CN109907827A - A kind of operation guiding system of mandibular angle bone cutting art - Google Patents
A kind of operation guiding system of mandibular angle bone cutting art Download PDFInfo
- Publication number
- CN109907827A CN109907827A CN201910305289.0A CN201910305289A CN109907827A CN 109907827 A CN109907827 A CN 109907827A CN 201910305289 A CN201910305289 A CN 201910305289A CN 109907827 A CN109907827 A CN 109907827A
- Authority
- CN
- China
- Prior art keywords
- preoperative
- patient
- osteotomy surface
- guiding system
- danger area
- 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
Links
Landscapes
- Dental Tools And Instruments Or Auxiliary Dental Instruments (AREA)
Abstract
The invention discloses a kind of operation guiding systems of mandibular angle bone cutting art, the following steps are included: S1, be based on multitask convolutional neural networks, according to the related data of the past mandibular angle bone cutting patient with operation, establish osteotomy surface prediction model study version, it is trained with a group related data for mandibular angle bone cutting patient with operation is newly entered, stable osteotomy surface prediction model is obtained, then is superimposed danger area data set, osteotomy surface prediction model is obtained and stablizes version；S2, the relevant information of mandibular angle bone cutting art patient is inputted to osteotomy surface prediction model, predicting surgical rear face 3D effect changes range；S3, according to the maximum osteotomy amount of mandibular angle bone cutting art patient and preoperative CT, drawn on eyeglass screen and project 3-D image see-through, with the fitting of visual area real-time imaging；S4, function superposition, and continuous testing improvement are constantly carried out to osteotomy surface prediction model, improve operation guiding system.This system predicts postoperative effect that real-time rendering in art improves operation precision by establishing model.
Description
Technical field
The present invention relates to Technology of surgery navigation fields, more particularly, to a kind of operation guiding system of mandibular angle bone cutting art.
Background technique
Due to the variation of mandibular angle bone cutting postoperative patient facial appearance, remove with Mandible Osteotomy amount phase outside the Pass, it is soft with part
Also some association of the variation of soft tissue volume amount caused by tissue tension changes.Therefore, mandibular angle bone cutting art goes bone amount
It is not to do subtraction between pre-operative patients facial appearance and Estimating the result, is previously scanned by means of three dimensional CT and face 3D merely
Photographic system carries out the 3D design of surgical effect, can not Accurate Prediction reach postoperative prediction in mandibular angle bone cutting art and imitate
Angle of mandible removes bone amount and osteotomy surface morphologic localization when fruit.
The existing operation guiding system for mandibular angle bone cutting art, by carrying out CT scan to patient, and by shadow
As data are rebuild and are handled, osteotomy line is designed by the past clinical experience, and be labeled in the three-dimensional data of mandibular
On model, osteotomy line is subdivided into multiple brill points by robot perceptual system, by carrying out many places along osteotomy line on bone face
Point drilling is bored to realize osteotomy.Meanwhile the system marks complex by being formed in lower jaw angular region drilling linkage flag module, or
Facing mould is customized according to denture form under patient, and mark module is connected on facing mould, is connected in art by patient-worn
The facing mould of mark module judges the relative position of mandibular.In art, system is by augmented reality by distinguishing label
Module judges mandibular relative position, and determines osteotomy line position, passes through RAS row mandibular angle bone cutting art to realize.The operation is led
Boat system passes through clinical practice application, and mean error is smaller, it is ensured that the safety of operation, while in auxiliary doctors experience product
It is tired etc. to have stronger advantage.But presently, there are three shortcomings for the system: (1) needing additional fixed signal point: by advising
Patient-worn is connected with the facing mould of mark module to determine osteotomy line, since connection type is non-rigid connection, gets the bid in art
Remember that there are the risks of higher relative shift between module and mandibular, there are errors so as to cause the judgement of osteotomy line, reduce
Operation safety；Or mark module is connected to lower jaw angular region by bore mode, though mark module is greatly improved under
The stability of relative positional relationship between jawbone body, but since the mandibular angle bone cutting art actual operation operating space of intra-oral approach is narrow
And it is deep, operation of fixation mark module difficulty itself is higher, while mark module is excessively huge with respect to visual area, can be realized properly
The applicable case for placing mark module is greatly limited.(2) operation guiding system is for the processing of osteotomy mode
Using the mode of mechanical interruption punching, due to the duct through bone tissue that punching is formed be it is linear, this mode for
It needing to carry out the case of mandibular osteo-distraction removal art simultaneously and is not suitable for, the osteotomy surface formed is a flat surface and non-curved,
Largely limit the application range of the operation guiding system.(3) operation guiding system is not to mandibular angle bone cutting art
Soft tissue change afterwards is paid attention to, the past experience of osteotomy line designed completely by patient, and no quantization index, this is
Although system improves the safety of operation, Accurate Prediction patient postoperative effect and raising patient satisfaction etc. are had no
Advantage.
Therefore, a kind of operation for mandibular angle bone cutting art based on artificial intelligence technology and augmented reality is designed to lead
Boat system is current urgent problem to be solved.
