CN109907732A - A kind of appraisal procedure and system of rupture of intracranial aneurysm risk - Google Patents

A kind of appraisal procedure and system of rupture of intracranial aneurysm risk Download PDF

Info

Publication number
CN109907732A
CN109907732A CN201910280012.7A CN201910280012A CN109907732A CN 109907732 A CN109907732 A CN 109907732A CN 201910280012 A CN201910280012 A CN 201910280012A CN 109907732 A CN109907732 A CN 109907732A
Authority
CN
China
Prior art keywords
parameter
virtual
target
risk
aneurysm
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN201910280012.7A
Other languages
Chinese (zh)
Other versions
CN109907732B (en
Inventor
区初斌
钱逸
李洲健
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Weizhi Medical Technology Foshan Co ltd
Original Assignee
Guangzhou Core Vein Technology Co Ltd
Guangzhou Xinmai Technology Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Guangzhou Core Vein Technology Co Ltd, Guangzhou Xinmai Technology Co Ltd filed Critical Guangzhou Core Vein Technology Co Ltd
Priority to CN201910280012.7A priority Critical patent/CN109907732B/en
Publication of CN109907732A publication Critical patent/CN109907732A/en
Application granted granted Critical
Publication of CN109907732B publication Critical patent/CN109907732B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Landscapes

  • Apparatus For Radiation Diagnosis (AREA)
  • Magnetic Resonance Imaging Apparatus (AREA)

Abstract

The invention discloses the appraisal procedures and system of a kind of rupture of intracranial aneurysm risk, this method is applied in the assessment system of rupture of intracranial aneurysm risk, this method comprises: being established according to encephalic image data including the aneurysmal threedimensional model on parent artery and parent artery；Virtual parent artery and virtual aneurysmal target morphology parameter are determined based on threedimensional model；Virtual parent artery and virtual aneurysmal target hemodynamic parameter are determined based on threedimensional model；Operation is carried out to target morphology parameter, target hemodynamic parameter and target clinical parameter based on preparatory trained machine learning model, obtains virtual aneurysmal assessment result, the assessment result is for assessing aneurysmal risk of rupture.As it can be seen that rupture of intracranial aneurysm risk can be automatically analyzed comprehensively by Morphologic Parameters, hemodynamic parameter and the clinical parameter to patient's intracranial aneurysm by implementing the present invention, the accuracy rate of aneurysm rupture risk assessment can be improved.

