CN114947916A - Method and device for calculating SYNTAX score of coronary artery lesion - Google Patents
Method and device for calculating SYNTAX score of coronary artery lesion Download PDFInfo
- Publication number
- CN114947916A CN114947916A CN202210570258.XA CN202210570258A CN114947916A CN 114947916 A CN114947916 A CN 114947916A CN 202210570258 A CN202210570258 A CN 202210570258A CN 114947916 A CN114947916 A CN 114947916A
- Authority
- CN
- China
- Prior art keywords
- target
- coronary artery
- lesion
- segment
- vessel
- 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
- 210000004351 coronary vessel Anatomy 0.000 title claims abstract description 272
- 230000003902 lesion Effects 0.000 title claims abstract description 229
- 238000000034 method Methods 0.000 title claims abstract description 39
- 230000002966 stenotic effect Effects 0.000 claims abstract description 83
- 210000004204 blood vessel Anatomy 0.000 claims abstract description 39
- 238000012216 screening Methods 0.000 claims abstract description 20
- 230000011218 segmentation Effects 0.000 claims abstract description 16
- 238000013528 artificial neural network Methods 0.000 claims description 17
- 238000004364 calculation method Methods 0.000 claims description 17
- 238000012549 training Methods 0.000 claims description 14
- 230000017531 blood circulation Effects 0.000 claims description 13
- 230000008901 benefit Effects 0.000 claims description 12
- 208000031481 Pathologic Constriction Diseases 0.000 claims description 7
- 208000037804 stenosis Diseases 0.000 claims description 7
- 230000036262 stenosis Effects 0.000 claims description 7
- 238000004590 computer program Methods 0.000 claims description 6
- 238000012545 processing Methods 0.000 claims description 6
- 238000013507 mapping Methods 0.000 claims description 5
- 238000002601 radiography Methods 0.000 claims description 5
- 238000009825 accumulation Methods 0.000 claims description 3
- 238000005192 partition Methods 0.000 claims description 3
- 238000000638 solvent extraction Methods 0.000 claims 1
- 238000010586 diagram Methods 0.000 description 12
- 238000002586 coronary angiography Methods 0.000 description 6
- 230000006870 function Effects 0.000 description 4
- 230000008569 process Effects 0.000 description 4
- 238000004891 communication Methods 0.000 description 3
- 230000008878 coupling Effects 0.000 description 3
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 230000000877 morphologic effect Effects 0.000 description 3
- 238000011282 treatment Methods 0.000 description 3
- 230000002792 vascular Effects 0.000 description 3
- 230000002411 adverse Effects 0.000 description 2
- 239000002872 contrast media Substances 0.000 description 2
- DDRJAANPRJIHGJ-UHFFFAOYSA-N creatinine Chemical compound CN1CC(=O)NC1=N DDRJAANPRJIHGJ-UHFFFAOYSA-N 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- 238000011156 evaluation Methods 0.000 description 2
- 230000009191 jumping Effects 0.000 description 2
- 238000013077 scoring method Methods 0.000 description 2
- 208000006545 Chronic Obstructive Pulmonary Disease Diseases 0.000 description 1
- 201000006306 Cor pulmonale Diseases 0.000 description 1
- 208000018262 Peripheral vascular disease Diseases 0.000 description 1
- 230000036770 blood supply Effects 0.000 description 1
- 238000004422 calculation algorithm Methods 0.000 description 1
- 238000007675 cardiac surgery Methods 0.000 description 1
- 208000029078 coronary artery disease Diseases 0.000 description 1
- 229940109239 creatinine Drugs 0.000 description 1
- 238000012217 deletion Methods 0.000 description 1
- 230000037430 deletion Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 239000003814 drug Substances 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000002372 labelling Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000013146 percutaneous coronary intervention Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 208000024891 symptom Diseases 0.000 description 1
- 238000011269 treatment regimen Methods 0.000 description 1
- 230000002861 ventricular Effects 0.000 description 1
- 238000012800 visualization Methods 0.000 description 1
Images
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B6/00—Apparatus or devices for radiation diagnosis; Apparatus or devices for radiation diagnosis combined with radiation therapy equipment
- A61B6/02—Arrangements for diagnosis sequentially in different planes; Stereoscopic radiation diagnosis
- A61B6/03—Computed tomography [CT]
- A61B6/032—Transmission computed tomography [CT]
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B6/00—Apparatus or devices for radiation diagnosis; Apparatus or devices for radiation diagnosis combined with radiation therapy equipment
- A61B6/48—Diagnostic techniques
- A61B6/481—Diagnostic techniques involving the use of contrast agents
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B6/00—Apparatus or devices for radiation diagnosis; Apparatus or devices for radiation diagnosis combined with radiation therapy equipment
- A61B6/50—Apparatus or devices for radiation diagnosis; Apparatus or devices for radiation diagnosis combined with radiation therapy equipment specially adapted for specific body parts; specially adapted for specific clinical applications
- A61B6/504—Apparatus or devices for radiation diagnosis; Apparatus or devices for radiation diagnosis combined with radiation therapy equipment specially adapted for specific body parts; specially adapted for specific clinical applications for diagnosis of blood vessels, e.g. by angiography
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B6/00—Apparatus or devices for radiation diagnosis; Apparatus or devices for radiation diagnosis combined with radiation therapy equipment
- A61B6/52—Devices using data or image processing specially adapted for radiation diagnosis
- A61B6/5211—Devices using data or image processing specially adapted for radiation diagnosis involving processing of medical diagnostic data
Landscapes
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Engineering & Computer Science (AREA)
- Medical Informatics (AREA)
- Biomedical Technology (AREA)
- Surgery (AREA)
- High Energy & Nuclear Physics (AREA)
- Physics & Mathematics (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Optics & Photonics (AREA)
- Pathology (AREA)
- Radiology & Medical Imaging (AREA)
- Veterinary Medicine (AREA)
- Heart & Thoracic Surgery (AREA)
- Molecular Biology (AREA)
- Biophysics (AREA)
- Animal Behavior & Ethology (AREA)
- General Health & Medical Sciences (AREA)
- Public Health (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Vascular Medicine (AREA)
- Dentistry (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Pulmonology (AREA)
- Theoretical Computer Science (AREA)
- Apparatus For Radiation Diagnosis (AREA)
Abstract
The application provides a method and a device for calculating a SYNTAX score of coronary artery lesion, which are used for acquiring blood vessel characteristic data of a target coronary artery of a target patient; inputting the target coronary artery central line, all target points on the target coronary artery central line and the node type of each target point in the blood vessel characteristic data into a blood vessel segment division model to automatically divide the target coronary artery blood vessel and determine a plurality of target blood vessel segments in the target coronary artery; screening out at least one stenotic vessel segment from the plurality of target vessel segments according to a lesion region in the target coronary artery and a fractional flow reserve of the lesion region; according to the segment scoring rule of the SYNTAX score, determining the lesion score of each narrow blood vessel segment and accumulating to determine the SYNTAX score of the target patient. By the method, full-automatic segmentation and full-automatic scoring of the coronary artery are achieved, and therefore the calculating efficiency of lesion scoring is improved.
Description
Technical Field
The application relates to the technical field of medicine, in particular to a method and a device for calculating a SYNTAX score of coronary artery lesion.