Summary of the invention
The object of the present invention is to provide a kind of operation guiding systems of mandibular angle bone cutting art, predict mould by establishing osteotomy surface
Type, and to model be trained with it is perfect, obtain operation guiding system, provide technical support for operation, predict postoperative effect, mention
Height operation precision, reduces operation risk.
Foregoing invention purpose of the invention is achieved by the following technical programs:
A kind of operation guiding system of mandibular angle bone cutting art, comprising the following steps:
S1, osteotomy surface is established according to the related data of the past mandibular angle bone cutting patient with operation based on multitask convolutional neural networks
Prediction model learns version, with newly entering a group related data for mandibular angle bone cutting patient with operation, to osteotomy surface prediction model learn version into
Row training, obtains stable osteotomy surface prediction model, then be superimposed danger area data set, obtains osteotomy surface prediction model and stablizes version；
S2, the relevant information of mandibular angle bone cutting art patient is inputted into osteotomy surface prediction model, predicts amputation line, face of boning, and
Predicting surgical rear face effect；
S3, according to the maximum osteotomy amount of mandibular angle bone cutting art patient and preoperative CT, drawn on eyeglass screen project it is see-through
, with visual area real-time imaging fitting 3-D image；
S4, function superposition, and continuous testing improvement are constantly carried out to osteotomy surface prediction model, improve operation guiding system.
The present invention is further arranged to: in step S1, the related data of the past mandibular angle bone cutting patient with operation includes
Preoperative CT image, postoperative CT image, preoperative mug shot, postoperative mug shot；To preoperative CT image, postoperative CT image pixel-class
It is compared after alignment, obtained difference is the first final osteotomy surface, quantifies to the first final osteotomy surface, obtains first
Final osteotomy surface parameter；According to preoperative CT image, nerve the first danger area 1 out of shape, arteriovenous the first danger area 2 out of shape are obtained,
First danger area 1, the first danger area 2 are quantified, the first danger area 1,2 parameter of the first danger area are obtained.
The present invention is further arranged to: in step S1, by the final osteotomy surface parameter data set of the past patient's different perspectives,
Preoperative mug shot data set, postoperative mug shot data set, preoperative CT image data set, composing training collection are inputted more
Task convolutional neural networks are trained, and are obtained osteotomy surface prediction model and are learnt version, i.e. operation guiding system 1.0 editions.
The present invention is further arranged to: by the second final osteotomy surface parameter data set of new patient in group's different perspectives, art
Front face picture data collection, postoperative mug shot data set constitute test set, survey to osteotomy surface prediction model study version
Examination, obtains stable osteotomy surface prediction model.
The present invention is further arranged to: danger area data set includes the first danger area data of the past patient, newly enters a group trouble
The second danger area data of person.
The present invention is further arranged to: in step S2, the relevant information of mandibular angle bone cutting art patient include preoperative CT image,
Preoperative mug shot, postoperative prediction mug shot；By preoperative CT image obtain mandibular angle bone cutting art patient lower denture rivet point,
The third danger area 2 in the third danger area 1 in slot nerve area out of shape and nervus mentalis area out of shape, facial artery and posterior facial vein area out of shape；
Avoid third danger area 1, third danger area 2, the maximum for obtaining mandibular angle bone cutting art patient is boned range.
The present invention is further arranged to: being boned range, preoperative mug shot, preoperative CT image, is obtained postoperative according to maximum
Facial 3D maximum change amount effect prediction.
The present invention is further arranged to: in step S3, obtain tooth rivet point from preoperative CT, by preoperative CT, preoperative photo,
After predicting surgical picture data collection, maximum bone range input osteotomy surface prediction model stablize version, predict amputation line, face of boning；
Further according to different perspectives, in conjunction with AR equipment, real-time rendering visual area amputates line, face of boning, third danger area 1, third danger area 2,
It is projeced into eyeglass screen after visualization of 3 d model is superimposed by AR system with practical visual area, completes operation guiding system 3.0
Version.
The present invention is further arranged to: in step S4, postoperative facial 3D effect being predicted, operation guiding system is superimposed on
3.0 editions, improvement system is improved, completes operation guiding system 4.0 editions；
The present invention is further arranged to: being debugged operation guiding system 4.0 editions repeatedly, is applied in clinical practice work, root
It needs further progress to upgrade according to actual conditions, increases the stability of system, improve in postoperative effect expection and surgical procedure
Precision, complete operation guiding system 5.0 editions.
Compared with prior art, advantageous effects of the invention are as follows:
1. the application analyzes and constructs osteotomy surface data set by utilizing historical data, mention to study the intelligent predicting of osteotomy surface
For data basis.
2. further, to the non-linear effects of final postoperative effect and modeled by study soft tissue variable, it is real
The now accurate estimation based on preoperative CT data, preoperative photo and simulation postoperative effect to osteotomy surface；
3. further, this system predicts postoperative effect by establishing model, real-time rendering in art, operation essence is improved
Degree reduces operation risk, shortens operating time, reduces postoperative complication, improves patient satisfaction.
