the specificity are obviously improved in the subsequent analysis after data cleaning, the identification of femoral artery in manual operation and the selection of plaques are successfully completed, and finally accurate puncture parts are provided for clinical operation.
4. The development of the artificial intelligent ultrasonic assisted positioning blood vessel plaque recognition method can be applied to femoral arteries, the subsequent application significance to other blood vessels is great, the development of the method is a very important foundation stone for the development of vascular robots, and the designed standardized flow also plays a very important role in further developing blood vessels such as radial artery puncture and the like.
5. The manual intelligent ultrasonic positioning puncture method not only replaces manual work and assists the manual work to solve the time and labor cost resources, but also is a very effective alternative scheme for solving the problem of high-risk operation of clinician irradiation under X rays at present.
The present invention is not limited to the preferred embodiments, but can be modified in any way according to the technical principles of the present invention, and all such modifications, equivalent variations and modifications are included in the scope of the present invention.
Claims (6)
1. The real-time femoral artery puncture ultrasonic image processing method is characterized by comprising the following steps of: which comprises the following steps:
(1) Constructing a shared backbone network so as to extract image features, adopting an end-to-end single-target detector, selecting resnet a pre-training model, and only reserving a small-target detection output module;
(2) Feature fusion: combining high-level semantic features and shallow detail features, filtering out plaque detection frames which are not coincident with the vessel wall to reduce false detection, tracking each detection frame, and carrying out short-distance compensation on missed detection by a single-target tracking (KCF) method through time sequence information before and after combination for possible missed detection between continuous frames, wherein single-target tracking is controlled within 2 frames;
(3) Outputting a plaque detection result, and decoding and outputting a plaque external rectangular frame;
(4) Outputting a blood vessel segmentation result, and decoding an enclosed region of the outer contour of the output blood vessel;
The training data are femoral artery blood vessel ultrasonic images and segmentation and detection frames corresponding to the images, and the training data are divided into a training set and a testing set according to a ratio of 4:1; while including background data that does not include any objects as negative samples;
the step (1) comprises the following sub-steps:
Initializing, loading resnet a pre-training model of the ultrasonic data, wherein the pre-training model is trained on a large amount of unlabeled ultrasonic data in a self-supervision mode by using a contrast learning method;
(1.2) initializing an optimizer SGD required by model training, configuring an initial learning rate and a learning rate reduction strategy of a network, and instantiating a corresponding loss function;
(1.3) loading training data, initializing a data enhancement strategy to increase the robustness of the model, and improving the convergence speed and accuracy of the model through a regularization strategy;
and (1.4) sending the data into a network, calculating loss, calculating gradient, updating model parameters, converging the models, and storing the best result model on a test set.
2. The method for processing femoral artery puncture ultrasonic images in real time according to claim 1, wherein the method comprises the following steps: in the step (1.3), the data enhancement policy includes: image horizontal flip, pan, rotate, contrast, brightness, saturation, scale, normalization.
3. The method for processing femoral artery puncture ultrasonic images in real time according to claim 2, wherein the method comprises the following steps: in the step (4), the artery and vein identification distinguishes whether the blood vessel area is an artery or a vein; the image is first turned into the HSV color space, and red and blue regions are identified by thresholds, respectively.
4. Femoral artery puncture ultrasonic image real-time processing apparatus, its characterized in that: it comprises the following steps:
the main network construction module is configured to construct a shared main network so as to extract image characteristics, adopts an end-to-end single-target detector, selects resnet a pre-training model, and only retains a small-target detection output module;
The feature fusion module is configured to fuse high-level semantic features with shallow detail features, filter out plaque detection frames which do not coincide with the vessel wall to reduce false detection, track each detection frame, and carry out short-distance compensation on missed detection by means of single-target tracking KCF (kcF) through time sequence information before and after fusion for possible missed detection among continuous frames, wherein the single-target tracking is controlled within 2 frames;
The first output module is configured to output a plaque detection result, and decode and output a plaque external rectangular frame;
a second output module configured to output a blood vessel segmentation result, decoding an enclosed region of an output blood vessel outer contour;
The training data are femoral artery blood vessel ultrasonic images and segmentation and detection frames corresponding to the images, and the training data are divided into a training set and a testing set according to a ratio of 4:1; while including background data that does not include any objects as negative samples;
The backbone network construction module performs the steps of:
Initializing, loading resnet a pre-training model of the ultrasonic data, wherein the pre-training model is trained on a large amount of unlabeled ultrasonic data in a self-supervision mode by using a contrast learning method;
(1.2) initializing an optimizer SGD required by model training, configuring an initial learning rate and a learning rate reduction strategy of a network, and instantiating a corresponding loss function;
(1.3) loading training data, initializing a data enhancement strategy to increase the robustness of the model, and improving the convergence speed and accuracy of the model through a regularization strategy;
(1.4) sending the data into a network, calculating loss, calculating gradient, updating model parameters, converging the models, and storing the best result model on a test set;
In the step (1.3), the data enhancement policy includes: image horizontal flip, pan, rotate, contrast, brightness, saturation, scale, normalization.
5. The femoral artery puncture ultrasound image real-time processing device of claim 4, wherein: in the second output module, artery and vein identification distinguishes whether a blood vessel area is an artery or a vein; the image is first turned into the HSV color space, and red and blue regions are identified by thresholds, respectively.