Description

A kind of appraisal procedure and system of rupture of intracranial aneurysm risk
Technical field
The present invention relates to intracranial aneurysm medicine technology field more particularly to a kind of assessments of rupture of intracranial aneurysm risk Method and system.
Background technique
Intracranial aneurysm, also known as encephalic angioma are to form a kind of warty of arterial wall by entocranial artery inner cavity abnormal dilatation Protrusion is a kind of common vascular conditions.According to statistics, in every 100 adults in China, just having 7 is aneurysm carrier. Intracranial aneurysm can be divided into non-ruptured aneurysm and ruptured aneurysm, and most intracranial aneurysms is non-ruptured aneurysm, It will not generally rupture throughout one's life, a year rupture rate is only 0.05%.However non-ruptured aneurysm can cause spontaneous spider once rupturing Gap bleeding under nethike embrane becomes ruptured aneurysm, and lethal disability rate is more than 50%, seriously threatens the life of patient.
The evaluation measures of rupture of intracranial aneurysm risk are mainly the evaluation measures based on PHASES score, the evaluation at present It is dynamic from the analysis of aneurysm site, aneurysm size, patient groups, patient's passing medical history and patient age that means are based on statistics Arteries and veins tumor, to deduce aneurysmal 5 years risk of rupture.However, practice discovery, which is only to aneurysm illness Crowd carries out statistical analysis, has ignored the analysis of aneurysm individual patients actual conditions, therefore to aneurysmal risk assessment Accuracy rate it is lower.
Summary of the invention
Technical problem to be solved by the present invention lies in provide a kind of appraisal procedure of rupture of intracranial aneurysm risk and be System, can be by the Morphologic Parameters, hemodynamic parameter and clinical parameter of intracranial aneurysm to the entocranial artery of patient Tumor risk of rupture is automatically analyzed comprehensively, to improve the accuracy rate of aneurysm rupture risk assessment.
In order to solve the above-mentioned technical problem, first aspect of the embodiment of the present invention discloses a kind of rupture of intracranial aneurysm risk Appraisal procedure, the method be applied to rupture of intracranial aneurysm risk assessment system in, which comprises
It is established according to encephalic image data including the aneurysmal threedimensional model on parent artery and the parent artery, institute Threedimensional model is stated to include the virtual parent artery to match with the parent artery and match with the aneurysm virtual Aneurysm；
Determine that target morphology parameter, the target morphology parameter include the virtual load tumor based on the threedimensional model The Morphologic Parameters of artery and the virtual aneurysmal Morphologic Parameters；
Determine that target hemodynamic parameter, the target hemodynamic parameter include described based on the threedimensional model The hemodynamic parameter of virtual parent artery and the virtual aneurysmal hemodynamic parameter；
The target morphology parameter, the target haemodynamics are joined based on preparatory trained machine learning model Several and target clinical parameter carries out operation, obtains the virtual aneurysmal assessment result, the assessment result is for assessing The aneurysmal risk of rupture, the target clinical parameter include predefining the corresponding use of the encephalic image data of going out The clinical parameter at family.
As an alternative embodiment, in first aspect of the embodiment of the present invention, it is described to be based on the threedimensional model Determine target morphology parameter, comprising:
Divide the threedimensional model based on predetermined model partitioning algorithm, the threedimensional model after being divided, and base Threedimensional model after the segmentation determines target morphology parameter；
Wherein, described that the threedimensional model, the three-dimensional after being divided are divided based on predetermined model partitioning algorithm Model, comprising:
The a certain pixel position on the virtual parent artery is determined based on predetermined model partitioning algorithm As the first wave source point of the first analog wave, and determine a certain pixel position in the virtual aneurysm as the Second wave source point of two analog waves, first analog wave and second analog wave are same type of analog wave；
Emit first analog wave and second analog wave simultaneously, and records first analog wave and described the The propagation duration of two analog waves, the starting for propagating duration propagate the moment to emit first analog wave and second mould At the time of quasi- wave, the termination propagation moment for propagating duration is the wave crest and second analog wave of first analog wave At the time of wave crest is overlapped for the first time；
Determine that first analog wave propagates covered region in the propagation duration and second analog wave passes Covered the sum of region is broadcast, as cut zone corresponding with the virtual parent artery and the virtual aneurysm, And the threedimensional model is divided according to the cut zone, the threedimensional model after being divided.
As an alternative embodiment, in first aspect of the embodiment of the present invention, it is described based on trained in advance Machine learning model transports the target morphology parameter, the target hemodynamic parameter and target clinical parameter It calculates, after obtaining the virtual aneurysmal assessment result, the method also includes:
The assessment report of the encephalic image data is generated according to the assessment result, the assessment report includes the mesh Mark the morphology risk analysis result of Morphologic Parameters, the haemodynamics risk analysis knot of the target hemodynamic parameter Fruit, the clinical risk analysis result of the target clinical parameter and integrated risk analysis are as a result, the integrated risk analysis knot Fruit is to be analyzed based on the morphology risk analysis result, the haemodynamics risk analysis result and the clinical risk As a result the result generated；
The integrated risk analysis result for determining that the assessment report includes based on default risk class rule is corresponding Risk class, and show the assessment report and the corresponding risk class of the integrated risk analysis result；
And the method also includes:
Based on the target morphology parameter, the target hemodynamic parameter and the target clinical parameter from pre- The determining goal-based assessment most like with the assessment report is reported in the aneurysm database first established, and shows that the target is commented Estimate report, the aneurysm database pre-established is for storing each aneurysm patient in all arteries tumor patient Assessment report.
As an alternative embodiment, in first aspect of the embodiment of the present invention, it is described to be based on the threedimensional model Determine target morphology parameter, comprising:
The dynamic of the virtual parent artery is extracted based on predetermined central line pick-up algorithm and the threedimensional model Arteries and veins center line；
Algorithm is distinguished based on predetermined region, region differentiation is carried out to the threedimensional model, obtain target area, institute State arterial inlet region, the virtual parent artery that target area includes the virtual parent artery artery exit region, The artery wall region of the virtual parent artery and the virtual aneurysmal aneurysm wall region；
Computational geometry analysis is carried out to the content that the target area includes based on predetermined Morphologic Parameters algorithm, Obtain target morphology parameter.
As an alternative embodiment, in first aspect of the embodiment of the present invention, it is described to be based on predetermined area Domain distinguishes algorithm and carries out region differentiation to the threedimensional model, obtains target area, comprising:
Algorithm, which is distinguished, based on predetermined region determines that the arterial inlet cross section of the virtual parent artery, artery go out Mouth cross section and the virtual aneurysmal tumor neck plane；
According to threedimensional model described in the arterial inlet cross-sectional cut, the arterial inlet of the virtual parent artery is obtained Region, and the threedimensional model is cut according to the exit cross-section, obtain the artery outlet area of the virtual parent artery Domain, and the threedimensional model according to the tumor neck plane cutting, obtain the virtual parent artery artery wall region and The virtual aneurysmal aneurysm wall region.
As an alternative embodiment, in first aspect of the embodiment of the present invention, it is described to be based on the threedimensional model Determine target hemodynamic parameter, comprising:
The threedimensional model is divided based on predetermined reseau-dividing algorithm, obtains multiple objective body grid models, institute Having the objective body grid model is polyhedron network lattice model；
Determine that the object boundary condition of the threedimensional model, the object boundary condition are dynamic including at least the virtual load tumor The boundary condition of the artery outlet border of the boundary condition on the arterial inlet boundary of arteries and veins, the virtual parent artery；
Based on virtual parent artery described in all objective body grid models and the object boundary condition simulation Blood flow and the virtual aneurysmal blood flow, obtain target hemodynamic parameter.
As an alternative embodiment, in first aspect of the embodiment of the present invention, it is described based on trained in advance Machine learning model transports the target morphology parameter, the target hemodynamic parameter and target clinical parameter It calculates, before obtaining the virtual aneurysmal assessment result, the method also includes:
Machine learning model is constructed, the machine learning model is for determining the virtual aneurysmal assessment result；
Wherein, the building machine learning model, comprising:
Obtain multiple sample encephalic image datas, each corresponding aneurysm patient of the sample encephalic image data；
Determine the sample characteristics parameter of each sample encephalic image data, each sample characteristics parameter includes sample This Morphologic Parameters, sample hemodynam ics param eter corresponding with the sample Morphologic Parameters and with the sample Morphologic Parameters Corresponding sample clinical parameter；
Each sample characteristics parameter is marked, the sample after sample characteristics parameter and the label after being marked The corresponding mark value of characteristic parameter；
Based on the sample characteristics parameter after each label of predetermined training algorithm training, machine learning mould is obtained Type；
Wherein, each sample characteristics parameter of label, the corresponding label of sample characteristics parameter after being marked Value, comprising:
When the aneurysm that the sample encephalic image data includes is ruptured aneurysm, then label and the sample encephalic image The mark value of the corresponding sample characteristics parameter of data is the first preset value；
When the aneurysm that the sample encephalic image data includes is non-ruptured aneurysm, then label and the sample encephalic shadow As the mark value of the corresponding sample characteristics parameter of data is the second preset value；
Wherein, first preset value and second preset value be not identical.
Second aspect of the embodiment of the present invention discloses a kind of assessment system of rupture of intracranial aneurysm risk, the assessment system System includes three-dimensional reconstruction module, Morphological measurement module, haemodynamics computing module and integrated risk computing module, In:
The three-dimensional reconstruction module includes that target parent artery and the target carry for being established according to encephalic image data The eparterial aneurysmal threedimensional model of tumor, the threedimensional model include the virtual load tumor to match with the target parent artery Artery and the virtual aneurysm to match with the aneurysm；
The Morphological measurement module, for determining target morphology parameter, the target shape based on the threedimensional model State parameter includes the Morphologic Parameters and the virtual aneurysmal Morphologic Parameters of the virtual parent artery；
The haemodynamics computing module, for determining target hemodynamic parameter, institute based on the threedimensional model State hemodynamic parameter that target hemodynamic parameter includes the virtual parent artery and described virtual aneurysmal Hemodynamic parameter；
The integrated risk computing module, for being based on preparatory trained machine learning model to the target morphology Parameter, the target hemodynamic parameter and target clinical parameter carry out operation, obtain the aneurysmal assessment result, The assessment result is the predetermined encephalic for assessing the aneurysmal risk of rupture, the target clinical parameter The clinical parameter of the corresponding user of image data.
As an alternative embodiment, in second aspect of the embodiment of the present invention, the Morphological measurement module packet It includes model segmentation submodule and morphology determines submodule, in which:
The model divides submodule, for dividing the threedimensional model based on predetermined model partitioning algorithm, obtains Threedimensional model after to segmentation；
The morphology determines submodule, for determining target morphology parameter based on the threedimensional model after the segmentation；
Wherein, the model segmentation submodule is based on the predetermined model partitioning algorithm segmentation threedimensional model, obtains The mode of threedimensional model after to segmentation specifically:
The a certain pixel position on the virtual parent artery is determined based on predetermined model partitioning algorithm As the first wave source point of the first analog wave, and determine a certain pixel position in the virtual aneurysm as the Second wave source point of two analog waves, first analog wave and second analog wave are same type of analog wave；
Emit first analog wave and second analog wave simultaneously, and records first analog wave and described the The propagation duration of two analog waves, the starting for propagating duration propagate the moment to emit first analog wave and second mould At the time of quasi- wave, the termination propagation moment for propagating duration is the wave crest and second analog wave of first analog wave At the time of wave crest is overlapped for the first time；
Determine that first analog wave propagates covered region in the propagation duration and second analog wave passes Covered the sum of region is broadcast, as cut zone corresponding with the virtual parent artery and the virtual aneurysm, And the threedimensional model is divided according to the cut zone, the threedimensional model after being divided.
As an alternative embodiment, in second aspect of the embodiment of the present invention, the integrated risk computing module, It is also used to joining the target morphology parameter, the target haemodynamics based on preparatory trained machine learning model Several and target clinical parameter carries out operation, after obtaining the virtual aneurysmal assessment result, according to the assessment result The assessment report of the encephalic image data is generated, the assessment report includes the morphology risk of the target morphology parameter Analysis result, the haemodynamics risk analysis result of the target hemodynamic parameter, the target clinical parameter face Bed risk analysis result and integrated risk analysis are as a result, the integrated risk analysis result is based on the morphology risk point Analyse the result that result, the haemodynamics risk analysis result and clinical risk analysis result generate；
The integrated risk computing module, be also used to determine that the assessment report includes based on predetermined level rule described in The corresponding risk class of integrated risk analysis result, and show that the assessment report and the integrated risk analysis result are corresponding Risk class；
And the integrated risk computing module, it is also used to based on the target morphology parameter, the target blood flow Mechanics parameter and the target clinical parameter determine and the assessment report most phase from the aneurysm database pre-established As goal-based assessment report that and show goal-based assessment report, the aneurysm database pre-established is for storing institute There is the assessment report of each aneurysm patient in aneurysm patient.
As an alternative embodiment, in second aspect of the embodiment of the present invention, the Morphological measurement module packet Include central line pick-up submodule, submodule and morphology computational submodule are distinguished in region, in which:
The central line pick-up submodule, for being based on predetermined central line pick-up algorithm and the threedimensional model Extract the artery center line of the virtual parent artery；
Submodule is distinguished in the region, carries out area to the threedimensional model for distinguishing algorithm based on predetermined region Domain is distinguished, and obtains target area, the target area includes the arterial inlet region of the virtual parent artery, the virtual load The artery exit region of tumor artery, the artery wall region of the virtual parent artery and the virtual aneurysmal aneurysm wall Region；
The morphology computational submodule, for being wrapped based on predetermined Morphologic Parameters algorithm to the target area The content included carries out computational geometry analysis, obtains target morphology parameter.
As an alternative embodiment, submodule base is distinguished in the region in second aspect of the embodiment of the present invention Algorithm is distinguished in predetermined region, region differentiation is carried out to the threedimensional model, obtain the mode of target area specifically:
Algorithm, which is distinguished, based on predetermined region determines that the arterial inlet cross section of the virtual parent artery, artery go out Mouth cross section and the virtual aneurysmal tumor neck plane；
According to threedimensional model described in the arterial inlet cross-sectional cut, the arterial inlet of the virtual parent artery is obtained Region, and the threedimensional model is cut according to the exit cross-section, obtain the artery outlet area of the virtual parent artery Domain, and the threedimensional model according to the tumor neck plane cutting, obtain the virtual parent artery artery wall region and The virtual aneurysmal aneurysm wall region.
As an alternative embodiment, the haemodynamics calculates mould in second aspect of the embodiment of the present invention Block determines the mode of target hemodynamic parameter based on the threedimensional model specifically:
The threedimensional model is divided based on predetermined reseau-dividing algorithm, obtains multiple objective body grid models, institute Having the objective body grid model is polyhedron network lattice model；
Determine that the object boundary condition of the threedimensional model, the object boundary condition are dynamic including at least the virtual load tumor The boundary condition of the artery outlet border of the boundary condition on the arterial inlet boundary of arteries and veins, the virtual parent artery；
Based on virtual parent artery described in all objective body grid models and the object boundary condition simulation Blood flow and the virtual aneurysmal blood flow, obtain target hemodynamic parameter.
As an alternative embodiment, the assessment system further includes mould in second aspect of the embodiment of the present invention Type constructs module, in which:
The model construction module, for being based on preparatory trained machine learning mould in the integrated risk computing module Type carries out operation to the target morphology parameter, the target hemodynamic parameter and target clinical parameter, obtains institute Before stating virtual aneurysmal assessment result, machine learning model is constructed, the machine learning model is described virtual for determining Aneurysmal assessment result；
Wherein, the mode of the model construction module building machine learning model specifically:
Obtain multiple sample encephalic image datas, each corresponding aneurysm patient of the sample encephalic image data；
Determine the sample characteristics parameter of each sample encephalic image data, each sample characteristics parameter includes sample This Morphologic Parameters, sample hemodynam ics param eter corresponding with the sample Morphologic Parameters and with the sample Morphologic Parameters Corresponding sample clinical parameter；
Each sample characteristics parameter is marked, the mark value of each sample characteristics parameter is obtained；
Based on each sample characteristics parameter of predetermined training algorithm training and each sample characteristics ginseng Several mark values, obtains machine learning model；
Wherein, the model construction module marks each sample characteristics parameter, obtains each sample characteristics ginseng The mode of several values of statistical indicant specifically:
When the aneurysm that the sample encephalic image data includes is ruptured aneurysm, then label and the sample encephalic image The mark value of the corresponding sample characteristics parameter of data is the first preset value；
When the aneurysm that the sample encephalic image data includes is non-ruptured aneurysm, then label and the sample encephalic shadow As the mark value of the corresponding sample characteristics parameter of data is the second preset value；
Wherein, first preset value and second preset value be not identical.
Third aspect present invention discloses the assessment system of another rupture of intracranial aneurysm risk, and described device includes:
It is stored with the memory of executable program code；
The processor coupled with the memory；
The processor calls the executable program code stored in the memory, executes first aspect present invention The appraisal procedure of disclosed rupture of intracranial aneurysm risk.
Fourth aspect present invention disclose a kind of computer can storage medium, the computer storage medium is stored with calculating Machine instruction, when the computer instruction is called, for executing rupture of intracranial aneurysm risk disclosed in first aspect present invention Appraisal procedure.
Compared with prior art, the embodiment of the present invention has the advantages that
In the embodiment of the present invention, it includes aneurysmal on parent artery and parent artery for being established according to encephalic image data Threedimensional model, the threedimensional model include the virtual parent artery to match with parent artery and match with aneurysm virtual Aneurysm；Target morphology parameter is determined based on threedimensional model, which includes the form of virtual parent artery Learn parameter and virtual aneurysmal Morphologic Parameters；Target hemodynamic parameter is determined based on threedimensional model, the target blood Hydromechanics parameter includes the hemodynamic parameter and virtual aneurysmal hemodynamic parameter of virtual parent artery；Base In preparatory trained machine learning model to target morphology parameter, target hemodynamic parameter and target clinical parameter Operation is carried out, virtual aneurysmal assessment result is obtained, for the assessment result for assessing aneurysmal risk of rupture, target is clinical Parameter includes the clinical parameter for predefining the corresponding user of encephalic image data.As it can be seen that implementing the present invention can pass through To the Morphologic Parameters, hemodynamic parameter and clinical parameter of patient's intracranial aneurysm to rupture of intracranial aneurysm risk into Row automatically analyzes comprehensively, can be improved the accuracy rate of aneurysm rupture risk assessment, to mention for doctor when making Treatment decsion For scientific reference, so that doctor quickly provides effective therapeutic scheme to patient.
Detailed description of the invention
To describe the technical solutions in the embodiments of the present invention more clearly, make required in being described below to embodiment Attached drawing is briefly described, it should be apparent that, drawings in the following description are only some embodiments of the invention, for For those of ordinary skill in the art, without creative efforts, it can also be obtained according to these attached drawings other Attached drawing.
Fig. 1 is a kind of flow diagram of the appraisal procedure of rupture of intracranial aneurysm risk disclosed by the embodiments of the present invention；
Fig. 2 is the structural representation of the threedimensional model of the virtual parent artery after a kind of segmentation disclosed by the embodiments of the present invention Figure；
Fig. 3 is the structural representation of the threedimensional model of the virtual parent artery after a kind of cutting disclosed by the embodiments of the present invention Figure；
Fig. 4 is a kind of three-dimensional mould of virtual parent artery using tumor neck plane as line of demarcation disclosed by the embodiments of the present invention The structural schematic diagram of type；
Fig. 5 is a kind of structural schematic diagram of the assessment system of rupture of intracranial aneurysm risk disclosed by the embodiments of the present invention；
Fig. 6 is the structural representation of the assessment system of another rupture of intracranial aneurysm risk disclosed by the embodiments of the present invention Figure；
Fig. 7 is the structural representation of the assessment system of another rupture of intracranial aneurysm risk disclosed by the embodiments of the present invention Figure；
Fig. 8 is the structural representation of the assessment system of another rupture of intracranial aneurysm risk disclosed by the embodiments of the present invention Figure.
Specific embodiment
In order to enable those skilled in the art to better understand the solution of the present invention, below in conjunction in the embodiment of the present invention Attached drawing, technical scheme in the embodiment of the invention is clearly and completely described, it is clear that described embodiment is only A part of the embodiment of the present invention, instead of all the embodiments.Based on the embodiments of the present invention, those of ordinary skill in the art Every other embodiment obtained without making creative work, shall fall within the protection scope of the present invention.
Description and claims of this specification and term " first " in above-mentioned attached drawing, " second " etc. are for distinguishing Different objects, are not use to describe a particular order.In addition, term " includes " and " having " and their any deformations, it is intended that It is to cover and non-exclusive includes.Such as process, method, device, product or the equipment for containing a series of steps or units do not have It is defined in listed step or unit, but optionally further comprising the step of not listing or unit, or optionally also wrap Include other step or units intrinsic for these process, methods, product or equipment.
Referenced herein " embodiment " is it is meant that a particular feature, structure, or characteristic described can wrap in conjunction with the embodiments Containing at least one embodiment of the present invention.Each position in the description occur the phrase might not each mean it is identical Embodiment, nor the independent or alternative embodiment with other embodiments mutual exclusion.Those skilled in the art explicitly and Implicitly understand, embodiment described herein can be combined with other embodiments.
The invention discloses the appraisal procedure and system of a kind of rupture of intracranial aneurysm risk, this method is dynamic applied to encephalic In the assessment system of arteries and veins tumor risk of rupture, Morphologic Parameters, the hemodynamic parameter to patient's intracranial aneurysm can be passed through And clinical parameter automatically analyzes rupture of intracranial aneurysm risk comprehensively, can be improved aneurysm rupture risk assessment Accuracy rate, so that scientific reference is provided when making Treatment decsion for doctor, so that doctor is quickly provided with to patient The therapeutic scheme of effect.It is described in detail separately below.
Embodiment one
Referring to Fig. 1, Fig. 1 is a kind of stream of the appraisal procedure of rupture of intracranial aneurysm risk disclosed by the embodiments of the present invention Journey schematic diagram.Wherein, method described in Fig. 1 can be applied in risk assessment terminal, and the risk assessment terminal includes All terminals that can assess the aneurysmal risk of rupture of patient.Further, the risk assessment terminal can also and user terminal Be wirelessly connected, wherein the user terminal may include smart phone (Android phone, iOS mobile phone etc.), smart phone wrist-watch, Tablet computer, palm PC, vehicle-mounted computer, desktop computer, net book, personal digital assistant (Personal Digital Assistant, PDA), Intelligent navigator and mobile internet device (Mobile Internet Devices, MID) etc. eventually End, the embodiment of the present invention without limitation, as shown in Figure 1, the appraisal procedure of the rupture of intracranial aneurysm risk may include following Operation:
101, risk assessment terminal is established according to encephalic image data including the aneurysm on parent artery and parent artery Threedimensional model, which includes the virtual parent artery to match with parent artery and the void to match with aneurysm Quasi- aneurysm.
In the embodiment of the present invention, encephalic image data includes but is not limited to CTA encephalic image data, MRA encephalic image number Any one accordingly and in DSA encephalic image data.
In the embodiment of the present invention, as an alternative embodiment, risk assessment terminal is built according to encephalic image data Vertical includes the aneurysmal threedimensional model on parent artery and parent artery, may include:
Risk assessment terminal acquires the encephalic image data of patient, and based on predetermined Threshold Segmentation Algorithm to encephalic Image data progress Threshold segmentation processing, the encephalic image data after being divided, and according to the encephalic image number after segmentation According to the aneurysmal threedimensional model that foundation includes on parent artery and parent artery.
Wherein, risk assessment terminal is based on predetermined Threshold Segmentation Algorithm and carries out Threshold segmentation to encephalic image data It handles, may include:
Risk assessment terminal is based on predetermined Threshold Segmentation Algorithm and carries out gray proces to encephalic image data, obtains Encephalic image data after gray proces, and gray value in encephalic image data is more than or equal to the pixel value for presetting gray threshold Labeled as the first preset value, the pixel value that gray value in encephalic image data is less than default gray threshold is preset labeled as second Value.
In the optional embodiment, the first preset value and the second preset value be not identical.Specifically, when the first preset value is When 1, the second preset value is 0；When the first preset value is 0, the second preset value is 1, and the embodiment of the present invention is without limitation.
In the optional embodiment, predetermined Threshold Segmentation Algorithm may include Otsu Threshold Segmentation Algorithm, from Adapt to Threshold Segmentation Algorithm, maximum entropy threshold partitioning algorithm, Roberts Threshold Segmentation Algorithm, Prewitt Threshold Segmentation Algorithm, Sobel Threshold Segmentation Algorithm, Marr-Hilderth Threshold Segmentation Algorithm, any one algorithm in Canny Threshold Segmentation Algorithm Or many algorithms combination, the embodiment of the present invention is without limitation.
As a kind of optional embodiment, risk assessment terminal is based on encephalic image data and establishes including parent artery and be somebody's turn to do It, can be on parent artery before aneurysmal threedimensional model:
Judge whether the quality of image index of encephalic image data reaches default quality of image index；
When the result judged is is, triggering execute it is above-mentioned according to encephalic image data establish include parent artery and should The operation of aneurysmal threedimensional model on parent artery；
When the judgment result is no, above-mentioned encephalic image data is pre-processed based on default image algorithm, until The quality of image index of the encephalic image data reaches default quality of image index, and it is above-mentioned according to encephalic image to trigger execution Data establish the operation including aneurysmal threedimensional model on parent artery and the parent artery.
In the optional embodiment, the quality of image index of encephalic image data includes but is not limited to encephalic image data Color depth, the resolution ratio of encephalic image data, the image fault of encephalic image data, the data format of encephalic image data, The file size of encephalic image data.
As it can be seen that the optional embodiment passes through in the aneurysmal threedimensional model established on parent artery and parent artery Before, judge whether the quality of image index of encephalic image data reaches preset standard, when not reaching preset standard, by cranium The quality of image index of interior image data is pre-processed, until reach default quality of image index just establish parent artery and Aneurysmal threedimensional model on parent artery, can reduce because quality of image index is not up to standard cause to establish parent artery with And aneurysmal threedimensional model it is not accurate a possibility that.
102, risk assessment terminal determines target morphology parameter based on above-mentioned threedimensional model, the target morphology parameter packet Include the Morphologic Parameters and above-mentioned virtual aneurysmal Morphologic Parameters of above-mentioned virtual parent artery.
In the embodiment of the present invention, target morphology parameter may include virtual aneurysmal knurl footpath, virtual aneurysmal tumor The outflow of high, virtual aneurysmal tumor wide, virtually aneurysmal neck breadth, virtual parent artery fluid inlet angle, virtual parent artery Angle, virtual aneurysmal aspherical index, virtual aneurysmal oval index, virtual aneurysmal length-width ratio, virtual aneurysm At least one of morphological parameters such as angle of deviation, the embodiment of the present invention is without limitation.
In the embodiment of the present invention, as an alternative embodiment, risk assessment terminal is true based on above-mentioned threedimensional model Set the goal Morphologic Parameters, may include:
Risk assessment terminal is based on predetermined model partitioning algorithm and divides above-mentioned threedimensional model, and three after being divided Dimension module, and target morphology parameter is determined based on the threedimensional model after segmentation；
In the optional embodiment, optionally, risk assessment terminal is divided based on predetermined model partitioning algorithm Above-mentioned threedimensional model, the threedimensional model after being divided may include:
Risk assessment terminal determines a certain picture on above-mentioned virtual parent artery based on predetermined model partitioning algorithm First wave source point (first wave source point referring to figure 2. in) of the vegetarian refreshments position as the first analog wave, and determination are above-mentioned Second wave source point (second referring to figure 2. in of a certain pixel position as the second analog wave in virtual aneurysm Wave source point), first analog wave and second analog wave are same type of analog wave.
Risk assessment terminal emits above-mentioned first analog wave and above-mentioned second analog wave simultaneously, and records the first analog wave The starting propagation moment with the propagation duration of the second analog wave, the propagation duration is the first analog wave of transmitting and the second analog wave The wave crest overlapping for the first time of the wave crest that the moment is the first analog wave and the second analog wave is propagated in moment, the termination of the propagation duration Moment；
Risk assessment terminal determines that the first analog wave propagates covered region and the second simulation in above-mentioned propagation duration Wave propagates covered the sum of region, as cut section corresponding with above-mentioned virtual parent artery and above-mentioned virtual aneurysm Domain, and above-mentioned threedimensional model is divided according to the cut zone, the threedimensional model after being divided.
In the optional embodiment, predetermined model partitioning algorithm may include that collision front-end algorithm (is also referred to as collided Detection algorithm), the embodiment of the present invention is without limitation.
In the embodiment of the present invention, above-mentioned threedimensional model is divided based on predetermined model partitioning algorithm, after obtaining segmentation Threedimensional model can with as shown in Fig. 2, Fig. 2 be the virtual parent artery after a kind of segmentation disclosed by the embodiments of the present invention three-dimensional The structural schematic diagram of model.
The optional embodiment divides threedimensional model by predetermined model partitioning algorithm, can more accurately determine Position and the virtual parent artery of reconstruction and virtual aneurysmal threedimensional model, so that it is dynamic to help to obtain more accurate virtual load tumor Arteries and veins and virtual aneurysmal Morphologic Parameters and hemodynamic parameter.
In the embodiment of the present invention, as another optional embodiment, risk assessment terminal is based on above-mentioned threedimensional model It determines target morphology parameter, may include:
Risk assessment terminal is based on predetermined central line pick-up algorithm and the extraction of above-mentioned threedimensional model is above-mentioned virtual The artery center line of parent artery；
Risk assessment terminal is based on predetermined region and distinguishes algorithm to the progress region differentiation of above-mentioned threedimensional model, obtains Target area, the target area include the artery in the arterial inlet region of above-mentioned virtual parent artery, above-mentioned virtual parent artery Exit region, the artery wall region of above-mentioned virtual parent artery and above-mentioned virtual aneurysmal aneurysm wall region；
Risk assessment terminal carries out the content that above-mentioned target area includes based on predetermined Morphologic Parameters algorithm Computational geometry analysis, obtains target morphology parameter.
In the optional embodiment, predetermined central line pick-up algorithm includes but is not limited to be based on topological thinning calculation Method, based on Hessian tracing algorithm, based on any one in distance transform algorithm.
It is further alternative in the optional embodiment, when predetermined central line pick-up algorithm is above-mentioned is based on When distance transform algorithm, risk assessment terminal is based on predetermined central line pick-up algorithm and above-mentioned threedimensional model extracts The artery center line for stating virtual parent artery may include:
Risk assessment terminal determines the beginning node and end segment of above-mentioned virtual parent artery based on distance transform algorithm Point；
Risk assessment terminal generates the Wei Nuotu of above-mentioned threedimensional model based on above-mentioned beginning node with above-mentioned endpoint node, and Record third analog wave from the beginning node travel to all destination nodes of Wei Nuotu needed for arrival time, wherein the institute Having destination node is all nodes on Wei Nuotu other than beginning node；
Risk assessment terminal determines the destination node set of above-mentioned Wei Nuotu, and by each section in the destination node set The line segment that point is connected in turn, the artery center line as above-mentioned virtual parent artery, wherein the destination node set Maximum space gradient direction by above-mentioned third analog wave from endpoint node along arrival time propagates all node groups passed through At set.
In the optional embodiment, each node in above-mentioned destination node set is sequentially connected by risk assessment terminal The line segment to get up, as the artery center line of above-mentioned virtual parent artery, specifically, virtual bracket implantation terminal determines mesh The node of inflows entry zone from virtual parent artery recently is marked in node set as start node, and from the starting The line segment that each node in destination node set is successively connected in turn by node is as above-mentioned virtual parent artery Artery center line.
In the optional embodiment, further optionally, when predetermined central line pick-up algorithm is above-mentioned base When topological Thinning Algorithm, risk assessment terminal is based on predetermined central line pick-up algorithm and above-mentioned threedimensional model extracts The artery center line of above-mentioned virtual parent artery may include:
Risk assessment is based on topological thinning method eventually and executes morphological erosion operation to above-mentioned virtual parent artery, until this The topological structure of virtual parent artery remains unchanged, and will be on topological structure from the starting point on the topological structure remained unchanged Artery center line of the line segment that each point is connected in turn as virtual parent artery.
In the optional embodiment, the starting point of topological structure be inflow entry zone apart from virtual parent artery most A close point.
In the optional embodiment, further optionally, when predetermined central line pick-up algorithm is above-mentioned base When Hessian tracing algorithm, risk assessment terminal is based on predetermined central line pick-up algorithm and above-mentioned threedimensional model The artery center line for extracting above-mentioned virtual parent artery may include:
Risk assessment terminal calculates the Hessian matrix of above-mentioned threedimensional model based on the method for tracing of Hessian matrix, obtains To the feature vector of the threedimensional model, and determine axis direction of the direction of this feature vector as virtual parent artery；
Risk assessment terminal obtains the local feature point set of above-mentioned virtual parent artery, and determines the local feature point set Each local feature region obtains the center of the local feature point set perpendicular to the central point in the section of above-mentioned axis direction in conjunction Point set, and each central point connection in the point set of center is successively formed by curve as virtual from starting central point The artery center line of parent artery.
In the optional embodiment, local feature region may include spot and/or angle on above-mentioned virtual parent artery Point, wherein spot may include the pixel that gray value is higher than default gray value in above-mentioned virtual parent artery.Angle point can wrap The branch for including the corner point of above-mentioned virtual parent artery and/or the aorta of above-mentioned virtual parent artery and the virtual parent artery is dynamic Arteries and veins is formed by a little.Originating central point is the point of inflow entry zone recently apart from above-mentioned virtual parent artery.
As it can be seen that passing through the extracting method of multiple intra-arterial center line, the extracting method of virtual parent artery can not only be enriched, Suitable artery center line extraction method can also be selected according to the actual conditions of virtual parent artery, to be conducive to quickly really Determine Morphologic Parameters and hemodynamic parameter.
Further alternative in the optional embodiment, risk assessment terminal is based on predetermined region and distinguishes calculation Method carries out region differentiation to above-mentioned threedimensional model, obtains the arterial inlet region of above-mentioned virtual parent artery, above-mentioned virtual load tumor The artery exit region of artery may include:
Risk assessment terminal determines the certain point on the inflow artery (also referred to as inflow blood vessel) of virtual parent artery as the Certain point on one target point, and the outflow artery (also referred to as outflow blood vessel) of determining virtual parent artery is as the second target Point；
Risk assessment terminal determines the first nodal point on its corresponding artery center line, Yi Jiji based on first object point The second central point on its corresponding artery center line is determined respectively in the second target point；
Risk assessment terminal cuts the inflow artery of virtual parent artery based on first object point and first nodal point, obtains The arterial inlet region of virtual parent artery and risk assessment terminal are based on the second target point and the cutting of the second central point is virtual The outflow artery of parent artery obtains the artery exit region of virtual parent artery.
In the optional embodiment, first nodal point is that the connecting line segment length on first object point and center line is minimum Point, same second central point is the smallest point of connecting line segment length on the second target point and center line.First object point with And second target point be also possible to medical staff selection point, the embodiment of the present invention is without limitation.The artery being cut into this way Entrance area and artery exit region are it is advantageously ensured that correct blood flow direction in calculation of boundary conditions, to help to calculate The hemodynamic parameter of virtual parent artery.
In the embodiment of the present invention, algorithm is distinguished based on predetermined region, region differentiation is carried out to above-mentioned threedimensional model, Threedimensional model after being divided can be as shown in figure 3, Fig. 3 be virtual load tumor after a kind of cutting disclosed by the embodiments of the present invention The structural schematic diagram of the threedimensional model of artery.
Further alternative in the optional embodiment, risk assessment terminal is based on predetermined region and distinguishes calculation Method to above-mentioned threedimensional model carry out region differentiation, obtain above-mentioned virtual parent artery artery wall region and above-mentioned virtual artery The aneurysm wall region of tumor may include:
Risk assessment terminal goes out to select two o'clock in the intersection of virtual aneurysm and virtual parent artery, connects this two o'clock A line segment is formed, then the line segment is scanned the face of being formed by as void along the normal orientation (positive and negative) of the view plane of threedimensional model Intend aneurysmal tumor neck plane (the tumor neck plane in referring to figure 4.), and distinguishes virtual carry using the tumor neck plane as line of demarcation Tumor artery and virtual aneurysm, obtain virtual parent artery artery wall region and virtual aneurysmal aneurysm wall region. As shown in figure 4, Fig. 4 is a kind of threedimensional model of the virtual parent artery using tumor neck plane as line of demarcation disclosed by the invention Structural schematic diagram.
103, risk assessment terminal determines target hemodynamic parameter based on above-mentioned threedimensional model, the target hemodynamic Learn the hemodynamic parameter and above-mentioned virtual aneurysmal hemodynamic parameter that parameter includes above-mentioned virtual parent artery.
In the embodiment of the present invention, step 102 and step 103 can occur simultaneously, and the embodiment of the present invention is without limitation.
In the embodiment of the present invention, hemodynamic parameter includes but is not limited to that the wall shear stress of virtual parent artery is average Value WSS-mean, the wall shear stress maximum value WSS-max of virtual parent artery, the wall shear stress of virtual parent artery are minimum The face in the wall surface high shear stress area of value WSS-min, the area in the wall surface Low shear stress area of virtual parent artery, virtual parent artery Product, the concussion coefficient OSI of the wall shear stress, energy-loss factor EL, pressure drop coefficient, shearing force spatial gradient, shearing Power time gradient.Wherein, wall surface Low shear stress area is region of the wall shear stress average value WSS-mean less than 12.8 μ Q/d3 Area and virtual aneurysm wall surface area ratio, wall surface high shear stress area are that wall shear stress average value WSS-mean is greater than 64 μ The area in the region of Q/d3 and virtual aneurysm wall surface area ratio.
In the embodiment of the present invention, risk assessment terminal determines target hemodynamic parameter based on above-mentioned threedimensional model, can To include:
Risk assessment terminal is based on predetermined reseau-dividing algorithm and divides above-mentioned threedimensional model, obtains multiple objective bodies Grid model, which is polyhedron network lattice model；
Risk assessment terminal determines the object boundary condition of above-mentioned threedimensional model, which includes at least above-mentioned The perimeter strip of the boundary condition on the arterial inlet boundary of virtual parent artery, the artery outlet border of above-mentioned virtual parent artery Part；
Risk assessment terminal is based on all above-mentioned objective body grid models and the above-mentioned above-mentioned void of object boundary condition simulation The blood flow of quasi- parent artery and above-mentioned virtual aneurysmal blood flow, obtain target hemodynamic parameter.
In the embodiment of the present invention, above-mentioned threedimensional model is divided based on predetermined reseau-dividing algorithm, obtains multiple mesh Standard type grid model can with as shown in figure 4, Fig. 4 be the virtual parent artery after a kind of grid dividing disclosed by the invention three-dimensional The structural schematic diagram of model.
In the embodiment of the present invention, risk assessment terminal is based on all above-mentioned objective body grid models and above-mentioned object boundary The blood flow of the above-mentioned virtual parent artery of condition simulation and above-mentioned virtual aneurysmal blood flow, obtain target blood flow Mechanics parameter.Specifically, to meet three dimensional fluid motion control equation as follows for blood flow:
Wherein, equation (1) is fluid mass conservation equation, and equation (2) is fluid momentum conservation equation.And in formula: its Middle ρ represents density of blood, and v represents the kinematic viscosity of blood, and u is the speed of blood, and p is blood pressure.
In the optional embodiment, fluid motion governing equation is one group of partial differential equation, and specific method for solving can To include one of finite element, finite difference calculus, finite volume method or multiple combinations, the embodiment of the present invention is without limitation.
104, risk assessment terminal is based on preparatory trained machine learning model to above-mentioned target morphology parameter, above-mentioned Target hemodynamic parameter and target clinical parameter carry out operation, obtain above-mentioned virtual aneurysmal assessment result, this is commented Result is estimated for assessing the aneurysmal risk of rupture.
In the embodiment of the present invention, target clinical parameter includes predefining facing for the corresponding user of encephalic image data Bed parameter.Wherein, whether clinical parameter includes but is not limited to blood pressure, blood lipid, the age, gender, passing medical history, family history, carries Whether tumor susceptibility gene is multiple intracranial aneurysm and intracranial aneurysm position.
As another optional embodiment, risk assessment terminal is based on preparatory trained machine learning model to above-mentioned Target morphology parameter, above-mentioned target hemodynamic parameter and target clinical parameter carry out operation, obtain above-mentioned virtual dynamic Before the assessment result of arteries and veins tumor, can also include:
Risk assessment terminal constructs machine learning model, and the machine learning model is for determining above-mentioned virtual aneurysmal comment Estimate result.
Wherein, risk assessment terminal constructs machine learning model, may include:
Risk assessment terminal obtains multiple sample encephalic image datas, the corresponding artery of each sample encephalic image data Tumor patient；
Risk assessment terminal determines the sample characteristics parameter of each sample encephalic image data, and each sample characteristics parameter can By include sample Morphologic Parameters, sample hemodynam ics param eter corresponding with the sample Morphologic Parameters and with the sample in the form of Learn the corresponding sample clinical parameter of parameter；
The each sample characteristics parameter of risk assessment terminal label, after the sample characteristics parameter and the label after being marked The corresponding mark value of sample characteristics parameter；
Risk assessment terminal obtains machine based on the sample characteristics parameter after each label of predetermined training pattern training Device learning model.
In the optional embodiment, each sample characteristics parameter of risk assessment terminal label, the sample spy after being marked The corresponding mark value of parameter is levied, may include:
When the aneurysm that above-mentioned sample encephalic image data includes is ruptured aneurysm, then mark and the sample encephalic shadow As the mark value of the corresponding sample characteristics parameter of data is the first preset value；
When the aneurysm that above-mentioned sample encephalic image data includes is non-ruptured aneurysm, then mark and the sample encephalic The mark value of the corresponding sample characteristics parameter of image data is the second preset value；
In the optional embodiment, above-mentioned first preset value and above-mentioned second preset value be not identical.Specifically, when above-mentioned the When one preset value is 1, then the second preset value is 0；When above-mentioned first preset value is 0, then the second preset value is 1, and the present invention is real Apply example without limitation.
In the optional embodiment, risk assessment terminal is based on each sample characteristics ginseng of predetermined training pattern training Number, obtains machine learning model, may include:
Risk assessment terminal divides the sample characteristics parameter after all labels, obtain training set, verifying collection and Test set, and based on the sample characteristics parameter after the above-mentioned label as training set of predetermined training pattern training, it obtains First sub- machine learning model, and using as the sub- machine learning of sample characteristics Verification first after the label of verifying collection Model obtains the second sub- machine learning model, finally based on the sample characteristics parameter testing second after the label as test set Sub- machine learning model, obtains machine learning model.
In the optional embodiment, predetermined training pattern may include Logic Regression Models, decision-tree model, shellfish Leaf this model, k- nearest neighbor algorithm model, Random Forest model, supporting vector machine model, neural network model, Adaboost mould One of type, GradientBoost model model or the combination of a variety of models, the embodiment of the present invention is without limitation.
As it can be seen that the optional embodiment passes through the engineering for pre-establishing and being used on and determining virtual aneurysmal assessment result Model is practised, can directly utilize the machine after the Morphologic Parameters, hemodynamic parameter and clinical parameter for determining patient Learning model carries out operation, to quickly determine the aneurysmal risk of rupture of patient, and then further quickly provides for patient Effective therapeutic scheme.
As another optional embodiment, risk assessment terminal is based on preparatory trained machine learning model to above-mentioned Target morphology parameter, above-mentioned target hemodynamic parameter and target clinical parameter carry out operation, obtain above-mentioned virtual dynamic It, can be with after the assessment result of arteries and veins tumor:
The assessment report of above-mentioned encephalic image data is generated according to above-mentioned assessment result, which includes above-mentioned target The haemodynamics risk analysis knot of the morphology risk analysis results of Morphologic Parameters, above-mentioned target hemodynamic parameter Fruit, the clinical risk analysis result of above-mentioned target clinical parameter and integrated risk analysis are as a result, the integrated risk analysis result To analyze the knot that result generates based on morphology risk analysis result, haemodynamics risk analysis result and clinical risk Fruit；
The corresponding risk of integrated risk analysis result that above-mentioned assessment report includes is determined based on default risk class rule Grade, and show above-mentioned assessment report and the corresponding risk class of integrated risk analysis result.
In the optional embodiment, which is the integrated risk analysis result according to intracranial aneurysm The risk class rule that is configured of slip gradient and default risk class may include several grades, such as: 1 grade -5 Grade or 1 grade -10 grades, such as: slip gradient is that the corresponding risk class of 0%-20% is 1 grade, slip gradient 80%100% It is the corresponding risk class of 0%-10% is 1 grade that corresponding risk class, which is 5 grades or slip gradient, slip gradient 90%- 100% corresponding risk class is 10 grades, and the embodiment of the present invention is without limitation.Specifically, rupture of intracranial aneurysm risk is higher Corresponding risk class is higher or the more high corresponding risk class of rupture of intracranial aneurysm risk is lower.
In the optional embodiment, further, the analysis in risk assessment terminal label assessment report is as a result, marked Analysis after note is as a result, and according to the risk of the different determining aneurysm ruptures of label.Specifically, will analysis result with different Color is marked, and color is deeper, and the risk for representing aneurysm rupture is higher, and the embodiment of the present invention is without limitation.Pass through in this way The analysis result of assessment report is come out with color and/or graphic context label, and by the risk of rupture of intracranial aneurysm to rupture The case where risk class is marked, and can intuitively understand aneurysm rupture.
In the optional embodiment, further, risk assessment terminal is based on above-mentioned target morphology parameter, above-mentioned mesh Mark the determining and above-mentioned assessment from the aneurysm database pre-established of hemodynamic parameter and above-mentioned target clinical parameter It reports most like goal-based assessment report, and shows that the goal-based assessment is reported, the aneurysm database pre-established is for depositing Store up the assessment report of each aneurysm patient in all arteries tumor patient.
In the optional embodiment, the determining and above-mentioned assessment from the aneurysm database pre-established of risk assessment terminal Report most like goal-based assessment report, i.e. risk assessment terminal determining and current disease from the aneurysm database pre-established The most like previous case of example, wherein being somebody's turn to do the previous case most like with current case is the target morphology according to current case Learn what parameter, target hemodynamic parameter and target clinical parameter were screened as reference from aneurysm database.And And the analysis of most like previous case may include the clinical information (such as: 50 years old) of the previous case, three with current case Tie up vascular pattern and similarity degree (such as: at least one of 80%), the embodiment of the present invention is without limitation.Further, The clinical information is the information after anonymization.In this way by by the previous case most like with current case analysis shows that commenting Estimate in report, reference can be provided for medical staff, to quickly make diagnostic result for current aneurysm patient.
In the optional embodiment, assessment report is sent to the use of the corresponding patient of the assessment report by risk assessment terminal Family terminal, is checked for patient.Assessment report is sent to patient in this way, saves assessment report convenient for patient, and real time inspection is commented Estimate report.
As it can be seen that the optional embodiment includes Morphologic Parameters assessment result, hemodynamic parameter assessment by generating As a result and the assessment report of clinical parameter assessment result and comprehensive eye exam result, can obtain more acurrate, more fully Aneurysmal assessment result.
As it can be seen that the appraisal procedure for implementing rupture of intracranial aneurysm risk described in Fig. 1 can be by dynamic to patient's encephalic Morphologic Parameters, hemodynamic parameter and the clinical parameter of arteries and veins tumor carry out automatic point comprehensively to rupture of intracranial aneurysm risk Analysis, can be improved the accuracy rate of aneurysm rupture risk assessment, to provide scientific ginseng when making Treatment decsion for doctor It examines, so that doctor quickly provides effective therapeutic scheme to patient.
Embodiment two
Referring to Fig. 5, Fig. 5 is a kind of knot of the assessment system of rupture of intracranial aneurysm risk disclosed by the embodiments of the present invention Structure schematic diagram.As shown in figure 5, the assessment system of the rupture of intracranial aneurysm risk may include three-dimensional reconstruction module 501, form Learn measurement module 502, haemodynamics computing module 503 and integrated risk computing module 504, in which:
Three-dimensional reconstruction module 501 includes that target parent artery and the target carry tumor for being established according to encephalic image data Eparterial aneurysmal threedimensional model, the threedimensional model include the virtual parent artery to match with target parent artery and The virtual aneurysm to match with aneurysm.
Morphological measurement module 502, for determining target morphology parameter, the target morphology based on above-mentioned threedimensional model Parameter includes the Morphologic Parameters and above-mentioned virtual aneurysmal Morphologic Parameters of above-mentioned virtual parent artery.
Haemodynamics computing module 503, for determining target hemodynamic parameter based on above-mentioned threedimensional model, the mesh Mark hemodynamic parameter includes the hemodynamic parameter and above-mentioned virtual aneurysmal blood flow of above-mentioned virtual parent artery Kinetic parameter.
Integrated risk computing module 504, for being based on preparatory trained machine learning model to above-mentioned target morphology Parameter, above-mentioned target hemodynamic parameter and target clinical parameter carry out operation, obtain above-mentioned virtual aneurysmal assessment As a result, the assessment result, for assessing aneurysmal risk of rupture, which is predetermined encephalic image number According to the clinical parameter of corresponding user.
As it can be seen that the assessment system for implementing rupture of intracranial aneurysm risk described in Fig. 5 can be by dynamic to patient's encephalic Morphologic Parameters, hemodynamic parameter and the clinical parameter of arteries and veins tumor carry out automatic point comprehensively to rupture of intracranial aneurysm risk Analysis, can be improved the accuracy rate of aneurysm rupture risk assessment, to provide scientific ginseng when making Treatment decsion for doctor It examines, so that doctor quickly provides effective therapeutic scheme to patient.
As a kind of optional embodiment, as shown in figure 5, integrated risk computing module 504, is also used to based on instruction in advance The machine learning model perfected is to above-mentioned target morphology parameter, above-mentioned target hemodynamic parameter and target clinical parameter Operation is carried out, after obtaining above-mentioned virtual aneurysmal assessment result, above-mentioned encephalic image data is generated according to the assessment result Assessment report, which includes the morphology risk analysis result of above-mentioned target morphology parameter, above-mentioned target blood flow The haemodynamics risk analysis result of kinetic parameter, the clinical risk analysis result of above-mentioned target clinical parameter and synthesis Risk analysis is as a result, the integrated risk analysis result is based on above-mentioned morphology risk analysis result, above-mentioned hemodynamic style of study The result that danger analysis result and clinical risk analysis result generate；
Integrated risk computing module 504 is also used to determine the synthesis that above-mentioned assessment report includes based on predetermined level rule The corresponding risk class of risk analysis result, and show the assessment report and the corresponding risk of integrated risk analysis result etc. Grade.
As it can be seen that the assessment system for implementing rupture of intracranial aneurysm risk described in Fig. 5 can also be by point of assessment report Analysis result is come out with color and/or graphic context label, and the risk of rupture of intracranial aneurysm is gone out with risk of rupture grade mark The case where coming, aneurysm rupture can be intuitively understood.
And integrated risk computing module 504, it is also used to based on above-mentioned target morphology parameter, above-mentioned target blood flow Mechanics parameter and above-mentioned target clinical parameter determine and above-mentioned assessment report most phase from the aneurysm database pre-established As goal-based assessment report that and show that the goal-based assessment is reported, the aneurysm database pre-established is all dynamic for storing The assessment report of each aneurysm patient in arteries and veins tumor patient.
As it can be seen that the assessment system for implementing rupture of intracranial aneurysm risk described in Fig. 5 can also be by will be with current disease The most like previous case of example analysis shows that in assessment report, reference can be provided for medical staff, to quickly be to work as Prerolandic artery Rolando tumor patient's makes diagnostic result.
As an alternative embodiment, Morphological measurement module 502 may include model segmentation submodule 5021 with And morphology determines submodule 5022, at this point, the structural schematic diagram of the assessment system of rupture of intracranial aneurysm risk can be such as Fig. 6 Shown, Fig. 6 is the structural schematic diagram of the assessment system of another rupture of intracranial aneurysm risk disclosed by the embodiments of the present invention, In:
Model divides submodule 5021, for dividing above-mentioned threedimensional model based on predetermined model partitioning algorithm, obtains Threedimensional model after to segmentation.
Morphology determines submodule 5022, for determining target morphology parameter based on the threedimensional model after above-mentioned segmentation.
Wherein, model segmentation submodule 5021 is based on the above-mentioned threedimensional model of predetermined model partitioning algorithm segmentation, obtains The mode of threedimensional model after to segmentation specifically:
The a certain pixel position on above-mentioned virtual parent artery is determined based on predetermined model partitioning algorithm As the first wave source point of the first analog wave, and determine a certain pixel position in above-mentioned virtual aneurysm as the Second wave source point of two analog waves, first analog wave and second analog wave are same type of analog wave；
Emit above-mentioned first analog wave and above-mentioned second analog wave simultaneously, and records first analog wave and second mould The propagation duration of quasi- wave should at the time of the starting of the propagation duration propagates the moment to emit the first analog wave and the second analog wave At the time of terminating the wave crest overlapping for the first time for propagating wave crest and the second analog wave that the moment is the first analog wave of propagation duration；
Determine that above-mentioned first analog wave propagates covered region in above-mentioned propagation duration and above-mentioned second analog wave passes Covered the sum of region is broadcast, as cut zone corresponding with above-mentioned virtual parent artery and above-mentioned virtual aneurysm, And the threedimensional model is divided according to the cut zone, the threedimensional model after being divided.
As an alternative embodiment, as is seen in fig. 6 or fig. 7, the assessment system of rupture of intracranial aneurysm risk is also It may include quality of image judgment module 505 and Yunnan snub-nosed monkey module 506, in which:
Quality of image judgment module 505, for being based on encephalic image data to establish including carrying in above-mentioned three-dimensional reconstruction module On tumor artery and the parent artery before aneurysmal threedimensional model, judge whether the quality of image index of encephalic image data reaches To default quality of image index；
Three-dimensional reconstruction module 501, is specifically used for:
When quality of image judgment module 505 judges that the quality of image index of encephalic image data reaches the default quality of image When index, the operation including aneurysmal threedimensional model on parent artery and the parent artery is established according to encephalic image data.
Yunnan snub-nosed monkey module 506, for judging the image matter of encephalic image data when quality of image judgment module 505 When figureofmerit not up to presets quality of image index, encephalic image data is pre-processed based on default image algorithm, until The quality of image index of the encephalic image data reaches default quality of image index, and triggers three-dimensional reconstruction module 501 and execute That states establishes the operation including aneurysmal threedimensional model on parent artery and the parent artery according to encephalic image data.
As it can be seen that implementing the assessment system of rupture of intracranial aneurysm risk described in Fig. 6 also by establishing parent artery And before the aneurysmal threedimensional model on parent artery, judge whether the quality of image index of encephalic image data reaches pre- Bidding is quasi-, and when not reaching preset standard, the quality of image index of encephalic image data is pre-processed, until reaching pre- If quality of image index just establishes the aneurysmal threedimensional model on parent artery and parent artery, can reduce because of image matter Figureofmerit is not up to standard and leads to a possibility that establishing parent artery and not accurate aneurysmal threedimensional model.
As an alternative embodiment, as shown in fig. 7, Morphological measurement module 502 includes central line pick-up submodule Submodule 5024 and morphology computational submodule 5025 are distinguished in block 5023, region, in which:
Central line pick-up submodule 5023, for being based on predetermined central line pick-up algorithm and above-mentioned threedimensional model Extract the artery center line of above-mentioned virtual parent artery.
Submodule 5024 is distinguished in region, carries out area to above-mentioned threedimensional model for distinguishing algorithm based on predetermined region Domain distinguish, obtain target area, the target area include above-mentioned virtual parent artery arterial inlet region, this virtually carries tumor move The artery exit region of arteries and veins, the artery wall region of the virtual parent artery and above-mentioned virtual aneurysmal aneurysm wall region.
Morphology computational submodule 5025, for being wrapped based on predetermined Morphologic Parameters algorithm to above-mentioned target area The content included carries out computational geometry analysis, obtains target morphology parameter.
In the optional embodiment, it is based in advance really as an alternative embodiment, submodule 1024 is distinguished in region Fixed region distinguishes algorithm and carries out region differentiation to above-mentioned threedimensional model, obtains the mode of target area specifically:
Algorithm, which is distinguished, based on predetermined region determines that the arterial inlet cross section of above-mentioned virtual parent artery, artery go out Mouth cross section and above-mentioned virtual aneurysmal tumor neck plane；
According to the above-mentioned above-mentioned threedimensional model of arterial inlet cross-sectional cut, the arterial inlet of above-mentioned virtual parent artery is obtained Region, and the threedimensional model is cut according to above-mentioned exit cross-section, obtain the artery outlet area of above-mentioned virtual parent artery Domain, and according to the above-mentioned above-mentioned threedimensional model of tumor neck plane cutting, obtain above-mentioned virtual parent artery artery wall region and Above-mentioned virtual aneurysmal aneurysm wall region.
As an alternative embodiment, haemodynamics computing module 503 determines target based on the threedimensional model The mode of hemodynamic parameter specifically:
Above-mentioned threedimensional model is divided based on predetermined reseau-dividing algorithm, obtains multiple objective body grid models, institute Having objective body grid model is polyhedron network lattice model；
Determine the object boundary condition of above-mentioned threedimensional model, which includes at least above-mentioned virtual parent artery The boundary condition on arterial inlet boundary, the virtual parent artery artery outlet border boundary condition；
Based on all above-mentioned objective body grid models and the above-mentioned virtual parent artery of above-mentioned object boundary condition simulation Blood flow and above-mentioned virtual aneurysmal blood flow, obtain target hemodynamic parameter.
As an alternative embodiment, as is seen in fig. 6 or fig. 7, the assessment system of the rupture of intracranial aneurysm risk It can also include model construction module 507, in which:
Model construction module 507, for being based on preparatory trained machine learning model in integrated risk computing module 504 Operation is carried out to above-mentioned target morphology parameter, above-mentioned target hemodynamic parameter and target clinical parameter, is obtained above-mentioned Before virtual aneurysmal assessment result, machine learning model is constructed, the machine learning model is for determining above-mentioned virtual artery The assessment result of tumor；
Wherein, model construction module 507 constructs the mode of machine learning model specifically:
Obtain multiple sample encephalic image datas, the corresponding aneurysm patient of each sample encephalic image data；
Determine that the sample characteristics parameter of each sample encephalic image data, each sample characteristics parameter include sample morphology The parameter and corresponding sample hemodynam ics param eter of the sample Morphologic Parameters and sample corresponding with the sample Morphologic Parameters Clinical parameter；
Each sample characteristics parameter is marked, the sample characteristics ginseng after the sample characteristics parameter and the label after being marked The corresponding mark value of number；
Based on the sample characteristics parameter after each label of predetermined training algorithm training, machine learning model is obtained；
Wherein, model construction module 507 marks each sample characteristics parameter, and the sample characteristics parameter after being marked is corresponding Mark value mode specifically:
When the aneurysm that above-mentioned sample encephalic image data includes is ruptured aneurysm, then label and the sample encephalic image The mark value of the corresponding sample characteristics parameter of data is the first preset value；
When the aneurysm that above-mentioned sample encephalic image data includes is non-ruptured aneurysm, then label and the sample encephalic shadow As the mark value of the corresponding sample characteristics parameter of data is the second preset value；
Wherein, first preset value and second preset value be not identical.
In the optional embodiment, when model construction module 507 has executed the operation of above-mentioned building machine learning model Later, can trigger integrated risk computing module 504 execute it is above-mentioned based on preparatory trained machine learning model to above-mentioned Target morphology parameter, above-mentioned target hemodynamic parameter and target clinical parameter carry out operation, obtain above-mentioned virtual dynamic The operation of the assessment result of arteries and veins tumor.
It is pre-established as it can be seen that the assessment system for implementing rupture of intracranial aneurysm risk described in Fig. 6 or Fig. 7 also passes through For determining the machine learning model of virtual aneurysmal assessment result, can determine patient Morphologic Parameters, blood flow After mechanics parameter and clinical parameter, operation directly is carried out using the machine learning model, to quickly determine the artery of patient The risk of rupture of tumor, and then further effective therapeutic scheme quickly is provided for patient.
Embodiment three
Referring to Fig. 8, Fig. 8 is the assessment system of another rupture of intracranial aneurysm risk disclosed by the embodiments of the present invention Structural schematic diagram.As shown in figure 8, the assessment system of the rupture of intracranial aneurysm risk may include:
It is stored with the memory 701 of executable program code；
The processor 702 coupled with memory 701；
Processor 702 calls the executable program code stored in memory 701, executes cranium described in embodiment one Step in the appraisal procedure of internal aneurysm risk of rupture.
Example IV
The embodiment of the invention discloses a kind of computer readable storage medium, storage is used for the calculating of electronic data interchange Machine program, wherein the computer program makes computer execute rupture of intracranial aneurysm risk described in embodiment one Step in appraisal procedure.
Embodiment five
The embodiment of the invention discloses a kind of computer program product, which includes storing computer The non-transient computer readable storage medium of program, and the computer program is operable to execute computer in embodiment one Step in the appraisal procedure of described rupture of intracranial aneurysm risk.
Installation practice described above is only illustrative, wherein the module as illustrated by the separation member can be with It is or may not be and be physically separated, the component shown as module may or may not be physical module, Can be in one place, or may be distributed on multiple network modules.It can select according to the actual needs wherein Some or all of the modules achieve the purpose of the solution of this embodiment.Those of ordinary skill in the art are not paying creativeness In the case where labour, it can understand and implement.
By the specific descriptions of above embodiment, those skilled in the art can be understood that each embodiment It can realize by means of software and necessary general hardware platform, naturally it is also possible to pass through hardware.Based on this understanding, Substantially the part that contributes to existing technology can be embodied in the form of software products above-mentioned technical proposal in other words, The computer software product may be stored in a computer readable storage medium, and storage medium includes read-only memory (Read- Only Memory, ROM), random access memory (Random Access Memory, RAM), programmable read only memory (Programmable Read-only Memory, PROM), Erasable Programmable Read Only Memory EPROM (Erasable Programmable Read Only Memory, EPROM), disposable programmable read-only memory (One-time Programmable Read-Only Memory, OTPROM), the electronics formula of erasing can make carbon copies read-only memory (Electrically-Erasable Programmable Read-Only Memory, EEPROM), CD-ROM (Compact Disc Read-Only Memory, CD-ROM) or other disc memories, magnetic disk storage, magnetic tape storage or can For carrying or any other computer-readable medium of storing data.
Finally, it should be noted that a kind of appraisal procedure of rupture of intracranial aneurysm risk disclosed by the embodiments of the present invention and being The disclosed only present pre-ferred embodiments of system, are only used to illustrate the technical scheme of the present invention, rather than its limitations；To the greatest extent Present invention has been described in detail with reference to the aforementioned embodiments for pipe, those skilled in the art should understand that；It is still It can modify to technical solution documented by aforementioned every embodiment, or part of technical characteristic is equally replaced It changes；And these modifications or substitutions, so that the essence of corresponding technical solution is detached from the every embodiment technical solution of the present invention Spirit and scope.

Claims (10)

1. a kind of appraisal procedure of rupture of intracranial aneurysm risk, which is characterized in that the described method includes:
According to encephalic image data establish include parent artery and the parent artery on aneurysmal threedimensional model, described three Dimension module includes the virtual parent artery to match with the parent artery and the virtual artery that matches with the aneurysm Tumor；
Determine that target morphology parameter, the target morphology parameter include the virtual parent artery based on the threedimensional model Morphologic Parameters and the virtual aneurysmal Morphologic Parameters；
Determine that target hemodynamic parameter, the target hemodynamic parameter include described virtual based on the threedimensional model The hemodynamic parameter of parent artery and the virtual aneurysmal hemodynamic parameter；
Based on preparatory trained machine learning model to the target morphology parameter, the target hemodynamic parameter with And target clinical parameter carries out operation, obtains the virtual aneurysmal assessment result, the assessment result is described for assessing Aneurysmal risk of rupture, the target clinical parameter include predefining the corresponding user of the encephalic image data of going out Clinical parameter.
2. the appraisal procedure of rupture of intracranial aneurysm risk according to claim 1, which is characterized in that described based on described Threedimensional model determines target morphology parameter, comprising:
Divide the threedimensional model based on predetermined model partitioning algorithm, the threedimensional model after being divided, and it is based on institute Threedimensional model after stating segmentation determines target morphology parameter；
Wherein, described based on the predetermined model partitioning algorithm segmentation threedimensional model, the threedimensional model after being divided, Include:
The a certain pixel position conduct on the virtual parent artery is determined based on predetermined model partitioning algorithm A certain pixel position on the first wave source point of first analog wave, and the determining virtual aneurysm is as the second mould Second wave source point of quasi- wave, first analog wave and second analog wave are same type of analog wave；
Emit first analog wave and second analog wave simultaneously, and records first analog wave and second mould The propagation duration of quasi- wave, the starting for propagating duration propagate the moment to emit first analog wave and second analog wave At the time of, the wave crest of wave crest and second analog wave that the moment is first analog wave is propagated in the termination for propagating duration At the time of overlapping for the first time；
Determine that first analog wave propagates covered region in the propagation duration and second analog wave propagates institute The sum of region of covering, as cut zone corresponding with the virtual parent artery and the virtual aneurysm, and root Divide the threedimensional model according to the cut zone, the threedimensional model after being divided.
3. the appraisal procedure of rupture of intracranial aneurysm risk according to claim 1 or 2, which is characterized in that described to be based on Preparatory trained machine learning model faces the target morphology parameter, the target hemodynamic parameter and target Bed parameter carries out operation, after obtaining the virtual aneurysmal assessment result, the method also includes:
The assessment report of the encephalic image data is generated according to the assessment result, the assessment report includes the target shape The morphology risk analysis result of state parameter, the haemodynamics risk analysis result of the target hemodynamic parameter, The clinical risk analysis result and integrated risk analysis of the target clinical parameter are as a result, the integrated risk analysis result is Result is analyzed based on the morphology risk analysis result, the haemodynamics risk analysis result and the clinical risk The result of generation；
The corresponding risk of the integrated risk analysis result that the assessment report includes is determined based on default risk class rule Grade, and show the assessment report and the corresponding risk class of the integrated risk analysis result；
And the method also includes:
It is built based on the target morphology parameter, the target hemodynamic parameter and the target clinical parameter from advance The determining goal-based assessment most like with the assessment report is reported in vertical aneurysm database, and shows the goal-based assessment report It accuses, the aneurysm database pre-established is used to store the assessment of each aneurysm patient in all arteries tumor patient Report.
4. the appraisal procedure of rupture of intracranial aneurysm risk according to claim 1, which is characterized in that described based on described Threedimensional model determines target morphology parameter, comprising:
In the artery for extracting the virtual parent artery based on predetermined central line pick-up algorithm and the threedimensional model Heart line；
Algorithm is distinguished based on predetermined region, region differentiation is carried out to the threedimensional model, obtain target area, the mesh Mark region includes the artery exit region, described in the arterial inlet region of the virtual parent artery, the virtual parent artery The artery wall region of virtual parent artery and the virtual aneurysmal aneurysm wall region；
Computational geometry analysis is carried out to the content that the target area includes based on predetermined Morphologic Parameters algorithm, is obtained Target morphology parameter.
5. the appraisal procedure of rupture of intracranial aneurysm risk according to claim 4, which is characterized in that described based on preparatory Determining region distinguishes algorithm and carries out region differentiation to the threedimensional model, obtains target area, comprising:
Algorithm, which is distinguished, based on predetermined region determines that the arterial inlet cross section of the virtual parent artery, artery outlet are horizontal Section and the virtual aneurysmal tumor neck plane；
According to threedimensional model described in the arterial inlet cross-sectional cut, the arterial inlet area of the virtual parent artery is obtained Domain, and the threedimensional model is cut according to the exit cross-section, the artery exit region of the virtual parent artery is obtained, And the threedimensional model according to the tumor neck plane cutting, obtain the artery wall region of the virtual parent artery and described Virtual aneurysmal aneurysm wall region.
6. the appraisal procedure of rupture of intracranial aneurysm risk according to claim 5, which is characterized in that described based on described Threedimensional model determines target hemodynamic parameter, comprising:
The threedimensional model is divided based on predetermined reseau-dividing algorithm, obtains multiple objective body grid models, Suo Yousuo Stating objective body grid model is polyhedron network lattice model；
Determine that the object boundary condition of the threedimensional model, the object boundary condition include at least the virtual parent artery The boundary condition of the artery outlet border of the boundary condition on arterial inlet boundary, the virtual parent artery；
Blood based on virtual parent artery described in all objective body grid models and the object boundary condition simulation Flowing and the virtual aneurysmal blood flow, obtain target hemodynamic parameter.
7. the according to claim 1, appraisal procedure of rupture of intracranial aneurysm risk described in 2,5 or 6, which is characterized in that described Based on preparatory trained machine learning model to the target morphology parameter, the target hemodynamic parameter and mesh It marks clinical parameter and carries out operation, before obtaining the virtual aneurysmal assessment result, the method also includes:
Machine learning model is constructed, the machine learning model is for determining the virtual aneurysmal assessment result；
Wherein, the building machine learning model, comprising:
Obtain multiple sample encephalic image datas, each corresponding aneurysm patient of the sample encephalic image data；
Determine the sample characteristics parameter of each sample encephalic image data, each sample characteristics parameter includes sample shape State parameter, sample hemodynam ics param eter corresponding with the sample Morphologic Parameters and corresponding with the sample Morphologic Parameters Sample clinical parameter；
Each sample characteristics parameter is marked, the sample characteristics after sample characteristics parameter and the label after being marked The corresponding mark value of parameter；
Based on the sample characteristics parameter after each label of predetermined training algorithm training, machine learning model is obtained；
Wherein, each sample characteristics parameter of label, the corresponding mark value of sample characteristics parameter after being marked, packet It includes:
When the aneurysm that the sample encephalic image data includes is ruptured aneurysm, then label and the sample encephalic image data The mark value of corresponding sample characteristics parameter is the first preset value；
When the aneurysm that the sample encephalic image data includes is non-ruptured aneurysm, then label and the sample encephalic image number Mark value according to corresponding sample characteristics parameter is the second preset value；
Wherein, first preset value and second preset value be not identical.
8. a kind of assessment system of rupture of intracranial aneurysm risk, which is characterized in that the assessment system includes Three-dimensional Gravity modeling Block, Morphological measurement module, haemodynamics computing module and integrated risk computing module, in which:
The three-dimensional reconstruction module includes that target parent artery and the target carry tumor and move for being established according to encephalic image data Aneurysmal threedimensional model on arteries and veins, the threedimensional model include the virtual parent artery to match with the target parent artery And the virtual aneurysm to match with the aneurysm；
The Morphological measurement module, for determining target morphology parameter, the target morphology based on the threedimensional model Parameter includes the Morphologic Parameters and the virtual aneurysmal Morphologic Parameters of the virtual parent artery；
The haemodynamics computing module, for determining target hemodynamic parameter, the mesh based on the threedimensional model Mark hemodynamic parameter includes the hemodynamic parameter and the virtual aneurysmal blood flow of the virtual parent artery Kinetic parameter；
The integrated risk computing module, for being joined based on preparatory trained machine learning model to the target morphology Several, the described target hemodynamic parameter and target clinical parameter carry out operation, obtain the aneurysmal assessment result, institute It is the predetermined encephalic shadow that assessment result, which is stated, for assessing the aneurysmal risk of rupture, the target clinical parameter As the clinical parameter of the corresponding user of data.
9. a kind of assessment system of rupture of intracranial aneurysm risk, which is characterized in that described device includes:
It is stored with the memory of executable program code；
The processor coupled with the memory；
The processor calls the executable program code stored in the memory, executes as claim 1-7 is any The appraisal procedure of rupture of intracranial aneurysm risk described in.
10. a kind of computer can storage medium, which is characterized in that the computer storage medium is stored with computer instruction, institute State computer instruction it is called when, for executing commenting for such as described in any item rupture of intracranial aneurysm risks of claim 1-7 Estimate method.
CN201910280012.7A 2019-04-09 2019-04-09 Intracranial aneurysm rupture risk assessment method and system Active CN109907732B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201910280012.7A CN109907732B (en) 2019-04-09 2019-04-09 Intracranial aneurysm rupture risk assessment method and system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201910280012.7A CN109907732B (en) 2019-04-09 2019-04-09 Intracranial aneurysm rupture risk assessment method and system

Publications (2)

Publication Number Publication Date
CN109907732A true CN109907732A (en) 2019-06-21
CN109907732B CN109907732B (en) 2022-12-02

Family

ID=66969088

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201910280012.7A Active CN109907732B (en) 2019-04-09 2019-04-09 Intracranial aneurysm rupture risk assessment method and system

Country Status (1)

Country Link
CN (1) CN109907732B (en)

Cited By (28)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110353639A (en) * 2019-07-16 2019-10-22 脑玺(上海)智能科技有限公司 A kind of blood supply area quantitative approach and system based on blood vessel enhancing radiography
CN110517780A (en) * 2019-09-02 2019-11-29 强联智创（北京）科技有限公司 A kind of aneurysm rupture methods of risk assessment and system
CN110534193A (en) * 2019-09-02 2019-12-03 强联智创（北京）科技有限公司 A kind of aneurysm rupture methods of risk assessment and system
CN110706808A (en) * 2019-09-12 2020-01-17 北京深睿博联科技有限责任公司 Aneurysm rupture state prediction method and device
CN110739078A (en) * 2019-09-02 2020-01-31 北京市神经外科研究所 aneurysm rupture risk assessment method and system
CN110742689A (en) * 2019-10-31 2020-02-04 北京理工大学 Method and device for evaluating arterial dissection operation, electronic equipment and storage medium
CN110782988A (en) * 2019-11-04 2020-02-11 北京理工大学 Intracranial aneurysm virtual support diagnosis and treatment system and diagnosis and treatment method thereof
CN110866914A (en) * 2019-11-21 2020-03-06 北京冠生云医疗技术有限公司 Method, system, device and medium for evaluating cerebral aneurysm hemodynamic index
CN110974415A (en) * 2019-12-19 2020-04-10 向欣 Method for establishing three-dimensional contrast aneurysm form database of aneurysm volume
CN111081378A (en) * 2019-11-22 2020-04-28 强联智创（北京）科技有限公司 Aneurysm rupture risk assessment method and system
CN111462047A (en) * 2020-03-06 2020-07-28 深圳睿心智能医疗科技有限公司 Blood vessel parameter measuring method, blood vessel parameter measuring device, computer equipment and storage medium
CN111584077A (en) * 2020-04-08 2020-08-25 北京市神经外科研究所 Aneurysm rupture risk assessment method and system
CN111584076A (en) * 2020-04-08 2020-08-25 北京市神经外科研究所 Aneurysm rupture risk assessment method and system
CN111671556A (en) * 2020-04-14 2020-09-18 北京市神经外科研究所 Aneurysm stent and method of construction thereof
CN112365473A (en) * 2020-11-12 2021-02-12 同心医联科技（北京）有限公司 AI medical image processing system, equipment and storage medium
CN112617770A (en) * 2020-12-28 2021-04-09 首都医科大学附属北京天坛医院 Intracranial aneurysm risk prediction method based on artificial intelligence and related device
WO2021081835A1 (en) * 2019-10-30 2021-05-06 未艾医疗技术(深圳)有限公司 Aneurysm ai processing method based on vrds 4d medical image, and product
CN112826457A (en) * 2020-12-31 2021-05-25 苏州爱琴生物医疗电子有限公司 Brain-based sign information detection device and method and related equipment
CN113133827A (en) * 2021-04-07 2021-07-20 昆明同心医联科技有限公司 Preoperative prediction method, system, terminal and medium for intracranial aneurysm operation
CN113143306A (en) * 2021-03-31 2021-07-23 广东医科大学附属医院 Be used for intracranial aneurysm CFD diagnosis visual device
CN113143458A (en) * 2021-03-15 2021-07-23 北京理工大学 Method and device for evaluating aortic dissection distal laceration occlusion treatment scheme
CN113192004A (en) * 2021-04-01 2021-07-30 温州医科大学附属第二医院（温州医科大学附属育英儿童医院） Simplified calculation method based on coronary artery multimode image crack formation and expansion and plaque rupture risk quantitative evaluation system
CN114145843A (en) * 2021-11-05 2022-03-08 北京市神经外科研究所 Novel intracranial aneurysm weak area evaluation method
CN114287954A (en) * 2021-12-13 2022-04-08 杭州脉流科技有限公司 Functional assessment method, system, electronic device and medium for intracranial vascular stenosis
CN114972165A (en) * 2022-03-24 2022-08-30 中山大学孙逸仙纪念医院 Method and device for measuring time-average shearing force
CN116485800A (en) * 2023-06-26 2023-07-25 杭州脉流科技有限公司 Automatic acquisition method, device, equipment and storage medium for morphological parameters of aneurysms
CN116919374A (en) * 2023-07-19 2023-10-24 西安交通大学 Intracranial aneurysm and method and system for evaluating blood flow dynamics parameters in aneurysm-carrying artery
CN117438092A (en) * 2023-12-20 2024-01-23 杭州脉流科技有限公司 Intracranial aneurysm rupture risk prediction device, computer device, and storage medium

Citations (14)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7986822B2 (en) * 2006-04-24 2011-07-26 Siemens Medical Solutions Usa, Inc. System and method for X-ray based assessment of aneurysm pulsation
US8170304B2 (en) * 2007-04-03 2012-05-01 Siemens Aktiengesellschaft Modeling cerebral aneurysms in medical images
CN105096388A (en) * 2014-04-23 2015-11-25 北京冠生云医疗技术有限公司 Computational Fluid Dynamics (CFD) based coronary artery blood flow simulating system and method
CN106127819A (en) * 2016-06-30 2016-11-16 上海联影医疗科技有限公司 Medical image extracts method and the device thereof of vessel centerline
CN107174219A (en) * 2010-08-12 2017-09-19 哈特弗罗公司 The method and system modeled for patient-specific blood flow
US9770303B2 (en) * 2013-12-18 2017-09-26 Heartflow, Inc. Systems and methods for predicting coronary plaque vulnerability from patient-specific anatomic image data
JP2018000981A (en) * 2017-08-23 2018-01-11 東芝メディカルシステムズ株式会社 Aneurysm rupture risk analysis program and system
CN108133478A (en) * 2018-01-11 2018-06-08 苏州润心医疗器械有限公司 A kind of method for extracting central line of coronary artery vessel
EP3404667A1 (en) * 2017-05-19 2018-11-21 Siemens Healthcare GmbH Learning based methods for personalized assessment, long-term prediction and management of atherosclerosis
CN108876794A (en) * 2017-05-16 2018-11-23 西门子保健有限责任公司 Aneurysm in volumetric image data with carry being isolated for tumor blood vessel
WO2018220573A1 (en) * 2017-05-31 2018-12-06 Fondazione Ri.Med Method and system for the evaluation of the risk of aortic rupture or dissection in an individual with an ascending thoracic aortic aneurysm
CN109493348A (en) * 2018-10-26 2019-03-19 强联智创（北京）科技有限公司 A kind of measurement method and system of the Morphologic Parameters of intracranial aneurysm image
CN109584997A (en) * 2018-10-26 2019-04-05 首都医科大学宣武医院 A kind of measurement method and system of the Morphologic Parameters of intracranial aneurysm image
CN109961850A (en) * 2019-03-19 2019-07-02 肖仁德 A kind of method, apparatus, computer equipment for assessing rupture of intracranial aneurysm risk

Patent Citations (14)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7986822B2 (en) * 2006-04-24 2011-07-26 Siemens Medical Solutions Usa, Inc. System and method for X-ray based assessment of aneurysm pulsation
US8170304B2 (en) * 2007-04-03 2012-05-01 Siemens Aktiengesellschaft Modeling cerebral aneurysms in medical images
CN107174219A (en) * 2010-08-12 2017-09-19 哈特弗罗公司 The method and system modeled for patient-specific blood flow
US9770303B2 (en) * 2013-12-18 2017-09-26 Heartflow, Inc. Systems and methods for predicting coronary plaque vulnerability from patient-specific anatomic image data
CN105096388A (en) * 2014-04-23 2015-11-25 北京冠生云医疗技术有限公司 Computational Fluid Dynamics (CFD) based coronary artery blood flow simulating system and method
CN106127819A (en) * 2016-06-30 2016-11-16 上海联影医疗科技有限公司 Medical image extracts method and the device thereof of vessel centerline
CN108876794A (en) * 2017-05-16 2018-11-23 西门子保健有限责任公司 Aneurysm in volumetric image data with carry being isolated for tumor blood vessel
EP3404667A1 (en) * 2017-05-19 2018-11-21 Siemens Healthcare GmbH Learning based methods for personalized assessment, long-term prediction and management of atherosclerosis
WO2018220573A1 (en) * 2017-05-31 2018-12-06 Fondazione Ri.Med Method and system for the evaluation of the risk of aortic rupture or dissection in an individual with an ascending thoracic aortic aneurysm
JP2018000981A (en) * 2017-08-23 2018-01-11 東芝メディカルシステムズ株式会社 Aneurysm rupture risk analysis program and system
CN108133478A (en) * 2018-01-11 2018-06-08 苏州润心医疗器械有限公司 A kind of method for extracting central line of coronary artery vessel
CN109493348A (en) * 2018-10-26 2019-03-19 强联智创（北京）科技有限公司 A kind of measurement method and system of the Morphologic Parameters of intracranial aneurysm image
CN109584997A (en) * 2018-10-26 2019-04-05 首都医科大学宣武医院 A kind of measurement method and system of the Morphologic Parameters of intracranial aneurysm image
CN109961850A (en) * 2019-03-19 2019-07-02 肖仁德 A kind of method, apparatus, computer equipment for assessing rupture of intracranial aneurysm risk

Non-Patent Citations (9)

* Cited by examiner, † Cited by third party
Title
BEN I. H. VERWER: "《An Efficient Uniform Cost Algorithm Applied to Distance Transforms》", 《IEEE TRANS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE》 *
IGNACIO LARRABIDE: "《Three-dimensional morphological analysis of intracranial aneurysms: A fully automated method for aneurysm sac isolation and quantification》", 《MEDICAL PHYSICS》 *
J. EGGER: "《Fast self-collision detection and simulation of bifurcated stents to treat abdominal aortic aneurysms (AAA)》", 《CONFERENCE OF THE IEEE EMBS》 *
于红玉等: "《基于影像特征的颅内动脉瘤三维血流动力学模型建立方法》", 《中国医疗设备》 *
于红玉等: "《颅内动脉瘤 3D 几何模型建立方法的研究》", 《计算机工程与应用》 *
刘超博等: "颅内动脉瘤破裂风险和动脉瘤形态学参数的关系", 《中国临床神经外科杂志》 *
殷复莲: "《数据分析与数据挖掘实用教程》", 30 September 2017 *
殷宏等: "《虚拟现实技术及应用》", 31 July 2018 *
江国权等: "颅内动脉瘤破裂的血流动力学和形态学因素", 《介入放射学杂志》 *

Cited By (42)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110353639B (en) * 2019-07-16 2022-04-01 脑玺(上海)智能科技有限公司 Blood supply area quantification method and system based on blood vessel enhanced radiography
CN110353639A (en) * 2019-07-16 2019-10-22 脑玺(上海)智能科技有限公司 A kind of blood supply area quantitative approach and system based on blood vessel enhancing radiography
CN110739078B (en) * 2019-09-02 2024-03-26 北京市神经外科研究所 Aneurysm rupture risk assessment method and system
CN110517780A (en) * 2019-09-02 2019-11-29 强联智创（北京）科技有限公司 A kind of aneurysm rupture methods of risk assessment and system
CN110534193A (en) * 2019-09-02 2019-12-03 强联智创（北京）科技有限公司 A kind of aneurysm rupture methods of risk assessment and system
CN110739078A (en) * 2019-09-02 2020-01-31 北京市神经外科研究所 aneurysm rupture risk assessment method and system
CN110706808A (en) * 2019-09-12 2020-01-17 北京深睿博联科技有限责任公司 Aneurysm rupture state prediction method and device
WO2021081835A1 (en) * 2019-10-30 2021-05-06 未艾医疗技术(深圳)有限公司 Aneurysm ai processing method based on vrds 4d medical image, and product
CN110742689A (en) * 2019-10-31 2020-02-04 北京理工大学 Method and device for evaluating arterial dissection operation, electronic equipment and storage medium
CN110782988A (en) * 2019-11-04 2020-02-11 北京理工大学 Intracranial aneurysm virtual support diagnosis and treatment system and diagnosis and treatment method thereof
CN110782988B (en) * 2019-11-04 2022-11-01 北京理工大学 Intracranial aneurysm virtual support simulation method
CN110866914B (en) * 2019-11-21 2023-10-03 北京冠生云医疗技术有限公司 Evaluation method, system, equipment and medium for cerebral aneurysm hemodynamic index
CN110866914A (en) * 2019-11-21 2020-03-06 北京冠生云医疗技术有限公司 Method, system, device and medium for evaluating cerebral aneurysm hemodynamic index
WO2021098768A1 (en) * 2019-11-22 2021-05-27 强联智创（北京）科技有限公司 Method and system for assessing aneurysm rupture risk
CN111081378A (en) * 2019-11-22 2020-04-28 强联智创（北京）科技有限公司 Aneurysm rupture risk assessment method and system
CN111081378B (en) * 2019-11-22 2022-05-20 强联智创（北京）科技有限公司 Aneurysm rupture risk assessment method and system
CN110974415A (en) * 2019-12-19 2020-04-10 向欣 Method for establishing three-dimensional contrast aneurysm form database of aneurysm volume
CN111462047A (en) * 2020-03-06 2020-07-28 深圳睿心智能医疗科技有限公司 Blood vessel parameter measuring method, blood vessel parameter measuring device, computer equipment and storage medium
CN111462047B (en) * 2020-03-06 2024-03-12 深圳睿心智能医疗科技有限公司 Vascular parameter measurement method, vascular parameter measurement device, vascular parameter measurement computer device and vascular parameter measurement storage medium
CN111584076A (en) * 2020-04-08 2020-08-25 北京市神经外科研究所 Aneurysm rupture risk assessment method and system
CN111584077A (en) * 2020-04-08 2020-08-25 北京市神经外科研究所 Aneurysm rupture risk assessment method and system
CN111671556A (en) * 2020-04-14 2020-09-18 北京市神经外科研究所 Aneurysm stent and method of construction thereof
CN112365473A (en) * 2020-11-12 2021-02-12 同心医联科技（北京）有限公司 AI medical image processing system, equipment and storage medium
CN112617770A (en) * 2020-12-28 2021-04-09 首都医科大学附属北京天坛医院 Intracranial aneurysm risk prediction method based on artificial intelligence and related device
CN112826457A (en) * 2020-12-31 2021-05-25 苏州爱琴生物医疗电子有限公司 Brain-based sign information detection device and method and related equipment
CN113143458A (en) * 2021-03-15 2021-07-23 北京理工大学 Method and device for evaluating aortic dissection distal laceration occlusion treatment scheme
CN113143306A (en) * 2021-03-31 2021-07-23 广东医科大学附属医院 Be used for intracranial aneurysm CFD diagnosis visual device
CN113192004B (en) * 2021-04-01 2022-01-18 温州医科大学附属第二医院（温州医科大学附属育英儿童医院） Simplified calculation method based on coronary artery multimode image crack formation and expansion and plaque rupture risk quantitative evaluation system
CN113192004A (en) * 2021-04-01 2021-07-30 温州医科大学附属第二医院（温州医科大学附属育英儿童医院） Simplified calculation method based on coronary artery multimode image crack formation and expansion and plaque rupture risk quantitative evaluation system
CN113133827B (en) * 2021-04-07 2022-08-05 昆明同心医联科技有限公司 Preoperative prediction method, system, terminal and medium for intracranial aneurysm operation
CN113133827A (en) * 2021-04-07 2021-07-20 昆明同心医联科技有限公司 Preoperative prediction method, system, terminal and medium for intracranial aneurysm operation
CN114145843A (en) * 2021-11-05 2022-03-08 北京市神经外科研究所 Novel intracranial aneurysm weak area evaluation method
CN114145843B (en) * 2021-11-05 2023-07-04 北京市神经外科研究所 Novel intracranial aneurysm weak area assessment method
CN114287954A (en) * 2021-12-13 2022-04-08 杭州脉流科技有限公司 Functional assessment method, system, electronic device and medium for intracranial vascular stenosis
CN114972165A (en) * 2022-03-24 2022-08-30 中山大学孙逸仙纪念医院 Method and device for measuring time-average shearing force
CN114972165B (en) * 2022-03-24 2024-03-15 中山大学孙逸仙纪念医院 Method and device for measuring time average shearing force
CN116485800A (en) * 2023-06-26 2023-07-25 杭州脉流科技有限公司 Automatic acquisition method, device, equipment and storage medium for morphological parameters of aneurysms
CN116485800B (en) * 2023-06-26 2023-09-08 杭州脉流科技有限公司 Automatic acquisition method, device, equipment and storage medium for morphological parameters of aneurysms
CN116919374A (en) * 2023-07-19 2023-10-24 西安交通大学 Intracranial aneurysm and method and system for evaluating blood flow dynamics parameters in aneurysm-carrying artery
CN116919374B (en) * 2023-07-19 2024-04-12 西安交通大学 Intracranial aneurysm and method and system for evaluating blood flow dynamics parameters in aneurysm-carrying artery
CN117438092A (en) * 2023-12-20 2024-01-23 杭州脉流科技有限公司 Intracranial aneurysm rupture risk prediction device, computer device, and storage medium
CN117438092B (en) * 2023-12-20 2024-03-22 杭州脉流科技有限公司 Intracranial aneurysm rupture risk prediction device, computer device, and storage medium

Also Published As

Publication number Publication date
CN109907732B (en) 2022-12-02

Similar Documents

Publication Publication Date Title
CN109907732A (en) A kind of appraisal procedure and system of rupture of intracranial aneurysm risk
CN110866914B (en) Evaluation method, system, equipment and medium for cerebral aneurysm hemodynamic index
CN109924956A (en) A kind of measurement method and device of the Morphologic Parameters of intracranial aneurysm image
CN104768465B (en) Blood flow reserve score (FFR) index
CN109166124A (en) A kind of retinal vascular morphologies quantization method based on connected region
CN112932663A (en) Intelligent auxiliary method and system for improving safety of laparoscopic cholecystectomy
CN110168613A (en) For using reduced-order model and machine learning to estimate the system and method for blood flow characteristic
WO2013031741A1 (en) System for diagnosing bloodflow characteristics, method thereof, and computer software program
KR20190105180A (en) Apparatus for Lesion Diagnosis Based on Convolutional Neural Network and Method thereof
CN104546007B (en) Anti-interference processing method and device for fetal movement detection
CN113133827B (en) Preoperative prediction method, system, terminal and medium for intracranial aneurysm operation
CN108511055A (en) Ventricular premature beat identifying system and method based on Multiple Classifier Fusion and diagnostic rule
CN109567868A (en) A kind of CTG fetal rhythm methods of marking and system
CN108665449A (en) Predict the deep learning model and device of the flow characteristic in blood flow vector path
KR20190114384A (en) Apparatus and method for skin lesion diagnosis based on neural network
CN109273093A (en) A kind of construction method and building system of Kawasaki disease risk evaluation model
Savaş et al. Performance comparison of carotid artery intima media thickness classification by deep learning methods
WO2023097780A1 (en) Classification method and device for classifying patient‑ventilator asynchrony phenomenon in mechanical ventilation process
CN115222674A (en) Detection device for intracranial aneurysm rupture risk based on multi-dimensional feature fusion
Zhang Voice feature selection to improve performance of machine learning models for voice production inversion
CN116919374B (en) Intracranial aneurysm and method and system for evaluating blood flow dynamics parameters in aneurysm-carrying artery
CN106963368A (en) QRS complex detection method and device
CN110378353A (en) A kind of tongue picture feature extracting method, system and computer readable storage medium
CN115953822A (en) Face video false distinguishing method and device based on rPPG physiological signal
CN115761024A (en) Blood vessel mask image generation method and device, electronic equipment and storage medium

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant
TR01 Transfer of patent right

Effective date of registration: 20230113

Address after: 1301, Floor 13, Building 3, Block 2, Heyang Plaza, No. 13, Changjiang Road, Guicheng Street, Nanhai District, Foshan City, Guangdong Province, 528000

Patentee after: Weizhi medical technology (Foshan) Co.,Ltd.

Address before: 510000 room 3025, No. 95, Jinling North Road, Nansha street, Nansha District, Guangzhou City, Guangdong Province

Patentee before: GUANGZHOU XINMAI TECHNOLOGY Co.,Ltd.

TR01 Transfer of patent right