4. carrying out real-time early warning in art further, in conjunction with AR equipment, touching danger zone is avoided, guarantees to pacify in art
Entirely.
5. further, the application improves the precision of prediction by convolutional network in conjunction with mandibular angle bone cutting art, make postoperative
Effect is more preferable.
Detailed description of the invention
Fig. 1 is the operation guiding system general construction schematic diagram of a specific embodiment of the invention；
Fig. 2 is the operation guiding system schematic diagram of a specific embodiment of the invention；
Fig. 3 is that the prediction model of a specific embodiment of the invention establishes schematic diagram；
Fig. 4 is the osteotomy surface prediction schematic diagram of a specific embodiment of the invention.
Specific embodiment
Below in conjunction with attached drawing, invention is further described in detail.
Fig. 1 is the scantling plan of this operation guiding system.
Specifically, a kind of operation guiding system of mandibular angle bone cutting art, as shown in Figure 2, comprising the following steps:
S1, according to the related data of the past mandibular angle bone cutting patient with operation, osteotomy surface prediction model study version is established, with newly entering group
The related data of mandibular angle bone cutting patient with operation tests osteotomy surface prediction model study version, obtains stable osteotomy surface
Prediction model；
Specifically, as shown in figure 3, including the following steps:
A1, the first osteotomy surface parameter is obtained according to the preoperative CT image of the past patient's mandibular angle bone cutting patient with operation, postoperative CT image,
In conjunction with preoperative photo, the postoperative film of the past patient, multitask convolutional neural networks are based on, obtain the study of osteotomy surface prediction model
Version, obtains operation guiding system 1.0 editions；
A2, the preoperative CT image of new patient in group, postoperative CT image are collected, the second osteotomy surface parameter is obtained, in conjunction with preoperative photo, art
Photo afterwards constructs test set, tests osteotomy surface prediction model, stable osteotomy surface prediction model is obtained, in conjunction with both
Toward the danger area of patient, the danger area of new patient in group, obtains osteotomy surface prediction model and stablize version, complete operation guiding system
2.0 version.
It is described further below:
The alignment that the preoperative CT image of the past mandibular angle bone cutting patient with operation, postoperative CT image are carried out to pixel scale, is compared
Compared with obtained difference is the final osteotomy surface of the past patient, i.e., the first final osteotomy surface tears the first final osteotomy surface open
Divide and demarcates.
Firstly, the first final osteotomy surface is divided into the first mandibular amputation line α 1 and the first mandibular osteo-distraction removal plane β 1
Two component parts, directly obtained on CT in the preoperative the first mandibular amputation line α 1 and the first mandibular osteo-distraction remove plane β 1,
The markup information of first lower tooth rivet point, the first lower tooth rivet point include multiple points, and multiple lower tooth rivet points determine under first
Denture rivet point reference planes γ 1.
Then, quantification treatment is carried out to the first final osteotomy surface, respectively indicates the first mandibular amputation line α 1, the with parameter
One mandibular osteo-distraction removes the positional relationship between the lower denture rivet point reference planes γ 1 of plane β 1 and first, that is, under first
On the basis of denture rivet point reference planes γ 1, the first mandibular amputation line α 1 is respectively indicated with parameter, the first mandibular osteo-distraction is gone
Except the positional relationship of plane β 1, specifically, indicate that the first mandibular amputates 1 the first lower tooth of geometric center distance of line α with parameter 11
The distance of 1 geometric center of column rivet point reference planes γ；The lower denture of the first mandibular amputation line α 1 and first is indicated with parameter 12
The deflection angle of rivet point reference planes γ 1；The first mandibular osteo-distraction removal 1 geometric center distance of plane β the is indicated with parameter 13
The distance of 1 geometric center of denture rivet point reference planes γ once；Indicate that the first mandibular osteo-distraction removes plane β 1 with parameter 14
With the deflection angle of the first lower denture rivet point reference planes γ 1.
Because final osteotomy surface is three-dimensional structure, thus the parameter at same visual angle constitutes a data set, the ginseng of different perspectives
Array is at different data sets.
The multiple perspective data collection of multiple perspective data collection, preoperative photo of the preoperative CT of the past patient, the multiple views of postoperative film
Angular data collection, final osteotomy surface data set, composing training collection, input multitask convolutional neural networks are trained, and obtain osteotomy
Face prediction model learns version, i.e. operation guiding system 1.0 editions.
In this step, learns soft tissue variable to the non-linear effects of final postoperative effect and model, realization is based on
The accurate estimation of preoperative CT data, preoperative photo and simulation postoperative effect to osteotomy surface.
Version is learnt for osteotomy surface prediction model, needs to carry out stability test.
Acquisition newly enters the data of group mandibular angle bone cutting art patient, forms test set.
Likewise, the preoperative CT image of new patient in group, postoperative CT image to be carried out to the alignment of pixel scale, compared
Compared with obtained difference is the final osteotomy surface of new patient in group, i.e., the second final osteotomy surface.Second final osteotomy surface is carried out
It splits and demarcates.
Firstly, the second final osteotomy surface is divided into the second mandibular amputation line α 2 and the second mandibular osteo-distraction removal plane β 2
Two component parts, directly obtained on CT in the preoperative the second mandibular amputation line α 2 and the second mandibular osteo-distraction remove plane β 2,
The markup information of second lower tooth rivet point, the second lower tooth rivet point include multiple points, and multiple lower tooth rivet points determine under second
Denture rivet point reference planes γ 2.
Then, quantification treatment is carried out to the second final osteotomy surface, respectively indicates the second mandibular amputation line α 2, the with parameter
Two mandibular osteo-distractions remove the correlation between the lower denture rivet point reference planes γ 2 of plane β 2 and second, that is, under second
On the basis of denture rivet point reference planes γ 2, the second mandibular amputation line α 2 is respectively indicated with parameter, the second mandibular osteo-distraction is gone
Except the positional relationship of plane β 2, specifically, indicate that the second mandibular amputates 2 the second lower tooth of geometric center distance of line α with parameter 21
The distance of 2 geometric center of column rivet point reference planes γ；The lower denture of the second mandibular amputation line α 2 and second is indicated with parameter 22
The deflection angle of rivet point reference planes γ 2；The second mandibular osteo-distraction removal 2 geometric center distance of plane β the is indicated with parameter 23
The distance of two lower 2 geometric centers of denture rivet point reference planes γ；Indicate that the second mandibular osteo-distraction removes plane β 2 with parameter 24
With the deflection angle of the second lower denture rivet point reference planes γ 2.
By the data set of the above parameter of different perspectives, the data set of the second final osteotomy surface of new patient in group is constituted.
By the data set of the second final osteotomy surface, preoperative CT, in conjunction with preoperative photo, the postoperative film of new patient in group, structure
At test set.
Test set data input osteotomy surface prediction model study version is tested, stable osteotomy surface prediction mould is obtained
Type improves the Stability and veracity of operation guiding system.
According to the preoperative CT of the past patient, inferior alveolar nerve area out of shape and the nervus mentalis Qu Wei out of shape of the past patient are marked
One danger area 1, marks facial artery and posterior facial vein area out of shape is the first danger area 2；Quantitative evaluation, building are carried out to each danger area
First danger area data set.
First danger area is quantified, indicates that denture rivet point reference planes are descended with first respectively in each danger area with parameter
Relationship between γ 1 is specifically indicated in 1 the first reference planes of geometric center distance γ of the first danger area, 1 geometry with parameter 15
The distance of the heart indicates the deflection angle in the first danger area 1 and the first reference planes γ 1 with parameter 16；The first danger is indicated with parameter 17
The distance of 2 the first reference planes of geometric center distance γ of danger zone, 1 geometric center, indicates the first danger area 2 and first with parameter 18
The deflection angle of reference planes γ 1.
According to the preoperative CT of new patient in group, marks the inferior alveolar nerve area out of shape of the past patient and nervus mentalis area out of shape is
Second danger area 1, marks facial artery and posterior facial vein area out of shape is the second danger area 2；Quantitative evaluation, structure are carried out to each danger area
Build the second risk data collection.
Similarly, the second danger area is quantified, indicates that denture rivet point is descended with second respectively in each danger area with parameter
Relationship between reference planes γ 2 specifically indicates 1 the first reference planes of geometric center distance of the second danger area with parameter 25
The distance of 2 geometric center of γ indicates the deflection angle in the second danger area 1 and the first reference planes γ 2 with parameter 26；With 27 table of parameter
The distance for showing 2 the first reference planes of geometric center distance γ of the second danger area, 2 geometric center, indicates the second danger area with parameter 28
2 and first reference planes γ 2 deflection angle.
The osteotomy surface prediction model for being added to stable by the first danger area data set, the second danger area data set, is cut
Bone face prediction model stablizes version, completes building operation guiding system 2.0 editions.
S2, the relevant information of mandibular angle bone cutting art patient is inputted into osteotomy surface prediction model, predicting surgical rear face 3D effect
Maximum change range.
To the patient that will carry out mandibular angle bone cutting art, i.e. actual patient, according to its preoperative CT, labeled as its lower tooth socket mind
It is third danger area 1 through area out of shape and nervus mentalis area out of shape, marks its facial artery and posterior facial vein area out of shape is third danger area
2；Quantitative evaluation is carried out to each danger area, constructs third danger area data set.
According to the preoperative CT image of mandibular angle bone cutting patient with operation, lower denture rivet point is grabbed, mandibular angle bone cutting art is obtained
Patient lower denture rivet point third reference planes γ 3.
According to the preoperative CT image of mandibular angle bone cutting patient with operation, third danger area 1, third danger area 2 are avoided, is obtained down
The maximum of jaw angle bone-culting operation patient is boned range.
The maximum of mandibular angle bone cutting patient with operation range of boning is quantified, including maximum range of boning is torn open
Divide, mark；Maximum bone range be split as third mandibular amputation line α 3 and third mandibular osteo-distraction remove plane β 3 two
Point, CT image Direct Mark third mandibular amputation line α 3 and third mandibular osteo-distraction remove plane β 3 in the preoperative.
Quantitative estimation third mandibular amputate line α 3, third mandibular osteo-distraction removal plane β 3 respectively with third reference planes
Relationship between γ 3.Specifically, indicate third mandibular amputation 3 geometric center of line α apart from third reference planes γ with parameter 31
The distance of 3 geometric centers；The deflection angle of third mandibular amputation line α 3 and third reference planes γ 3 is indicated with parameter 32；With ginseng
Number 33 indicates distance of third mandibular osteo-distraction removal 3 geometric center of plane β apart from 3 geometric center of third reference planes γ；With
Parameter 34 indicates the deflection angle of third mandibular osteo-distraction removal plane β 3 and third reference planes γ 3.
Quantitative evaluation is carried out to each danger area, indicates relationship of each danger area respectively between reference planes γ 3 with parameter,
Wherein, distance of 1 geometric center of third danger area apart from 3 geometric center of third reference planes γ is indicated with parameter 35, use parameter
36 indicate the deflection angle in third danger area 1 and reference planes γ 3；2 geometric center distance of third danger area is indicated with parameter 37
The distance of three reference planes γ, 3 geometric center indicates the deflection angle in third danger area 2 Yu third reference planes γ 3 with parameter 38.
Maximum range of boning is not represented as osteotomy surface of finally performing the operation.
By the preoperative CT, preoperative photo, maximum of mandibular angle bone cutting patient with operation bone range input osteotomy surface prediction model
Stablize version, obtain postoperative facial maximum change amount 3D effect prediction, is i.e. predicting surgical rear face effect picture changes range.
S3, according to the preoperative CT of mandibular angle bone cutting art patient, amputate line, face of boning, draw and project on eyeglass screen
3-D image see-through, with the fitting of visual area real-time imaging.
Specifically, as shown in figure 4, including the following steps:
B1, the preoperative CT image according to mandibular angle bone cutting patient with operation obtain the lower denture rivet of mandibular angle bone cutting patient with operation
Point, danger area, and demarcate maximum and bone range；
B2, range that the preoperative CT image of mandibular angle bone cutting patient with operation, preoperative photo, prediction postoperative film, maximum are boned, it is defeated
Enter osteotomy surface prediction model, predict the amputation line of actual patient, face of boning, in conjunction with AR equipment, real-time rendering visual area amputate line,
It bones face, danger area.
It is described further below: according to weight of equipment, performance degree of stability, wearing mode soundness, whether to meet operation
The conditions such as sterile principle, but it tests, select, purchasing suitable wearable augmented reality equipment and its software for being used for secondary development
Platform selects suitable AR(Augmented Reality) equipment, also referred to as wearable augmented reality equipment.
Using augmented reality, the real-time rendering and fitting in osteotomy surface and danger area under visual area are realized, improve angle of mandible
The osteotomy precision of bone-culting operation realizes the forewarning function to patient, avoids touching danger zone.
The range input of boning of the preoperative CT of mandibular angle bone cutting patient with operation, preoperative photo, prediction postoperative film, maximum is cut
Bone face prediction model stablizes version, predicts the amputation line of actual patient, face of boning；Further according to different perspectives amputation line, bone
Face, danger area, in conjunction with AR equipment, real-time rendering visual area amputates line, face of boning, third danger area 1, third danger area 2, will be visual
Change after threedimensional model is superimposed by AR system with practical visual area and be projeced into eyeglass screen, completes operation guiding system 3.0 editions.
Specifically, it is based on AR equipment, establishes a set of visualization of 3 d model for osteotomy surface in mandibular angle bone cutting art, and
In conjunction with three-dimensional CT image, more of labelling side lower tooth, and multiple rivet points are set accordingly, according to preoperative CT image, label is dangerous
Area, and in the three-dimensional mode, determine the spatial relationship of possible rivet point, danger area and osteotomy surface threedimensional model.
During actual operation, the camera carried by the AR equipment that patient dresses shoots visual area and is grabbed
Default rivet point is taken, according to the three-dimensional space position relationship built, see-through and art is projected on AR device screen
Osteotomy surface, danger area 1,2 image of danger area after the fitting of wild real-time imaging, realize amputated in visual area line, face of boning, danger area 1,
The real-time rendering in danger area 2.According to image in a large amount of practical art, filters out and be easy to grab and do not influence spatial relation
Construct 3 or so rivet points of stability.Image automatic identification technology based on AR equipment, is constructed in real time to visual area image
Analysis identification, automatically grab the system function of default rivet point, and make the system function with build in advance rivet point, danger
The 3-dimensional image in area 1, danger area 2 and osteotomy surface combines, and makes patient in actual operation by wearing AR equipment and auxiliary at its
Help down, realize the threedimensional model of osteotomy surface project see-throughly on AR device screen, and with from wearer visual angle penetrate AR
The function of mandibular portions fitting in the patient visual area that device screen is observed, while being thrown see-throughly on AR device screen
Shadow goes out third danger area 1,2 image of third danger area, after corresponding to inferior alveolar nerve and nervus mentalis area out of shape and facial artery and face
Vein area out of shape realizes the forewarning function to patient.
Meanwhile during actual operation, operation guiding system is tested and is adjusted, reaches system accurate positioning
And projection can be stablized, to realize navigation function of the operation guiding system in mandibular angle bone cutting operation.
S4, function superposition, and continuous testing improvement are constantly carried out to osteotomy surface prediction model, improve operation guiding system.
Operation guiding system 2.0 editions that postoperative facial 3D effect prediction in step S2 is added in step S3, it is perfect
Improvement system completes operation guiding system 4.0 editions；
After operation guiding system 4.0 editions debugging repeatedly, it is applied in clinical practice work, need according to the actual situation
It wants further progress to upgrade, increases the stability of system, improve the precision in postoperative effect expection and surgical procedure, complete hand
Art navigation system 5.0 editions.
For the patient, the application demarcates maximum according to the preoperative CT of actual patient and danger area 1,2 and bones range, passes through
It inputs preoperative photo, preoperative CT and maximum and goes bone amount, obtain postoperative facial 3D effect maximum change figure by artificial intelligence technology,
Prediction face contour adjustable range realization postoperative to patient is estimated, and the achievable high accurancy and precision to patient's postoperative effect is completed
Personalized designs, cost is linked up before desmopyknosis, improves patient satisfaction.
For doctor, the application by label danger area, bone by the maximum that operation guiding system prejudges patient automatically
Range, the postoperative prediction face contour adjustable range of backstepping, realize based on 3D photograph and processing system, can be real to postoperative effect
It is existing, accurately personalized designs carry out visual area projection to amputation line, face of boning, improve operation precision in conjunction with AR equipment,
Danger area is marked under visual area, suggesting effect is played to doctor, reduces operation risk, shortens operating time, reduces postoperative complication.
The embodiment of present embodiment is presently preferred embodiments of the present invention, not limits protection of the invention according to this
Range, therefore: the equivalence changes that all structures under this invention, shape, principle are done, should all be covered by protection scope of the present invention it
It is interior.
Claims (10)
1. a kind of operation guiding system of mandibular angle bone cutting art, it is characterised in that: the following steps are included:
S1, osteotomy surface is established according to the related data of the past mandibular angle bone cutting patient with operation based on multitask convolutional neural networks
Prediction model learns version, with newly entering a group related data for mandibular angle bone cutting patient with operation, to osteotomy surface prediction model learn version into
Row training, obtains stable osteotomy surface prediction model, then be superimposed danger area data set, obtains osteotomy surface prediction model and stablizes version；
S2, the relevant information of mandibular angle bone cutting art patient is inputted to osteotomy surface prediction model, predicting surgical rear face 3D effect is maximum
Change range；
S3, according to the maximum osteotomy amount of mandibular angle bone cutting art patient and preoperative CT, preoperative photo, prediction postoperative effect, in eyeglass
It is drawn on screen and projects 3-D image see-through, with the fitting of visual area real-time imaging；
S4, function superposition, and continuous testing improvement are constantly carried out to osteotomy surface prediction model, improve operation guiding system.
2. operation guiding system according to claim 1, it is characterised in that: in step S1, the past mandibular angle bone cutting
The related data of patient with operation includes preoperative CT image, postoperative CT image, preoperative mug shot, postoperative mug shot；
To being compared after preoperative CT image, the alignment of postoperative CT image pixel-class, obtained difference is the first final osteotomy surface,
First final osteotomy surface is quantified, the first final osteotomy surface parameter is obtained；
According to preoperative CT image, nerve the first danger area 1 out of shape, arteriovenous the first danger area 2 out of shape are obtained, to the first danger area
1, the first danger area 2 is quantified, and obtains the first danger area 1,2 parameter of the first danger area.
3. operation guiding system according to claim 1, it is characterised in that: in step S1, by the past patient's different perspectives
Final osteotomy surface parameter data set, preoperative mug shot data set, postoperative mug shot data set, preoperative CT image data
Collection, composing training collection are inputted multitask convolutional neural networks and are trained, and obtain osteotomy surface prediction model and learn version, i.e.,
Operation guiding system 1.0 editions.
4. operation guiding system according to claim 3, it is characterised in that: by new patient in group's different perspectives second most
Whole osteotomy surface parameter data set, preoperative mug shot data set, postoperative mug shot data set constitute test set, to osteotomy surface
Prediction model study version is tested, and stable osteotomy surface prediction model is obtained.
5. operation guiding system according to claim 1, it is characterised in that: danger area data set includes the of the past patient
One danger area data, the second danger area data of new patient in group.
6. operation guiding system according to claim 1, it is characterised in that: in step S2, mandibular angle bone cutting art patient's
Relevant information includes preoperative CT image, preoperative mug shot, postoperative prediction mug shot；Angle of mandible is obtained by preoperative CT image to cut
It is quiet behind the lower denture rivet point of bone art patient, the third danger area 1 in slot nerve area out of shape and nervus mentalis area out of shape, facial artery and face
The third danger area 2 in arteries and veins area out of shape；Third danger area 1, third danger area 2 are avoided, the maximum of mandibular angle bone cutting art patient is obtained
It bones range.
7. operation guiding system according to claim 6, it is characterised in that: according to maximum bone range, it is preoperative face shine
Piece, preoperative CT image obtain postoperative face 3D maximum change amount effect prediction.
8. operation guiding system according to claim 6, it is characterised in that: in step S3, obtain tooth riveting from preoperative CT
Follow closely point, by picture data collection, maximum after preoperative CT, preoperative photo, predicting surgical bone range input osteotomy surface prediction model stablize
Version predicts amputation line, face of boning；Further according to different perspectives, in conjunction with AR equipment, real-time rendering visual area amputates line, face of boning, the
Three danger areas 1, third danger area 2 are projeced into eyeglass screen after being superimposed visualization of 3 d model with practical visual area by AR system
Curtain completes operation guiding system 3.0 editions.
9. operation guiding system according to claim 1, it is characterised in that: in step S4, postoperative facial 3D effect is pre-
It surveys, is superimposed on operation guiding system 3.0 editions, improve improvement system, complete operation guiding system 4.0 editions.
10. operation guiding system according to claim 9, it is characterised in that: operation guiding system 4.0 editions are debugged repeatedly,
It is applied in clinical practice work, further progress upgrades according to the needs of actual conditions, increases the stability of system, mentions
Precision in high postoperative effect expection and surgical procedure, completes operation guiding system 5.0 editions.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910305289.0A CN109907827B (en) | 2019-04-16 | 2019-04-16 | Operation navigation system for mandibular angle osteotomy |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910305289.0A CN109907827B (en) | 2019-04-16 | 2019-04-16 | Operation navigation system for mandibular angle osteotomy |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109907827A true CN109907827A (en) | 2019-06-21 |
| CN109907827B CN109907827B (en) | 2020-07-14 |
Family
ID=66977351
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910305289.0A Active CN109907827B (en) | 2019-04-16 | 2019-04-16 | Operation navigation system for mandibular angle osteotomy |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109907827B (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112257912A (en) * | 2020-10-15 | 2021-01-22 | 北京爱康宜诚医疗器材有限公司 | Method and device for predicting operation evaluation information, processor and electronic device |
| CN113052864A (en) * | 2021-03-02 | 2021-06-29 | 四川大学 | Method for predicting body appearance after plastic surgery based on machine learning |
| CN113143457A (en) * | 2021-02-09 | 2021-07-23 | 席庆 | Maxillofacial operation auxiliary system and method based on MR head-mounted equipment |
| CN113768627A (en) * | 2021-09-14 | 2021-12-10 | 武汉联影智融医疗科技有限公司 | Method and device for acquiring receptive field of visual navigator and surgical robot |
| CN116211458A (en) * | 2022-12-12 | 2023-06-06 | 高峰医疗器械(无锡)有限公司 | Implant planning method, device, equipment and storage medium |
| CN116883428A (en) * | 2023-07-07 | 2023-10-13 | 东北大学 | Mandible spiral CT image partition segmentation method |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1578646A (en) * | 2001-10-31 | 2005-02-09 | 画像诊断株式会社 | Medical simulation apparatus and method for controlling 3-dimensional image display in the medical simulation apparatus |
| CN105608741A (en) * | 2015-12-17 | 2016-05-25 | 四川大学 | Computer simulation method for predicting soft tissue appearance change after maxillofacial bone plastic surgery |
| CN105943113A (en) * | 2016-04-13 | 2016-09-21 | 南方医科大学 | Mandible angle osteotomy navigation template preparation method |
| US20170014169A1 (en) * | 2014-03-11 | 2017-01-19 | Ohio State Innovation Foundation | Methods, devices, and manufacture of the devices for musculoskeletal reconstructive surgery |
| CN106529117A (en) * | 2015-09-11 | 2017-03-22 | 西门子保健有限责任公司 | Physiology-driven decision support for therapy planning |
| WO2019056059A1 (en) * | 2017-09-21 | 2019-03-28 | Tmj Orthopaedics Pty Ltd | A surgical procedure for cancerous mandibular reconstruction and a temporary mandibular spacer therefor |
| CN109567942A (en) * | 2018-10-31 | 2019-04-05 | 上海盼研机器人科技有限公司 | Using the craniomaxillofacial surgery robot assisted system of artificial intelligence technology |
-
2019
- 2019-04-16 CN CN201910305289.0A patent/CN109907827B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1578646A (en) * | 2001-10-31 | 2005-02-09 | 画像诊断株式会社 | Medical simulation apparatus and method for controlling 3-dimensional image display in the medical simulation apparatus |
| US20170014169A1 (en) * | 2014-03-11 | 2017-01-19 | Ohio State Innovation Foundation | Methods, devices, and manufacture of the devices for musculoskeletal reconstructive surgery |
| CN106529117A (en) * | 2015-09-11 | 2017-03-22 | 西门子保健有限责任公司 | Physiology-driven decision support for therapy planning |
| CN105608741A (en) * | 2015-12-17 | 2016-05-25 | 四川大学 | Computer simulation method for predicting soft tissue appearance change after maxillofacial bone plastic surgery |
| CN105943113A (en) * | 2016-04-13 | 2016-09-21 | 南方医科大学 | Mandible angle osteotomy navigation template preparation method |
| WO2019056059A1 (en) * | 2017-09-21 | 2019-03-28 | Tmj Orthopaedics Pty Ltd | A surgical procedure for cancerous mandibular reconstruction and a temporary mandibular spacer therefor |
| CN109567942A (en) * | 2018-10-31 | 2019-04-05 | 上海盼研机器人科技有限公司 | Using the craniomaxillofacial surgery robot assisted system of artificial intelligence technology |
Non-Patent Citations (2)
| Title |
|---|
| 刘杰: "基于机器学习的面部整形手术的模拟", 《中国优秀硕士学位论文全文数据库 医药卫生科技辑》 * |
| 李明哲等: "计算机导航辅助下口内入路髁突切除术精确性", 《北京大学学报》 * |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112257912A (en) * | 2020-10-15 | 2021-01-22 | 北京爱康宜诚医疗器材有限公司 | Method and device for predicting operation evaluation information, processor and electronic device |
| CN113143457A (en) * | 2021-02-09 | 2021-07-23 | 席庆 | Maxillofacial operation auxiliary system and method based on MR head-mounted equipment |
| CN113052864A (en) * | 2021-03-02 | 2021-06-29 | 四川大学 | Method for predicting body appearance after plastic surgery based on machine learning |
| CN113052864B (en) * | 2021-03-02 | 2022-12-23 | 四川大学 | Method for predicting body appearance after plastic surgery based on machine learning |
| CN113768627A (en) * | 2021-09-14 | 2021-12-10 | 武汉联影智融医疗科技有限公司 | Method and device for acquiring receptive field of visual navigator and surgical robot |
| CN116211458A (en) * | 2022-12-12 | 2023-06-06 | 高峰医疗器械(无锡)有限公司 | Implant planning method, device, equipment and storage medium |
| CN116211458B (en) * | 2022-12-12 | 2023-10-03 | 高峰医疗器械(无锡)有限公司 | Implant planning method, device, equipment and storage medium |
| CN116883428A (en) * | 2023-07-07 | 2023-10-13 | 东北大学 | Mandible spiral CT image partition segmentation method |
| CN116883428B (en) * | 2023-07-07 | 2024-05-31 | 东北大学 | Mandible spiral CT image partition segmentation method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109907827B (en) | 2020-07-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109907827A (en) | A kind of operation guiding system of mandibular angle bone cutting art | |
| CN105264459B (en) | Tactile enhancing for simulation surgery and virtual reality system | |
| US20230346507A1 (en) | Augmented reality display for cardiac and vascular procedures with compensation for cardiac motion | |
| RU2714665C2 (en) | Guide system for positioning patient for medical imaging | |
| CN109875683A (en) | The method of osteotomy surface prediction model is established in a kind of mandibular angle bone cutting art | |
| JP7543380B2 (en) | Methods for cephalometric analysis | |
| CN104939925B (en) | Depths and surface visualization based on triangulation | |
| KR20170007309A (en) | Method for 3-d cephalometric analysis | |
| US12097064B2 (en) | Method and system for 3D cephalometric analysis | |
| CN106806021A (en) | A kind of VR surgery simulation systems and method based on human organ 3D models | |
| JP2008516727A (en) | Digital ophthalmic workstation | |
| Mamone et al. | Monitoring wound healing with contactless measurements and augmented reality | |
| CN109700550A (en) | A kind of augmented reality method and device for dental operation | |
| CN112885436B (en) | Dental surgery real-time auxiliary system based on augmented reality three-dimensional imaging | |
| KR20190108923A (en) | Plastic surgery/surgery support system using augmented reality | |
| CN114667538A (en) | Viewing system for use in a surgical environment | |
| JP2023505956A (en) | Anatomical feature extraction and presentation using augmented reality | |
| Debarba et al. | Augmented reality visualization of joint movements for physical examination and rehabilitation | |
| WO2022089051A1 (en) | Skull correction scheme generation system and construction method therefor, and skull correction scheme acquisition method and apparatus | |
| JP2006061196A (en) | Constructing/displaying device for three-dimensional eye fundus image | |
| JP2004024739A (en) | Method and apparatus for three-dimensional display and coordinate measurement of fundus oculi | |
| US20210137601A1 (en) | Device and method for imaging during implantation of retina implants | |
| Condino et al. | Registration Sanity Check for AR-guided Surgical Interventions: Experience From Head and Face Surgery | |
| CN111658142A (en) | MR-based focus holographic navigation method and system | |
| KR101988531B1 (en) | Navigation system for liver disease using augmented reality technology and method for organ image display |
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 |