Background
The SYNTAX (The Synergy of Percutaneous Coronary Intervention treatment and Cardiac Surgery) scoring system is a scoring system for evaluating The complexity of Coronary lesions according to The anatomical characteristics of The Coronary, and a score is finally obtained by adopting a 16-segment Coronary tree method according to The result of Coronary angiography and combining The distribution advantage type of The Coronary, The lesion part, The stenosis degree, The lesion number and The specific characteristics of The lesion to comprehensively analyze The Coronary lesions, thereby providing an objective evaluation index which can be accurately quantified for clinical treatment and having stronger clinical practicability.
The conventional SYNTAX score generally requires a physician to correspond each coronary vessel in a two-dimensional coronary angiography image to a coronary segment on a standard coronary anatomical pattern map, and needs to make an accurate judgment on the position, structure and degree of each coronary lesion and whether an adverse sign exists. However, coronary angiography is an invasive examination, and requires a special contrast medium to be injected into a coronary artery opening after a puncture of a coronary artery of a body, and then detection is performed. Moreover, identification of coronary segments and lesion features in the segments by means of manual marking generally has the problems of erroneous identification, low efficiency and the like.
Disclosure of Invention
In view of this, an object of the present application is to provide a method and an apparatus for calculating a syncax score of coronary artery lesion, so as to implement full-automatic segmentation and full-automatic scoring of coronary artery, thereby improving the efficiency of calculating the syncax score and reducing the pressure of doctors.
The embodiment of the application provides a method for calculating a coronary artery lesion SYNTAX score, which comprises the following steps:
acquiring blood vessel characteristic data of a target coronary artery determined according to a coronary artery three-dimensional medical image of a target patient; at least a target coronary artery centerline, all target points on the target coronary artery centerline, a lesion region in the target coronary artery, and a fractional flow reserve of the lesion region are included in the vessel feature data;
determining the node type of each target point on the central line of the target coronary artery according to a preset node type division rule; wherein the target point is a point determined when extracting the target coronary artery centerline; the node type division rule is a rule for determining the node type according to the number of the adjacent points of the target point;
inputting the target coronary artery central line, all target points on the target coronary artery central line and the node type of each target point into a vessel segment division model for automatic division of a target coronary artery vessel, and determining a plurality of target vessel segments in the target coronary artery;
screening out at least one stenotic vessel segment from the plurality of target vessel segments according to a lesion region in the target coronary artery and a fractional flow reserve of the lesion region;
determining a lesion score of each stenotic vessel segment according to a segment scoring rule of SYNTAX score based on the coronary distribution advantage type of the target coronary artery and lesion data of the stenotic vessel segment; wherein, the lesion data at least comprises lesion areas, lesion quantity and specific characteristics of lesions;
the lesion scores for each stenotic vessel segment are accumulated to determine the syncax score of the target patient.
Optionally, the determining the node type of each target point on the centerline of the target coronary artery according to a preset node type division rule includes:
for each target point on the centerline of the target coronary artery, determining the number of neighbors of the target point;
when the target point has only one neighbor point, determining the node type of the target point as an end point;
when the target point only has two adjacent points, determining the node type of the target point as a connection point;
when the target point has only three adjacent points, determining the node type of the target point as a bifurcation point;
when the target point has only four neighbors, the node type of the target point is determined to be a bifurcation point.
Optionally, the vessel segment division model is obtained by training through the following steps:
acquiring original three-dimensional medical images of a plurality of coronary arteries to be trained; the original three-dimensional medical image is coronary artery CT radiography;
aiming at each coronary artery to be trained, processing an original three-dimensional medical image of the coronary artery to be trained, and determining a central line of the coronary artery to be trained, all target points on the central line and a node type of each target point;
aiming at the central line of each coronary artery to be trained, segmenting the central line of the coronary artery to be trained according to a coronary artery 16 segmentation method, and determining a plurality of vessel segments to be trained of the coronary artery to be trained and segment numbers of the vessel segments to be trained;
and performing iterative training on the vessel segment division neural network by taking the central line of each coronary artery to be trained subjected to segment division, all target points on the central line and the node type of each target point as input features and taking the segment number of each vessel segment to be trained of each coronary artery to be trained as output features to obtain a vessel segment division model.
Optionally, the screening out at least one stenotic vessel segment from the plurality of target vessel segments according to a lesion region in the target coronary artery and a fractional flow reserve of the lesion region includes:
determining a target blood vessel segment meeting preset conditions as a stenosis segment; wherein the preset conditions include: the blood flow reserve fraction of the lesion area is not more than the preset blood flow reserve fraction, and the standard vessel diameter of the target vessel segment is more than the preset diameter.
Optionally, determining the coronary distribution dominance type of the target coronary artery by:
determining a segment number for each target vessel segment while determining a plurality of target vessel segments in the target coronary artery;
determining a coronary distribution dominance type of the target coronary artery based on the segment number and the mapping relation of the segment number and the dominance type of each target vessel segment in the target coronary artery; wherein the coronary distribution dominance type of the target coronary artery is used to determine a scoring weight coefficient for each target vessel segment.
Optionally, before determining the lesion score of each stenotic vessel segment, the calculation method further comprises:
for each stenotic vessel segment, determining a number of lesion regions included in the stenotic vessel segment;
when the number of the lesion areas included in the stenotic vessel segment is not less than the preset number, determining the distance between adjacent lesion areas in the stenotic vessel segment;
when the distance between the adjacent lesion areas is smaller than the preset distance between the stenotic vessel sections, combining the adjacent lesion areas into a new lesion area, and updating the lesion data of the stenotic vessel sections.
Optionally, the vessel segment dividing neural network is a Point-Transformer deep neural network.
The application embodiment also provides a computing device for coronary artery lesion SYNTAX scoring, which comprises:
the acquisition module is used for acquiring blood vessel characteristic data of a target coronary artery determined according to a coronary artery three-dimensional medical image of a target patient; at least a target coronary artery centerline, all target points on the target coronary artery centerline, a lesion region in the target coronary artery, and a fractional flow reserve of the lesion region are included in the vessel feature data;
the first determining module is used for determining the node type of each target point on the central line of the target coronary artery according to a preset node type division rule; wherein the target point is a point determined when extracting the target coronary artery centerline; the node type division rule is a rule for determining the node type according to the number of the adjacent points of the target point;
the dividing module is used for inputting the target coronary artery central line, all target points on the target coronary artery central line and the node type of each target point into a vessel segment dividing model to automatically divide a target coronary artery vessel and determine a plurality of target vessel segments in the target coronary artery;
a screening module for screening out at least one stenotic vessel segment from the plurality of target vessel segments according to a lesion region in the target coronary artery and a fractional flow reserve of the lesion region;
a second determining module, configured to determine, for each stenotic vessel segment, a lesion score of each stenotic vessel segment according to a segment scoring rule of a SYNTAX score, based on a coronary distribution advantage type of the target coronary artery and lesion data of the stenotic vessel segment; wherein, the lesion data at least comprises lesion areas, lesion quantity and specific characteristics of lesions;
and an accumulation module for accumulating the lesion scores of each stenotic vessel segment and determining a SYNTAX score of the target patient.
Optionally, when the dividing module is configured to determine the node type of each target point on the centerline of the target coronary artery according to a preset node type dividing rule, the dividing module is configured to:
for each target point on the centerline of the target coronary artery, determining the number of neighbors of the target point;
when the target point has only one neighbor point, determining the node type of the target point as an end point;
when the target point only has two adjacent points, determining the node type of the target point as a connection point;
when the target point has only three adjacent points, determining the node type of the target point as a bifurcation point;
when the target point has only four neighbors, the node type of the target point is determined to be a bifurcation point.
Optionally, the computing apparatus further includes a model training module, and the model training module is configured to:
acquiring original three-dimensional medical images of a plurality of coronary arteries to be trained; the original three-dimensional medical image is coronary artery CT radiography;
aiming at each coronary artery to be trained, processing an original three-dimensional medical image of the coronary artery to be trained, and determining a central line of the coronary artery to be trained, all target points on the central line and a node type of each target point;
aiming at the central line of each coronary artery to be trained, carrying out segment division on the central line of the coronary artery to be trained according to a coronary artery 16 segmentation method, and determining a plurality of blood vessel segments to be trained of the coronary artery to be trained and segment numbers of each blood vessel segment to be trained;
and performing iterative training on the vessel segment division neural network by taking the central line of each coronary artery to be trained subjected to segment division, all target points on the central line and the node type of each target point as input features and taking the segment number of each vessel segment to be trained of each coronary artery to be trained as output features to obtain a vessel segment division model.
Optionally, the screening module, when being configured to screen out at least one stenotic vessel segment from the plurality of target vessel segments according to a lesion region in the target coronary artery and a fractional flow reserve of the lesion region, is configured to:
determining a target blood vessel segment meeting preset conditions as a stenosis segment; wherein the preset conditions include: the blood flow reserve fraction of the lesion area is not more than the preset blood flow reserve fraction, and the standard vessel diameter of the target vessel segment is more than the preset diameter.
Optionally, the computing device further comprises a third determining module, configured to determine the coronary distribution dominance type of the target coronary artery by:
determining a segment number for each target vessel segment while determining a plurality of target vessel segments in the target coronary artery;
determining a coronary distribution dominance type of the target coronary artery based on the segment number and the mapping relation of the segment number and the dominance type of each target vessel segment in the target coronary artery; wherein the coronary distribution dominance type of the target coronary artery is used to determine a scoring weight coefficient for each target vessel segment.
Optionally, the computing apparatus further includes a merging module, where the merging module is configured to:
for each stenotic vessel segment, determining a number of lesion regions included in the stenotic vessel segment;
when the number of lesion areas included in the stenotic vessel segment is not less than the preset number, determining the distance between adjacent lesion areas in the stenotic vessel segment;
when the distance between the adjacent lesion areas is smaller than the preset distance between the stenotic vessel sections, combining the adjacent lesion areas into a new lesion area, and updating the lesion data of the stenotic vessel sections.
Optionally, the vessel segment dividing neural network is a Point-Transformer deep neural network.
An embodiment of the present application further provides an electronic device, including: a processor, a memory and a bus, the memory storing machine-readable instructions executable by the processor, the processor and the memory communicating via the bus when the electronic device is operating, the machine-readable instructions when executed by the processor performing the steps of the computing method as described above.
Embodiments of the present application further provide a computer-readable storage medium, on which a computer program is stored, where the computer program is executed by a processor to perform the steps of the computing method as described above.
The embodiment of the application provides a method and a device for calculating a SYNTAX score of coronary artery lesion, wherein the method comprises the following steps: acquiring blood vessel characteristic data of a target coronary artery determined according to a coronary artery three-dimensional medical image of a target patient; at least a target coronary artery centerline, all target points on the target coronary artery centerline, a lesion region in the target coronary artery, and a fractional flow reserve of the lesion region are included in the vessel feature data; determining the node type of each target point on the central line of the target coronary artery according to a preset node type division rule; wherein the target point is a point determined when extracting the target coronary artery centerline; the node type division rule is a rule for determining the node type according to the number of the adjacent points of the target point; inputting the target coronary artery central line, all target points on the target coronary artery central line and the node type of each target point into a vessel segment division model for automatic division of a target coronary artery vessel, and determining a plurality of target vessel segments in the target coronary artery; screening out at least one stenotic vessel segment from the plurality of target vessel segments according to a lesion region in the target coronary artery and a fractional flow reserve of the lesion region; determining a lesion score of each stenotic vessel segment according to a segment scoring rule of SYNTAX score based on the coronary distribution advantage type of the target coronary artery and lesion data of the stenotic vessel segment; wherein, the lesion data at least comprises lesion areas, lesion quantity and specific characteristics of lesions; the lesion scores for each stenotic vessel segment are accumulated to determine the syncax score of the target patient.
Therefore, the vascular characteristic data of the coronary artery of the target patient are obtained through the non-invasive three-dimensional medical image of the coronary artery, and the harm to the patient can be reduced; the automatic segmentation of the coronary artery is realized through the blood vessel segment division model, so that the segmentation efficiency can be improved, and the pressure of doctors can be reduced; the accuracy and efficiency of the score can be improved by screening the narrow blood vessel segments through the lesion areas and the blood flow reserve scores, and determining the SYNTAX score of the target patient according to the lesion score of each narrow blood vessel segment.
In order to make the aforementioned objects, features and advantages of the present application more comprehensible, preferred embodiments accompanied with figures are described in detail below.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings that are required to be used in the embodiments will be briefly described below, it should be understood that the following drawings only illustrate some embodiments of the present application and therefore should not be considered as limiting the scope, and for those skilled in the art, other related drawings can be obtained from the drawings without inventive effort.
Fig. 1 is a flowchart of a method for calculating a syncax score of a coronary artery lesion according to an embodiment of the present application;
FIG. 2 is a schematic illustration of a characteristic image of a target coronary artery acquired as provided herein;
FIG. 3 is a schematic illustration of a vessel diameter curve provided herein;
FIG. 4 is a schematic diagram of a right dominant coronary artery;
FIG. 5 is a schematic diagram of the structure of the left dominant coronary artery;
FIG. 6 is a schematic illustration of the merging of lesion areas provided herein;
fig. 7 is a schematic structural diagram of a device for calculating a coronary artery lesion SYNTAX score according to an embodiment of the present application;
fig. 8 is a second schematic structural diagram of a calculating apparatus for calculating a coronary artery lesion SYNTAX score according to an embodiment of the present application;
fig. 9 is a schematic structural diagram of an electronic device according to an embodiment of the present application.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only a part of the embodiments of the present application, and not all the embodiments. The components of the embodiments of the present application, generally described and illustrated in the figures herein, can be arranged and designed in a wide variety of different configurations. Thus, the following detailed description of the embodiments of the present application, presented in the accompanying drawings, is not intended to limit the scope of the claimed application, but is merely representative of selected embodiments of the application. Every other embodiment that can be obtained by a person skilled in the art without making creative efforts based on the embodiments of the present application falls within the protection scope of the present application.
The conventional SYNTAX score usually requires a physician to correspond each coronary vessel in a two-dimensional coronary angiography image with a coronary segment on a standard coronary anatomical pattern map, and needs to make an accurate judgment on the position, structure and degree of each coronary lesion and whether there is an adverse symptom. However, coronary angiography is an invasive examination, and requires a special contrast medium to be injected into a coronary artery opening after a puncture of a coronary artery of a body, and then detection is performed. Moreover, identification of coronary segments and lesion features in segments by means of manual labeling generally has problems of erroneous identification, low efficiency, and the like.
Based on the above, the embodiment of the application provides a method for calculating a syncax score of coronary artery lesion, which realizes full-automatic segmentation and full-automatic scoring of coronary arteries, thereby improving the efficiency of calculating the syncax score and reducing the pressure of doctors.
Here, the method for calculating the syncax score of the coronary artery lesion provided by the present application is a syncax generation III calculation method. It should be noted that the SYNTAX score has a significant guiding significance for the subsequent treatment regimen of patients with coronary heart disease. With the co-development of technology and clinical, SYNTAX has also undergone the course from generation I to generation III. The SYNTAX I generation is based entirely on coronary two-dimensional morphological information obtained by invasive coronary angiography, and a doctor usually performs manual calculation on the segmentation score of each coronary segment. In the SYNTAX generation II, some additional factors such as clinical variables, sex, age, left ventricular ejection fraction, creatinine clearance, left main lesion, chronic obstructive pulmonary disease, peripheral vascular disease and the like are added on the basis of the SYNTAX generation I to establish a multivariate proportion risk model for comprehensive generation II SYNTAX scoring, but at present, the statistics is also carried out manually by doctors. In the SYNTAX generation III, the scoring method based on the lesion degree of the blood vessel segment in the generation I is replaced by the scoring method based on the functional index CT-FFR (fractional flow reserve), but there is no automatic calculation scheme disclosed at present. The calculation method for the SYNTAX score of coronary artery lesion disclosed by the application is a SYNTAX III generation automatic calculation method.
Referring to fig. 1, fig. 1 is a flowchart illustrating a method for calculating a syncax score of coronary artery lesion according to an embodiment of the present application. As shown in fig. 1, the method for calculating a coronary artery lesion SYNTAX score provided in the embodiment of the present application includes:
s101, obtaining blood vessel characteristic data of a target coronary artery determined according to a coronary artery three-dimensional medical image of a target patient.
Here, the coronary artery three-dimensional medical image may be a coronary artery three-dimensional medical image acquired in a non-invasive manner; the vessel characteristic data at least comprises a target coronary artery central line, all target points on the target coronary artery central line, lesion areas in the target coronary artery and the fractional flow reserve of the lesion areas, and the acquired vessel characteristic data of the target coronary artery is morphological data and functional data of the target coronary artery;
wherein the morphological data of the target coronary artery refers to visualization data of the target coronary artery; the functional data of the target coronary artery refers to medical evaluation parameters of the target coronary artery.
The vessel characteristic data of the target coronary artery can also comprise a three-dimensional image of the vessel, a lesion type of a lesion region in the target coronary artery and contour information of the vessel at each target point. The fractional flow reserve (CT-FFR) of the diseased region may be calculated by a hydrodynamic simulation method.
For example, referring to fig. 2, fig. 2 is a schematic diagram of a feature image of a target coronary artery obtained according to the present application. As shown in fig. 2, the information includes information of a region in which a lesion exists in a coronary artery, a vessel diameter of the lesion region, a lesion type of the lesion region, and a fractional flow reserve of the lesion region.
S102, determining the node type of each target point on the central line of the target coronary artery according to a preset node type division rule.
Here, the target point is a point determined when the target coronary artery centerline is extracted; the node type division rule is a rule for determining the node type according to the number of the adjacent points of the target point.
When extracting the target coronary artery central line, dotting and marking the central position of a vascular tubular object in the target coronary artery, and sequentially connecting all the dotted points in sequence to obtain the target coronary artery central line, wherein the dotted point is the target point. There will be a target point at the vessel intersection location.
In an embodiment provided by the present application, the determining a node type of each target point on the centerline of the target coronary artery according to a preset node type partition rule includes: for each target point on the centerline of the target coronary artery, determining the number of neighbors of the target point; when the target point has only one neighbor point, determining the node type of the target point as an end point; when the target point only has two adjacent points, determining the node type of the target point as a connection point; when the target point has only three adjacent points, determining the node type of the target point as a bifurcation point; when the target point has only four neighbors, the node type of the target point is determined to be a bifurcation point.
Therefore, the node type corresponding to each target point on the central line can be determined according to the node type division rule.
S103, inputting the target coronary artery central line, all target points on the target coronary artery central line and the node type of each target point into a vessel segment division model to automatically divide the target coronary artery vessel, and determining a plurality of target vessel segments in the target coronary artery.
Here, the vessel segment division model determines the target vessel segment included in the target coronary artery by identifying the input target coronary artery centerline, all target points on the target coronary artery centerline, and the node type of each target point, thereby achieving automatic division of the target coronary artery vessel segment. Wherein the vessel segments are also coronary segments, and the segment number of each target vessel segment is determined at the same time when the target vessel segment is determined.
In determining the target vessel segment, redundant segments not belonging to the target vessel segment (i.e. vessels without segment numbers) are also identified. For these segments, automatic deletion may be performed, leaving only the targeted vessel segments that affect the SYNTAX score.
In one embodiment provided by the present application, the vessel segment division model is obtained by training through the following steps: acquiring original three-dimensional medical images of a plurality of coronary arteries to be trained; the original three-dimensional medical image is coronary artery CT radiography; aiming at each coronary artery to be trained, processing an original three-dimensional medical image of the coronary artery to be trained, and determining a central line of the coronary artery to be trained, all target points on the central line and a node type of each target point; aiming at the central line of each coronary artery to be trained, segmenting the central line of the coronary artery to be trained according to a coronary artery 16 segmentation method, and determining a plurality of vessel segments to be trained of the coronary artery to be trained and segment numbers of the vessel segments to be trained; and performing iterative training on the vessel segment division neural network by taking the central line of each coronary artery to be trained subjected to segment division, all target points on the central line and the node type of each target point as input features and taking the segment number of each vessel segment to be trained of each coronary artery to be trained as output features to obtain a vessel segment division model.
This step is the training process of the vessel segment segmentation model. Here, the coronary CT contrast is an image obtained by a non-invasive examination.
Here, the centerline of each coronary artery to be trained after segment division, all target points on the centerline, and the node type of each target point are used as input features, the segment number of each vessel segment to be trained of each coronary artery to be trained is used as an output feature, when the vessel segment neural network is subjected to iterative training, different weight coefficients can be given to target points of different node types according to the node types, for example, an endpoint weight coefficient is given as 2, a connection point weight coefficient is given as 1, and the like, then inputting the information into the vessel segment dividing neural network for training, comparing the predicted central line number prediction result with a real label, calculating the value of a loss function, and continuously iterating the network parameters through a back propagation algorithm until convergence, namely the value of the loss function is not reduced any more, so as to obtain the vessel segment division model.
Here, the vessel segment dividing neural network may be a Point-Transformer deep neural network. The Point-Transformer mainly comprises three types of modules, namely an encoding module, a decoding module and a characteristic combination module. The coding module is responsible for down-sampling and high-dimensional feature; and the decoding part combines the coded abstract features with the coding module, so that the details of the high-dimensional abstract result are richer. Wherein each encoding module and each decoding module possesses a self-attention mechanism.
For example, for dividing the coronary artery according to the coronary artery 16 segmentation method, all the target vessel segments that can be divided are shown in table 1, and in addition to the names and segment numbers of the vessel segments after division, the weighting coefficients of each vessel segment under different coronary artery dominance types are also recorded in table 1.
Table 1:
s104, screening out at least one narrow blood vessel segment from the plurality of target blood vessel segments according to the lesion area in the target coronary artery and the fractional flow reserve of the lesion area.
In one embodiment provided herein, the screening out at least one stenotic vessel segment from the plurality of target vessel segments according to a lesion region in the target coronary artery and a fractional flow reserve of the lesion region includes: determining a target blood vessel segment meeting preset conditions as a stenosis segment; wherein the preset conditions include: the blood flow reserve fraction of the lesion area is not more than the preset blood flow reserve fraction, and the standard vessel diameter of the target vessel segment is more than the preset diameter.
Here, the preset fractional flow reserve and the preset diameter may be selected according to the actual environment, wherein the preset diameter may be selected according to the SYNTAX scoring rule, for example, a blood vessel segment with a standard blood vessel diameter of 1.5mm or less is not used as a blood vessel for determining the stenotic blood vessel segment.
The standard vessel diameter of the target vessel segment is obtained by fitting the real diameter of the vessel corresponding to the target vessel segment.
For example, referring to fig. 3, fig. 3 is a schematic diagram of a vessel diameter curve provided by the present application, as shown in fig. 3, after a target vessel segment is determined, that is, a centerline with a segment number is determined, and a contour diameter of each target point on the centerline is plotted as a distance-diameter curve, as shown in fig. 3. The true diameter curve (true vessel diameter curve) was fitted to a standard diameter curve (standard vessel diameter curve) using the Spline curve fitting method, and vessel segments with a reference diameter of 1.5mm or less were cut off as defined by SYNTAX without entering into scoring. In the real diameter curve, if the minimum value of a continuous region is lower than 50% of the standard curve, the region is considered to have a lesion, the region is found, the functional index (blood flow reserve fraction) of the CT-FFR of the region is read, if the functional index is less than 0.8, the lesion is considered to be a lesion having a significant influence on the blood supply function, the segment is determined to be a narrow blood vessel segment, and if the CT-FFR is greater than 0.8, the assessment is not included.
S105, aiming at each narrow blood vessel segment, determining a lesion score of each narrow blood vessel segment according to a segment scoring rule of SYNTAX score based on the coronary distribution advantage type of the target coronary artery and lesion data of the narrow blood vessel segment; wherein, the lesion data at least comprises lesion areas, lesion quantity and specific characteristics of lesions.
Here, the coronary distribution dominance type is a weighting coefficient for determining a stenosis blood vessel section when performing the SYNTAX score.
Please refer to table 2, wherein table 2 is a segment scoring rule table for SYNTAX score.
Table 2:
in one embodiment provided herein, the coronary distribution dominance type of the target coronary artery is determined by: determining a segment number for each target vessel segment while determining a plurality of target vessel segments in the target coronary artery; determining a coronary distribution dominance type of the target coronary artery based on the segment number and the mapping relation of the segment number and the dominance type of each target vessel segment in the target coronary artery; wherein the coronary distribution dominance type of the target coronary is used to determine a scoring weight coefficient for each target vessel segment.
Here, the coronary distribution advantage type of the coronary artery includes a left dominant type and a right dominant type, which can be determined by the segment number of each target vessel segment included in the target coronary artery.
For example, if the target coronary artery includes 4-segment numbered target vessel segments and 16-segment numbered target vessel segments, but does not include 15-segment numbered target vessel segments, the coronary distribution dominance type of the target coronary artery is a right dominance type, please refer to fig. 4, and fig. 4 is a schematic structural diagram of the right dominance type coronary artery. If the target coronary artery does not include the target vessel segments with 4 segments and 16 segments but includes the target vessel segments with 15 segments, the coronary distribution dominant type of the target coronary artery is the left dominant type, please refer to fig. 5, and fig. 5 is a schematic structural diagram of the left dominant type coronary artery.
In another embodiment provided herein, prior to determining the lesion score for each stenotic vessel segment, the computing method further comprises: for each stenotic vessel segment, determining a number of lesion regions included in the stenotic vessel segment; when the number of the lesion areas included in the stenotic vessel segment is not less than the preset number, determining the distance between adjacent lesion areas in the stenotic vessel segment; when the distance between the adjacent lesion areas is smaller than the preset distance between the stenotic vessel sections, combining the adjacent lesion areas into a new lesion area, and updating the lesion data of the stenotic vessel sections.
This step is a merging process of the lesion areas in the vessel segment, the preset spacing being selected based on the standard diameter of the stenotic vessel segment. Wherein two lesions are combined into one lesion treatment if the distance between adjacent lesion areas is < 3 times the standard diameter of the stenotic vessel segment according to the scoring rule of SYNTAX, and the method is also applicable to three or more consecutive lesions.
For example, referring to fig. 6, fig. 6 is a schematic view of a lesion region merging provided in the present application. As shown in fig. 6, according to the merging rule, two lesion regions in the stenotic vessel segment 6 in the right drawing can be merged into one lesion region, and two lesion regions in the stenotic vessel segment 6 in the left drawing cannot be merged.
Thus, after lesion region merging, for each stenotic vessel segment, a lesion score for each stenotic vessel segment is then determined according to tables 1 and 2 based on lesion data of the stenotic vessel segment, the corresponding dominant distribution type. The method specifically comprises the following steps:
(1) and judging whether the stenotic vessel segment is completely occluded or not. If the actual diameter of the stenotic vessel segment is 0, it is considered to be a total occlusion. Firstly, determining a weight coefficient corresponding to the stenotic vessel segment, determining the weight coefficient from table 1, assuming that the weight coefficient is a, and determining the total occlusion score to be 5a from table 2; continuously scoring with additional conditions of complete occlusion, entering the (2-5) judgment, and directly jumping to the (6) condition if the real diameter is greater than 0;
(2) if the occlusion recorded in the case report is more than 3 months or the time is unknown, scoring is +1 on the original basis;
(3) aiming at the occlusion, the method is divided into three types, if the occlusion is a blunt stump or a bridge side branch, the occlusion is scored as +1, and the occlusion outside the two types is a normal occlusion and is directly jumped into (4). The blunt stump has no central line after being blocked, and the bridge side branch has two bifurcate points before and after being blocked, and two sections of central lines share the bifurcate points to be judged as the bridge side branch;
(4) if normal occlusion exists, determining that a plurality of visible segments exist after occlusion according to the occlusion position, if the segments are invisible, determining that the visible segments are +1, and each invisible segment is + 1;
(5) judging whether side branches exist after occlusion according to the number of bifurcate points of the central line, indicating the number of the side branches if the side branches exist, reading the real diameters of the side branches, scoring +1 if the diameters are less than 1.5mm, and scoring and judging each side branch as above;
(6) if only CT-FFR < ═ 0.8 instead of lesions with a diameter of 0, then a score of 2a was scored;
(7) judging whether a four-branch point exists in the region of the lesion, if the bifurcation of the lesion does not have other lesions, scoring by +3, if the other lesions exist, scoring by +1 and at most +3 according to the scoring rule that the other three bifurcations exist and one bifurcation has the lesion; it is noteworthy that only the following branches are defined as three-pronged lesions: 3/4/16/16a, 5/6/11/12, 11/12a/12b/13, 6/7/9/9a, and 7/8/10/10 a;
(8) judging whether a region where the lesion is located has a bifurcation point, if so, entering (9-10) to judge, otherwise, jumping to (11); it is noteworthy that only the following branches are defined as bifurcate lesions: 5/6/11, 6/7/9, 7/8/10, 11/13/12a, 13/14/14a, 3/4/16, and 13/14/15;
(9) bifurcate lesion typing, which is classified into types A, B, C, D, E, F, G according to Duke and Institute cardio pulmonary disease matched (ICPS) bifurcation lesion typing system, wherein ABC score +1 and DEFG score + 2;
(10) whether the bifurcation lesion angle is <70 °. Determining the direction vector of the central line from the central line of the bifurcation lesion, and solving the included angle theta between the central line and the central line: if theta is less than 70 degrees, scoring is + 1;
(11) if the lesion is located at 1 or 5 segments, or 6 or 11 segments, it is considered an open lesion, scoring + 1;
(12) and (5) judging the distortion. Calculating the curvature of a point on the central line of a segment where the lesion is located, if the angle of the segment is larger than a certain value (for example, 90 degrees), determining that distortion exists, and scoring + 2;
(13) calculating the distance from the starting point to the end point of the lesion according to the central line of the lesion area, and scoring +1 if the distance is more than 20 mm;
(14) reading the plaque information of the lesion, and scoring +2 if the lesion is seriously calcified; if yes, scoring + 1;
(15) (ii) if the lesion length is > 75% of the total vessel length and the vessel diameter is <2mm, then a diffuse lesion or small vessel lesion is considered, each segment of the vessel is scored + 1;
in this way, lesion scoring for each stenotic vessel segment is completed.
And S106, accumulating the lesion scores of each narrow blood vessel segment, and determining the SYNTAX score of the target patient.
Here, the calculated lesion scores of each stenotic vessel segment are added, and the total score after addition is determined as the SYNTAX score of the target patient.
After the SYNTAX score of the target patient is determined, the operation mode more suitable for the target patient can be determined.
The calculation method for the coronary artery lesion SYNTAX score provided by the embodiment of the application comprises the following steps: acquiring blood vessel characteristic data of a target coronary artery determined according to a coronary artery three-dimensional medical image of a target patient; at least a target coronary artery centerline, all target points on the target coronary artery centerline, a lesion region in the target coronary artery, and a fractional flow reserve of the lesion region are included in the vessel feature data; determining the node type of each target point on the central line of the target coronary artery according to a preset node type division rule; wherein the target point is a point determined when extracting the target coronary artery centerline; the node type division rule is a rule for determining the node type according to the number of the adjacent points of the target point; inputting the target coronary artery central line, all target points on the target coronary artery central line and the node type of each target point into a vessel segment division model for automatic division of a target coronary artery vessel, and determining a plurality of target vessel segments in the target coronary artery; screening out at least one stenotic vessel segment from the plurality of target vessel segments according to a lesion region in the target coronary artery and a fractional flow reserve of the lesion region; determining a lesion score of each stenotic vessel segment according to a segment scoring rule of SYNTAX score based on the coronary distribution advantage type of the target coronary artery and lesion data of the stenotic vessel segment; wherein, the lesion data at least comprises lesion areas, lesion quantity and specific characteristics of lesions; the lesion scores for each stenotic vessel segment are accumulated to determine the syncax score of the target patient.
Therefore, the vascular characteristic data of the coronary artery of the target patient are obtained through the non-invasive three-dimensional medical image of the coronary artery, and the harm to the patient can be reduced; the automatic segmentation of the coronary artery is realized through the blood vessel segment division model, so that the segmentation efficiency can be improved, and the pressure of doctors can be reduced; the accuracy and efficiency of the score can be improved by screening the narrow blood vessel segments through the lesion areas and the blood flow reserve scores, and determining the SYNTAX score of the target patient according to the lesion score of each narrow blood vessel segment.
Referring to fig. 7 and 8, fig. 7 is a first structural schematic diagram of a device for calculating a coronary artery lesion SYNTAX score according to an embodiment of the present application, and fig. 8 is a second structural schematic diagram of the device for calculating a coronary artery lesion SYNTAX score according to the embodiment of the present application. As shown in fig. 7, the computing device 700 includes:
an obtaining module 701, configured to obtain blood vessel characteristic data of a target coronary artery determined according to a coronary artery three-dimensional medical image of a target patient; at least a target coronary artery centerline, all target points on the target coronary artery centerline, a lesion region in the target coronary artery, and a fractional flow reserve of the lesion region are included in the vessel feature data;
a first determining module 702, configured to determine a node type of each target point on the centerline of the target coronary artery according to a preset node type partition rule; wherein the target point is a point determined when extracting the target coronary artery centerline; the node type division rule is a rule for determining the node type according to the number of the adjacent points of the target point;
a dividing module 703, configured to input the target coronary artery centerline, all target points on the target coronary artery centerline, and a node type of each target point into a vessel segment division model for automatic division of a target coronary artery, and determine a plurality of target vessel segments in the target coronary artery;
a screening module 704 for screening out at least one stenotic vessel segment from the plurality of target vessel segments according to a lesion region in the target coronary artery and a fractional flow reserve of the lesion region;
a second determining module 705, configured to determine, for each stenotic vessel segment, a lesion score of each stenotic vessel segment according to a segment scoring rule of a SYNTAX score based on a coronary distribution advantage type of the target coronary artery and lesion data of the stenotic vessel segment; wherein, the lesion data at least comprises lesion areas, lesion quantity and specific characteristics of lesions;
an accumulation module 706 for accumulating the lesion scores of each stenotic vessel segment, determining a syncax score of the target patient.
Optionally, when the dividing module 703 is configured to determine the node type of each target point on the centerline of the target coronary artery according to a preset node type dividing rule, the dividing module 703 is configured to:
for each target point on the centerline of the target coronary artery, determining the number of neighbors of the target point;
when the target point has only one neighbor point, determining the node type of the target point as an end point;
when the target point only has two adjacent points, determining the node type of the target point as a connection point;
when the target point has only three adjacent points, determining the node type of the target point as a bifurcation point;
when the target point has only four neighbors, the node type of the target point is determined to be a bifurcation point.
Optionally, as shown in fig. 8, the computing apparatus 700 further includes a model training module 707, where the model training module 707 is configured to:
acquiring original three-dimensional medical images of a plurality of coronary arteries to be trained; the original three-dimensional medical image is coronary artery CT radiography;
aiming at each coronary artery to be trained, processing an original three-dimensional medical image of the coronary artery to be trained, and determining a central line of the coronary artery to be trained, all target points on the central line and a node type of each target point;
aiming at the central line of each coronary artery to be trained, segmenting the central line of the coronary artery to be trained according to a coronary artery 16 segmentation method, and determining a plurality of vessel segments to be trained of the coronary artery to be trained and segment numbers of the vessel segments to be trained;
and performing iterative training on the vessel segment division neural network by taking the central line of each coronary artery to be trained subjected to segment division, all target points on the central line and the node type of each target point as input features and taking the segment number of each vessel segment to be trained of each coronary artery to be trained as output features to obtain a vessel segment division model.
Optionally, when the screening module 704 is configured to screen out at least one stenotic vessel segment from the plurality of target vessel segments according to a lesion region in the target coronary artery and a fractional flow reserve of the lesion region, the screening module 704 is configured to:
determining a target blood vessel segment meeting preset conditions as a stenosis segment; wherein the preset conditions include: the blood flow reserve fraction of the lesion area is not more than the preset blood flow reserve fraction, and the standard vessel diameter of the target vessel segment is more than the preset diameter.
Optionally, the computing apparatus 700 further comprises a third determining module 708, wherein the third determining module 708 is configured to determine the coronary distribution dominance type of the target coronary artery by:
determining a segment number for each target vessel segment while determining a plurality of target vessel segments in the target coronary artery;
determining a coronary distribution dominance type of the target coronary artery based on the segment number and the mapping relation of the segment number and the dominance type of each target vessel segment in the target coronary artery; wherein the coronary distribution dominance type of the target coronary artery is used to determine a scoring weight coefficient for each target vessel segment.
Optionally, the computing apparatus 700 further comprises a merging module 709, wherein the merging module 709 is configured to:
for each stenotic vessel segment, determining a number of lesion regions included in the stenotic vessel segment;
when the number of lesion areas included in the stenotic vessel segment is not less than the preset number, determining the distance between adjacent lesion areas in the stenotic vessel segment;
when the distance between the adjacent lesion areas is smaller than the preset distance between the stenotic vessel sections, combining the adjacent lesion areas into a new lesion area, and updating the lesion data of the stenotic vessel sections.
Optionally, the vessel segment dividing neural network is a Point-Transformer deep neural network.
Referring to fig. 9, fig. 9 is a schematic structural diagram of an electronic device according to an embodiment of the present disclosure. As shown in fig. 9, the electronic device 900 includes a processor 910, a memory 920, and a bus 930.
The memory 920 stores machine-readable instructions executable by the processor 910, when the electronic device 900 runs, the processor 910 communicates with the memory 920 through the bus 930, and when the machine-readable instructions are executed by the processor 910, the steps in the method embodiments shown in fig. 1 to fig. 6 can be performed.
An embodiment of the present application further provides a computer-readable storage medium, where a computer program is stored on the computer-readable storage medium, and when the computer program is executed by a processor, the steps in the method embodiments shown in fig. 1 to fig. 6 may be executed.
It is clear to those skilled in the art that, for convenience and brevity of description, the specific working processes of the above-described systems, apparatuses and units may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.
In the several embodiments provided in the present application, it should be understood that the disclosed system, apparatus and method may be implemented in other ways. The above-described embodiments of the apparatus are merely illustrative, and for example, the division of the units is only one logical division, and there may be other divisions when actually implemented, and for example, a plurality of units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed coupling or direct coupling or communication connection between each other may be through some communication interfaces, indirect coupling or communication connection between devices or units, and may be in an electrical, mechanical or other form.
The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.
In addition, functional units in the embodiments of the present application may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit.
The functions, if implemented in software functional units and sold or used as a stand-alone product, may be stored in a non-transitory computer-readable storage medium executable by a processor. Based on such understanding, the technical solutions of the present application or portions thereof that substantially contribute to the prior art may be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the methods described in the embodiments of the present application. And the aforementioned storage medium includes: various media capable of storing program codes, such as a usb disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk, or an optical disk.
Finally, it should be noted that: the above-mentioned embodiments are only specific embodiments of the present application, and are used for illustrating the technical solutions of the present application, but not limiting the same, and the scope of the present application is not limited thereto, and although the present application is described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: any person skilled in the art can modify or easily conceive the technical solutions described in the foregoing embodiments or equivalent substitutes for some technical features within the technical scope disclosed in the present application; such modifications, changes or substitutions do not depart from the spirit and scope of the exemplary embodiments of the present application, and are intended to be covered by the scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.
Claims (10)
1. A method for calculating a syncax score of a coronary lesion, the method comprising:
acquiring blood vessel characteristic data of a target coronary artery determined according to a coronary artery three-dimensional medical image of a target patient; at least a target coronary artery centerline, all target points on the target coronary artery centerline, a lesion region in the target coronary artery, and a fractional flow reserve of the lesion region are included in the vessel feature data;
determining the node type of each target point on the central line of the target coronary artery according to a preset node type division rule; wherein the target point is a point determined when extracting the target coronary artery centerline; the node type division rule is a rule for determining the node type according to the number of the adjacent points of the target point;
inputting the target coronary artery central line, all target points on the target coronary artery central line and the node type of each target point into a vessel segment division model for automatic division of a target coronary artery vessel, and determining a plurality of target vessel segments in the target coronary artery;
screening out at least one stenotic vessel segment from the plurality of target vessel segments according to a lesion region in the target coronary artery and a fractional flow reserve of the lesion region;
determining a lesion score of each stenotic vessel segment according to a segment scoring rule of SYNTAX score based on the coronary distribution advantage type of the target coronary artery and lesion data of the stenotic vessel segment; wherein, the lesion data at least comprises lesion areas, lesion quantity and specific characteristics of lesions;
the lesion scores for each stenotic vessel segment are accumulated to determine the syncax score of the target patient.
2. The method according to claim 1, wherein the determining the node type of each target point on the centerline of the target coronary artery according to a preset node type partition rule comprises:
for each target point on the centerline of the target coronary artery, determining the number of neighbors of the target point;
when the target point has only one neighbor point, determining the node type of the target point as an end point;
when the target point only has two adjacent points, determining the node type of the target point as a connection point;
when the target point has only three adjacent points, determining the node type of the target point as a bifurcation point;
when the target point has only four neighbors, the node type of the target point is determined to be a bifurcation point.
3. The computing method of claim 1, wherein the vessel segment segmentation model is trained by:
acquiring original three-dimensional medical images of a plurality of coronary arteries to be trained; the original three-dimensional medical image is coronary artery CT radiography;
aiming at each coronary artery to be trained, processing an original three-dimensional medical image of the coronary artery to be trained, and determining a central line of the coronary artery to be trained, all target points on the central line and a node type of each target point;
aiming at the central line of each coronary artery to be trained, segmenting the central line of the coronary artery to be trained according to a coronary artery 16 segmentation method, and determining a plurality of vessel segments to be trained of the coronary artery to be trained and segment numbers of the vessel segments to be trained;
and performing iterative training on the vessel segment division neural network by taking the central line of each coronary artery to be trained subjected to segment division, all target points on the central line and the node type of each target point as input features and taking the segment number of each vessel segment to be trained of each coronary artery to be trained as output features to obtain a vessel segment division model.
4. The method of claim 1, wherein the screening out at least one stenotic vessel segment from the plurality of target vessel segments according to a lesion region in the target coronary artery and a fractional flow reserve of the lesion region comprises:
determining a target blood vessel segment meeting preset conditions as a stenosis segment; wherein the preset conditions include: the blood flow reserve fraction of the lesion area is not more than the preset blood flow reserve fraction, and the standard vessel diameter of the target vessel segment is more than the preset diameter.
5. The computing method of claim 1, wherein the coronary distribution dominance type of the target coronary artery is determined by:
determining a segment number for each target vessel segment while determining a plurality of target vessel segments in the target coronary;
determining a coronary distribution dominance type of the target coronary artery based on the segment number and the mapping relation of the segment number and the dominance type of each target vessel segment in the target coronary artery; wherein the coronary distribution dominance type of the target coronary artery is used to determine a scoring weight coefficient for each target vessel segment.
6. The computing method of claim 1, wherein prior to determining a lesion score for each stenotic vessel segment, the computing method further comprises:
for each stenotic vessel segment, determining a number of lesion regions included in the stenotic vessel segment;
when the number of lesion areas included in the stenotic vessel segment is not less than the preset number, determining the distance between adjacent lesion areas in the stenotic vessel segment;
when the distance between the adjacent lesion areas is smaller than the preset distance between the stenotic vessel sections, combining the adjacent lesion areas into a new lesion area, and updating the lesion data of the stenotic vessel sections.
7. The computing method of claim 3, wherein the vessel segment partitioning neural network is a Point-Transformer deep neural network.
8. A computing device for a SYNTAX score of a coronary lesion, the computing device comprising:
the acquisition module is used for acquiring blood vessel characteristic data of a target coronary artery determined according to a coronary artery three-dimensional medical image of a target patient; at least a target coronary artery centerline, all target points on the target coronary artery centerline, a lesion region in the target coronary artery, and a fractional flow reserve of the lesion region are included in the vessel feature data;
the first determining module is used for determining the node type of each target point on the central line of the target coronary artery according to a preset node type division rule; wherein the target point is a point determined when extracting the target coronary artery centerline; the node type division rule is a rule for determining the node type according to the number of the adjacent points of the target point;
the dividing module is used for inputting the target coronary artery central line, all target points on the target coronary artery central line and the node type of each target point into a vessel segment dividing model to automatically divide a target coronary artery vessel and determine a plurality of target vessel segments in the target coronary artery;
a screening module for screening out at least one stenotic vessel segment from the plurality of target vessel segments according to a lesion region in the target coronary artery and a fractional flow reserve of the lesion region;
a second determining module, configured to determine, for each stenotic vessel segment, a lesion score of each stenotic vessel segment according to a segment scoring rule of a SYNTAX score, based on a coronary distribution advantage type of the target coronary artery and lesion data of the stenotic vessel segment; wherein, the lesion data at least comprises lesion areas, lesion quantity and specific characteristics of lesions;
and an accumulation module for accumulating the lesion scores of each stenotic vessel segment and determining a SYNTAX score of the target patient.
9. An electronic device, comprising: a processor, a memory and a bus, the memory storing machine-readable instructions executable by the processor, the processor and the memory communicating over the bus when the electronic device is operated, the machine-readable instructions being executable by the processor to perform the steps of the computing method of any of claims 1 to 7.
10. A computer-readable storage medium, characterized in that a computer program is stored on the computer-readable storage medium, which computer program, when being executed by a processor, performs the steps of the calculation method according to one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210570258.XA CN114947916A (en) | 2022-05-24 | 2022-05-24 | Method and device for calculating SYNTAX score of coronary artery lesion |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210570258.XA CN114947916A (en) | 2022-05-24 | 2022-05-24 | Method and device for calculating SYNTAX score of coronary artery lesion |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114947916A true CN114947916A (en) | 2022-08-30 |
Family
ID=82955244
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210570258.XA Pending CN114947916A (en) | 2022-05-24 | 2022-05-24 | Method and device for calculating SYNTAX score of coronary artery lesion |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114947916A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115953495A (en) * | 2023-03-14 | 2023-04-11 | 北京唯迈医疗设备有限公司 | Intelligent path planning device, system and storage medium based on two-dimensional radiography image |
| CN116913456A (en) * | 2023-09-12 | 2023-10-20 | 四川省医学科学院·四川省人民医院 | Interventional diagnosis and treatment evaluation method and system based on clinical indexes |
-
2022
- 2022-05-24 CN CN202210570258.XA patent/CN114947916A/en active Pending
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115953495A (en) * | 2023-03-14 | 2023-04-11 | 北京唯迈医疗设备有限公司 | Intelligent path planning device, system and storage medium based on two-dimensional radiography image |
| CN116913456A (en) * | 2023-09-12 | 2023-10-20 | 四川省医学科学院·四川省人民医院 | Interventional diagnosis and treatment evaluation method and system based on clinical indexes |
| CN116913456B (en) * | 2023-09-12 | 2023-12-12 | 四川省医学科学院·四川省人民医院 | Interventional diagnosis and treatment evaluation method and system based on clinical indexes |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11490963B2 (en) | Route selection assistance system, recording medium on which route selection assistance program is recorded, route selection assistance method, and diagnosis method | |
| KR102248498B1 (en) | Systems and methods for estimating blood flow characteristics from vessel geometry and physiology | |
| KR101910233B1 (en) | Systems and methods for numerically evaluating vasculature | |
| CN110490040B (en) | Method for identifying local vascular stenosis degree in DSA coronary artery image | |
| CN114947916A (en) | Method and device for calculating SYNTAX score of coronary artery lesion | |
| US20180243033A1 (en) | Systems and Methods for Identifying Anatomically Relevant Blood Flow Characteristics in a Patient | |
| CN114424290B (en) | Apparatus and method for providing a longitudinal display of coronary calcium loading | |
| CN108324300A (en) | Method and apparatus for segmentation of blood vessels | |
| CN111513845A (en) | Systems and methods for facilitating pre-operative prediction of post-operative tissue function | |
| Bruyninckx et al. | Segmentation of liver portal veins by global optimization | |
| US11596352B2 (en) | Route selection assistance system, recording medium on which route selection assistance program is recorded, route selection assistance method, and diagnosis method | |
| US11341645B2 (en) | Methods and systems for determining coronary hemodynamic characteristic(s) that is predictive of myocardial infarction | |
| US20220273180A1 (en) | Functional impact of vascular lesions | |
| Gharleghi et al. | Computed tomography coronary angiogram images, annotations and associated data of normal and diseased arteries | |
| CN114119645A (en) | Method, system, device and medium for determining image segmentation quality | |
| Markiewicz et al. | Computerized system for quantitative assessment of atherosclerotic plaques in the femoral and iliac arteries visualized by multislice computed tomography | |
| WO2022270152A1 (en) | Image processing device, method, and program | |
| CN115760961B (en) | Vascular image processing method, vascular image processing device, server and post-processing image generation system | |
| CN117958965A (en) | Preoperative path planning method and device for vascular intervention operation | |
| KR102446929B1 (en) | Aortic dissection visualization apparatus and method | |
| CN112307804A (en) | Blood vessel state evaluation method and blood vessel state evaluation device | |
| Tokunaga et al. | Extraction of the thoracic aorta territory and aneurysm from CT images | |
| JP2024515188A (en) | Systems and methods for identifying coronary artery disease - Patents.com | |
| CN115115657A (en) | Focus segmentation method and device, electronic device and storage medium | |
| KR20240039479A (en) | Medical image visualization apparatus and method for diagnosis of aorta |
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 |