6. The femoral artery puncture ultrasound image real-time processing device of claim 5, wherein: the working method comprises the following steps:
(I) The ultrasonic probe is placed at the root of the thigh to find the transverse cutting of the femoral artery; performing a blood vessel segmentation algorithm to determine a femoral artery transection profile; according to the abscissa of the blood vessel contour center and the abscissa of the image center, calculating the translation direction of the probe, executing the probe translation action, slowly translating the probe, and arranging the blood vessel abscissa center in the image abscissa center;
(II) when the blood vessel abscissa center is at the image abscissa center, performing a probe rotating operation, and rotating the probe by 90 degrees to obtain a femoral artery longitudinal cutting image; performing a translation probe operation, translating the probe along the direction of the probe, and finding out a vascular bifurcation to be used as an endpoint of the femoral artery;
(III) operating the ultrasound device into a doppler mode to determine that the vicinity of the puncture is on a femoral artery vessel and that there is no vein above; performing translation probe operation, searching a position suitable for puncture from the femoral artery end point forwards, and determining the end point reached by the puncture needle head; executing a puncture needle end point calculation algorithm and a blood vessel contour segmentation algorithm, and determining a puncture end point and a puncture path; executing a plaque detection algorithm to ensure that no plaque exists on the puncture path, and if no plaque exists, finishing the search of the puncture position;
(IV) if the puncture point meets the condition, executing the adjustment of the angle of the puncture needle and performing puncture according to the calculated puncture angle and depth; and (3) if the puncture point is not in accordance with the condition, executing the operation of translating the probe, continuing to translate the probe forward, and repeating the step (III).
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202410480109.3A CN118071750B (en) | 2024-04-22 | Femoral artery puncture ultrasonic image real-time processing method, device and working method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202410480109.3A CN118071750B (en) | 2024-04-22 | Femoral artery puncture ultrasonic image real-time processing method, device and working method |
Publications (2)
Publication Number | Publication Date |
---|---|
CN118071750A CN118071750A (en) | 2024-05-24 |
CN118071750B true CN118071750B (en) | 2024-07-09 |
Family
ID=
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108171698A (en) * | 2018-02-12 | 2018-06-15 | 数坤(北京)网络科技有限公司 | A kind of method of automatic detection human heart Coronary Calcification patch |
CN111950388A (en) * | 2020-07-22 | 2020-11-17 | 上海市同仁医院 | Vulnerable plaque tracking and identifying system and method |
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108171698A (en) * | 2018-02-12 | 2018-06-15 | 数坤(北京)网络科技有限公司 | A kind of method of automatic detection human heart Coronary Calcification patch |
CN111950388A (en) * | 2020-07-22 | 2020-11-17 | 上海市同仁医院 | Vulnerable plaque tracking and identifying system and method |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
KR102212499B1 (en) | Ivus image analysis method | |
CN112652052B (en) | Coronary artery three-dimensional reconstruction method and system based on blood vessel branch registration | |
CN109685809B (en) | Liver infusorian focus segmentation method and system based on neural network | |
CN109199322B (en) | Yellow spot detection method and storage device | |
CN110648338B (en) | Image segmentation method, readable storage medium, and image processing apparatus | |
CN111932554A (en) | Pulmonary blood vessel segmentation method, device and storage medium | |
Gil et al. | Automatic segmentation of artery wall in coronary IVUS images: a probabilistic approach | |
CN109087310B (en) | Meibomian gland texture region segmentation method and system, storage medium and intelligent terminal | |
CN112991315A (en) | Identification method and system of vascular lesion, storage medium and electronic device | |
CN111657883A (en) | Coronary artery SYNTAX score automatic calculation method and system based on sequence radiography | |
KR102250689B1 (en) | Method and device for automatic vessel extraction of fundus photography using registration of fluorescein angiography | |
CN113935976A (en) | Method and system for automatically segmenting blood vessels in internal organs by enhancing CT (computed tomography) image | |
CN109003280A (en) | Inner membrance dividing method in a kind of blood vessel of binary channels intravascular ultrasound image | |
CN118071750B (en) | Femoral artery puncture ultrasonic image real-time processing method, device and working method | |
Jiang et al. | Dopus-net: Quality-aware robotic ultrasound imaging based on doppler signal | |
WO2024037358A1 (en) | Method for automatically identifying and positioning perforator vessel, device, and storage medium | |
CN118071750A (en) | Femoral artery puncture ultrasonic image real-time processing method, device and working method | |
CN116363311A (en) | Coronary Leiden score calculation and risk classification method and system | |
CN116245867A (en) | Vascular plaque and thrombus identification method and system based on unsupervised learning | |
WO2022096867A1 (en) | Image processing of intravascular ultrasound images | |
CN113951932A (en) | Scanning method and device for ultrasonic equipment | |
CN114022405A (en) | Intravascular ultrasound image processing method based on deep learning | |
CN112614141A (en) | Method and device for planning blood vessel scanning path, storage medium and terminal equipment | |
Mi et al. | Detecting carotid intima-media from small-sample ultrasound images | |
Mahesh et al. | Intelligent Systems for Medical Diagnostics with the Detection of Diabetic Retinopathy at Reduced Entropy |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant |