CN108846830A - The method, apparatus and storage medium be automatically positioned to lumbar vertebrae in CT - Google Patents
The method, apparatus and storage medium be automatically positioned to lumbar vertebrae in CT Download PDFInfo
- Publication number
- CN108846830A CN108846830A CN201810517798.5A CN201810517798A CN108846830A CN 108846830 A CN108846830 A CN 108846830A CN 201810517798 A CN201810517798 A CN 201810517798A CN 108846830 A CN108846830 A CN 108846830A
- Authority
- CN
- China
- Prior art keywords
- lumbar vertebrae
- image
- initial
- measured
- vertebra
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0012—Biomedical image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/70—Determining position or orientation of objects or cameras
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10072—Tomographic images
- G06T2207/10081—Computed x-ray tomography [CT]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20084—Artificial neural networks [ANN]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30004—Biomedical image processing
- G06T2207/30008—Bone
- G06T2207/30012—Spine; Backbone
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Computing Systems (AREA)
- Software Systems (AREA)
- Evolutionary Computation (AREA)
- Computational Linguistics (AREA)
- Molecular Biology (AREA)
- Biophysics (AREA)
- General Engineering & Computer Science (AREA)
- Biomedical Technology (AREA)
- Mathematical Physics (AREA)
- Data Mining & Analysis (AREA)
- Artificial Intelligence (AREA)
- Medical Informatics (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Radiology & Medical Imaging (AREA)
- Quality & Reliability (AREA)
- Life Sciences & Earth Sciences (AREA)
- Image Analysis (AREA)
- Apparatus For Radiation Diagnosis (AREA)
Abstract
Method, apparatus and storage medium, method the embodiment of the invention discloses lumbar vertebrae automatic positioning in a kind of couple of CT include：Training step：Initial vertebra CT stereo-picture is acquired, initial vertebra CT stereo-picture is handled to obtain the positional parameter of lumbar vertebrae position；Prediction steps：Acquire vertebra CT stereo-picture to be measured；Projection process is carried out to vertebra CT stereo-picture to be measured using digital reconstruction irradiation image technology and geometrical relationship, obtains DRR image to be measured；DRR image to be measured is predicted using convolutional neural networks and positional parameter, to obtain the lumbar vertebrae location information in DRR image to be measured；Using digital reconstruction irradiation image technology, the lumbar vertebrae location information in DRR image to be measured is mapped in vertebra CT stereo-picture to be measured, to obtain the lumbar vertebrae location information in vertebra CT stereo-picture to be measured.Implement the embodiment of the present invention, robustness, real-time and the accuracy of lumbar vertebrae automatic positioning can be improved.
Description
Technical field
The present invention relates to apply medical imaging graph and image processing technical field, and in particular to lumbar vertebrae is automatic in a kind of couple of CT
The method, apparatus and storage medium of positioning.
Background technique
Medicine CT (Computed Tomography) is successively swept using x-ray is certain thickness to the progress of human body check point
It retouches, obtains a series of images.Clinical application observes the tissues such as bone tissue, liver by CT and makes pathological analysis.Vertebra is as people
Body one relatively obvious and fixed tissue, in the course of surgery can be using vertebra as optimal reference point.Therefore to CT body data
The fast and accurately positioning of middle vertebra becomes the key that such as hepatic pathology is automatically analyzed with diagnosed.Lumbar vertebrae automatic positioning at present can
It is divided into three kinds：1, it manuallys locate；2, the CT body data vertebra based on traditional images processing positions 3, based on conventional machines study
The positioning of CT body data vertebra.It is taken time and effort by software interactive manual positioning, increases doctor's workload.Due to the individual difference of human body
Positioning is realized in the opposite sex, traditional images processing by technologies such as binaryzation, simple data statistics, and robustness needs to be mentioned
It rises；Conventional machines study can position it by position regression function by extracting one or more kinds of features, this side
The Feature Selection quality contraposition of method puts back into the precision returned and has decisive impact, however chooses one or more kinds of features not
CT body data can be more fully described, robustness has to be hoisted, and since CT body Data Data amount is big, the speed of positioning has
It is to be hoisted.
Convolutional neural networks CNN (Convolutional Neural Network) is most represented in depth learning technology
One of network structure, be widely applied in field of image processing acquirement, such as recognition of face, vehicle tracking etc..Convolutional Neural net
Network achievees the effect that robustness is most strong, real-time is best in such as Face datection, vehicle detection application in terms of the positioning.
Digital reconstruction irradiation image DRR (Digitally Reconstructured Radiograph), by vertical to CT
Volume data carries out the result of the x-ray target direction reconstruction image of close copy localization machine.Development and CT with computer technology
The progress of scanning technique, DRR are widely used in outside CT simulator locating, image guided radiation therapy (IGRT) and area of computer aided
The fields such as section.
Based on foregoing description, it is necessary to a kind of lumbar vertebrae automatic positioning scheme based on DRR and CNN technology is provided, to solve
Defect present in the prior art.
Summary of the invention
The embodiment of the present invention is designed to provide the method, apparatus and computer that lumbar vertebrae is automatically positioned in a kind of couple of CT
Readable storage medium storing program for executing, to improve robustness, real-time and the accuracy of lumbar vertebrae automatic positioning.
To achieve the above object, in a first aspect, the embodiment of the invention provides the sides that lumbar vertebrae in a kind of couple of CT is automatically positioned
Method, including：
Training step：Initial vertebra CT stereo-picture is acquired, the initial vertebra CT stereo-picture is handled to obtain
To the positional parameter of lumbar vertebrae position；
Prediction steps：
Acquire vertebra CT stereo-picture to be measured；
The vertebra CT stereo-picture to be measured is carried out at projection using digital reconstruction irradiation image technology and geometrical relationship
Reason, obtains DRR image to be measured；
The DRR image to be measured is predicted using convolutional neural networks and positional parameter, to obtain the DRR to be measured
Lumbar vertebrae location information in image；
Using digital reconstruction irradiation image technology, the lumbar vertebrae location information in the DRR image to be measured is mapped to described
In vertebra CT stereo-picture to be measured, to obtain the lumbar vertebrae location information in vertebra CT stereo-picture to be measured.
As a kind of preferred embodiment of the application, the initial vertebra CT stereo-picture is handled to obtain waist
The positional parameter of vertebra position, specifically includes：
The initial vertebra CT stereo-picture is carried out at projection using digital reconstruction irradiation image technology and geometrical relationship
Reason, obtains initial DRR image；
Lumbar vertebrae position in the initial DRR image is labeled, and records the width of every initial DRR image
Degree, height and the boundary coordinate up and down of lumbar vertebrae position, form training dataset；
The initial DRR image and training dataset are trained using convolutional neural networks, to obtain the positioning
Parameter.
As the application another preferred embodiment, the initial vertebra CT stereo-picture is handled to obtain
The positional parameter of lumbar vertebrae position, specifically includes：
The initial vertebra CT stereo-picture is carried out at projection using digital reconstruction irradiation image technology and geometrical relationship
Reason, obtains initial DRR image, and the initial DRR image is single channel image；
Single pass DRR image is converted into triple channel RGB image；
Lumbar vertebrae position in the initial DRR image is labeled, and records the width of every initial DRR image
Degree, height and the boundary coordinate up and down of lumbar vertebrae position, form training dataset；
By after conversion initial DRR image and training dataset input convolutional neural networks be trained, it is described to obtain
Positional parameter.
Specifically, the DRR image to be measured includes two, wherein one corresponds to the coronal-plane of the CT image to be measured,
Another corresponds to the sagittal plane of the CT image to be measured.
Second aspect, the embodiment of the invention provides the devices that lumbar vertebrae in a kind of couple of CT is automatically positioned, including：
Training module is handled the initial vertebra CT stereo-picture for acquiring initial vertebra CT stereo-picture
To obtain the positional parameter of lumbar vertebrae position；
Prediction module is used for：
Acquire vertebra CT stereo-picture to be measured；
The vertebra CT stereo-picture to be measured is carried out at projection using digital reconstruction irradiation image technology and geometrical relationship
Reason, obtains DRR image to be measured；
The DRR image to be measured is predicted using convolutional neural networks and positional parameter, to obtain the DRR to be measured
Lumbar vertebrae location information in image；
Using digital reconstruction irradiation image technology, the lumbar vertebrae location information in the DRR image to be measured is mapped to described
In vertebra CT stereo-picture to be measured, to obtain the lumbar vertebrae location information in vertebra CT stereo-picture to be measured.
As a kind of preferred embodiment of the application, the prediction module is specifically used for：
The initial vertebra CT stereo-picture is carried out at projection using digital reconstruction irradiation image technology and geometrical relationship
Reason, obtains initial DRR image；
Lumbar vertebrae position in the initial DRR image is labeled, and records the width of every initial DRR image
Degree, height and the boundary coordinate up and down of lumbar vertebrae position, form training dataset；
The initial DRR image and training dataset are trained using convolutional neural networks, to obtain the positioning
Parameter.
As the application another preferred embodiment, the prediction module is specifically used for：
The initial vertebra CT stereo-picture is carried out at projection using digital reconstruction irradiation image technology and geometrical relationship
Reason, obtains initial DRR image, and the initial DRR image is single channel image；
Single pass DRR image is converted into triple channel RGB image；
Lumbar vertebrae position in the initial DRR image is labeled, and records the width of every initial DRR image
Degree, height and the boundary coordinate up and down of lumbar vertebrae position, form training dataset；
By after conversion initial DRR image and training dataset input convolutional neural networks be trained, it is described to obtain
Positional parameter.
The third aspect, the embodiment of the invention provides the device that lumbar vertebrae in a kind of couple of CT is automatically positioned, including it is processor, defeated
Enter equipment, output equipment and memory, the processor, input equipment, output equipment and memory are connected with each other, wherein institute
Memory is stated for storing computer program, the computer program includes program instruction, and the processor is configured for adjusting
It is instructed with described program, the method for executing above-mentioned first aspect.
Fourth aspect, the embodiment of the invention provides a kind of computer readable storage medium, the computer-readable storage
Media storage has computer program, and the computer program includes program instruction, and described program instructs when being executed by a processor
The processor is set to execute method described in above-mentioned first aspect.
Implement the embodiment of the present invention, obtains the positional parameter of lumbar vertebrae position in the training stage, in forecast period, pass through positioning
Parameter and convolutional neural networks predict DRR image to be measured, obtain the lumbar vertebrae location information of DRR image to be measured, finally, will
Lumbar vertebrae location information in DRR image to be measured is mapped in vertebra CT stereo-picture to be measured, realize lumbar vertebrae in CT image from
Dynamic positioning；That is, CT human body 3D Volume reconstruction is 2D DRR image by combining DRR and CNN technology by the embodiment of the present invention,
The positioning of DRR lumbar vertebrae is carried out by CNN again, finally CT 3D volume data coordinate is mapped to by 2D DRR positioning coordinate, realizes essence
Quasi- quickly CT 3D volume data lumbar vertebrae positioning, improves robustness, real-time and the accuracy of lumbar vertebrae automatic positioning.
Detailed description of the invention
It, below will be to specific in order to illustrate more clearly of the specific embodiment of the invention or technical solution in the prior art
Embodiment or attached drawing needed to be used in the description of the prior art are briefly described.In all the appended drawings, similar element
Or part is generally identified by similar appended drawing reference.In attached drawing, each element or part might not be drawn according to actual ratio.
Fig. 1 is the signal of the training step in the method being automatically positioned to lumbar vertebrae in CT that first embodiment of the invention provides
Flow chart；
Fig. 2 is that digital reconstruction irradiation image generates schematic diagram；
Fig. 3 is the two-dimensional virtual X-ray picture obtained using number using digital reconstruction irradiation image technology；
Fig. 4 is the schematic diagram after being labeled to two-dimensional virtual X-ray picture；
Fig. 5 is the architecture diagram of convolutional neural networks；
Fig. 6 is the signal of the prediction steps in the method being automatically positioned to lumbar vertebrae in CT that first embodiment of the invention provides
Flow chart；
Fig. 7 is the structural schematic diagram for the device being automatically positioned to lumbar vertebrae in CT that first embodiment of the invention provides；
Fig. 8 is the structural schematic diagram for the device being automatically positioned to lumbar vertebrae in CT that second embodiment of the invention provides.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation description, it is clear that described embodiments are some of the embodiments of the present invention, instead of all the embodiments.Based on this hair
Embodiment in bright, every other implementation obtained by those of ordinary skill in the art without making creative efforts
Example, shall fall within the protection scope of the present invention.
It should be appreciated that ought use in this specification and in the appended claims, term " includes " and "comprising" instruction
Described feature, entirety, step, operation, the presence of element and/or component, but one or more of the other feature, whole is not precluded
Body, step, operation, the presence or addition of element, component and/or its set.It is also understood that the institute in this description of the invention
The term used is merely for the sake of being not intended to limit the present invention for the purpose of describing particular embodiments.Such as in description of the invention
With it is used in the attached claims like that, other situations unless the context is clearly specified, otherwise singular
" one ", "one" and "the" are intended to include plural form.
It will be further appreciated that the term "and/or" used in description of the invention and the appended claims is
Refer to any combination and all possible combinations of one or more of associated item listed, and including these combinations.
As used in this specification and in the appended claims, term " if " can be according to context quilt
Be construed to " when ... " or " once " or " in response to determination " or " in response to detecting ".Similarly, phrase " if it is determined that " or
" if detecting [described condition or event] " can be interpreted to mean according to context " once it is determined that " or " in response to true
It is fixed " or " once detecting [described condition or event] " or " in response to detecting [described condition or event] ".
The method being automatically positioned provided by the embodiment of the present invention to lumbar vertebrae in CT mainly includes that training step and prediction walk
Suddenly.In training step, initial vertebra CT stereo-picture is handled, the positional parameter of lumbar vertebrae position is obtained；It is walked in prediction
In rapid, vertebra CT stereo-picture to be measured is handled based on positional parameter and convolutional neural networks, it is vertical to obtain vertebra CT to be measured
Lumbar vertebrae location information in body image realizes the automatic positioning of lumbar vertebrae in CT stereo-picture.
Specifically, referring to FIG. 1, being the flow diagram of training step in first embodiment of the invention, as shown, instruction
Practicing step may include steps of：
S101 acquires initial vertebra CT stereo-picture.
S102 throws the initial vertebra CT stereo-picture using digital reconstruction irradiation image technology and geometrical relationship
Shadow processing, obtains initial DRR image.
To more fully understand step S102, first digital reconstruction irradiation image technology and the mapping of CT coordinate are described below：
Digital reconstruction irradiation image cardinal principle, as shown in Fig. 2, being that simulation X-ray is shown in a certain Propagation
The energy loss being made of photoelectric absorption and scattering come.The decaying of X-ray line can use absorption optics in digital reconstruction irradiation image
Model tormulation, such as following formula：
Wherein, S is the length parameter in optical projection direction；I (s) is the optical strength at distance S；T (a) is light intensity
Attenuation coefficient.
By DRR principle it is found that CT body data coordinates can be mapped to DRR coordinate according to formula 2：
Wherein, vl is the number of the corresponding CT body coordinate direction pixel of projection coordinate, and dl is the corresponding pixel of DRR coordinate
Number, ds are DRR pixel spatial resolution, and loDr is that DRR positions coordinate, and SID is distance of the light source to rebuilding plane, and SAD is
Light source is to CT body centre distance, and vzl is projecting direction CT body data pixels number, and vzs is projecting direction CT spatial resolution, vs
CT body data spatial resolution is corresponded to for positioning coordinate points.
Specifically, respectively that every set CT is three-dimensional using digital reconstruction irradiation image technology and geometrical relationship in step S102
Coronal-plane and sagittal plane in image project on the plate on the right side of in Fig. 2, obtain two width DRR images, as shown in Figure 3.It needs
Bright, two width DRR images in Fig. 3, a width is coronal-plane, and a width is sagittal plane.
S103 is labeled the lumbar vertebrae position in the initial DRR image.
Specifically, the position of lumbar vertebrae in DRR image is marked out using annotation tool labelimg, as shown in Figure 4.Wherein,
The mark of lumbar vertebrae position needs certain medical practice, needs to carry out under the guidance of medical practitioner.It, can in specific operation process
Using according to the prominent shape of tail bone in picture, rib cage and ridge and position as referring to.
S104, width, height and the boundary up and down of lumbar vertebrae position for recording every initial DRR image are sat
Mark forms training dataset.
S105 is trained the initial DRR image and training dataset using convolutional neural networks, described to obtain
Positional parameter.
It should be noted that convolutional neural networks single shot detection (SSD) net used in the present embodiment
Network, specific structure are as shown in Figure 5.SSD algorithm is the algorithm of object category and position in a kind of direct forecast image, such as Fig. 5
Shown, the master network structure of algorithm is VGG16, most will be changed to convolutional layer by latter two full articulamentum, and then increase 4 convolution
Layer carrys out tectonic network structure.The output of wherein 5 kinds of different convolutional layers is carried out with two 3 × 3 different convolution kernels respectively
Convolution, the confidence level of an output object category；The location information of another output object generates 4 of reflection object space
Coordinate value.The algorithm realizes trains end to end, even if the resolution ratio of image is lower, also can guarantee the precision of detection.
Specifically, in the present embodiment, initial DRR image is triple channel, by the initial DRR image of triple channel and training
Data set inputs convolutional neural networks as shown in Figure 5 together, and can be generated a set of can be automatically positioned lumbar vertebrae position in DRR image
Positional parameter.
Optionally, in other embodiment of the present invention, initial DRR image obtained in step S102 is single pass,
It needs single channel DRR image being converted to triple channel RGB image at this time, so that it better adapts to the mind of the convolution in the present embodiment
Framework through network.
Referring to FIG. 6, being the flow diagram of prediction steps in first embodiment of the invention, as shown, prediction steps
It may include steps of：
S201 acquires vertebra CT stereo-picture to be measured.
S202 throws the vertebra CT stereo-picture to be measured using digital reconstruction irradiation image technology and geometrical relationship
Shadow processing, obtains DRR image to be measured.
S203 predicts the DRR image to be measured using convolutional neural networks and positional parameter, with obtain it is described to
Survey the lumbar vertebrae location information in DRR image.
Lumbar vertebrae location information in the DRR image to be measured is mapped to by S204 using digital reconstruction irradiation image technology
In the vertebra CT stereo-picture to be measured, to obtain the lumbar vertebrae location information in vertebra CT stereo-picture to be measured.
It should be noted that the detailed process of forecast period please refers to the description of training stage, details are not described herein.
The method for implementing to be automatically positioned lumbar vertebrae in CT provided by the embodiment of the present invention, obtains lumbar vertebrae position in the training stage
The positional parameter set is predicted DRR image to be measured by positional parameter and convolutional neural networks, is obtained in forecast period
The lumbar vertebrae location information of DRR image to be measured, finally, the lumbar vertebrae location information in DRR image to be measured is mapped to vertebra CT to be measured
In stereo-picture, the automatic positioning of lumbar vertebrae in CT image is realized；That is, the embodiment of the present invention is by combining DRR and CNN technology,
It is 2D DRR image by CT human body 3D Volume reconstruction, then the positioning of DRR lumbar vertebrae is carried out by CNN, is finally positioned by 2D DRR
Coordinate is mapped to CT 3D volume data coordinate, realizes precisely quickly CT 3D volume data lumbar vertebrae positioning, it is automatic to improve lumbar vertebrae
Robustness, real-time and the accuracy of positioning.
Correspondingly, on the basis of the method being automatically positioned to lumbar vertebrae in CT provided by above-described embodiment, the present invention is real
It applies example and additionally provides the device that lumbar vertebrae is automatically positioned in a kind of couple of CT.As shown in fig. 7, the apparatus may include：
Training module 100, for acquiring initial vertebra CT stereo-picture, at the initial vertebra CT stereo-picture
Reason is to obtain the positional parameter of lumbar vertebrae position；
Prediction module 200, is used for：
Acquire vertebra CT stereo-picture to be measured；
The vertebra CT stereo-picture to be measured is carried out at projection using digital reconstruction irradiation image technology and geometrical relationship
Reason, obtains DRR image to be measured；
The DRR image to be measured is predicted using convolutional neural networks and positional parameter, to obtain the DRR to be measured
Lumbar vertebrae location information in image；
Using digital reconstruction irradiation image technology, the lumbar vertebrae location information in the DRR image to be measured is mapped to described
In vertebra CT stereo-picture to be measured, to obtain the lumbar vertebrae location information in vertebra CT stereo-picture to be measured.
Specifically, the prediction module 200 is specifically used for：
The initial vertebra CT stereo-picture is carried out at projection using digital reconstruction irradiation image technology and geometrical relationship
Reason, obtains initial DRR image；
Lumbar vertebrae position in the initial DRR image is labeled, and records the width of every initial DRR image
Degree, height and the boundary coordinate up and down of lumbar vertebrae position, form training dataset；
The initial DRR image and training dataset are trained using convolutional neural networks, to obtain the positioning
Parameter.
Optionally, the prediction module 200 is specifically used for：
The initial vertebra CT stereo-picture is carried out at projection using digital reconstruction irradiation image technology and geometrical relationship
Reason, obtains initial DRR image, and the initial DRR image is single channel image；
Single pass DRR image is converted into triple channel RGB image；
Lumbar vertebrae position in the initial DRR image is labeled, and records the width of every initial DRR image
Degree, height and the boundary coordinate up and down of lumbar vertebrae position, form training dataset；
By after conversion initial DRR image and training dataset input convolutional neural networks be trained, it is described to obtain
Positional parameter.
It should be noted that the specific workflow of the device please refers to the description of preceding method embodiment part, herein
It repeats no more.
Implement the device being automatically positioned provided by the embodiment of the present invention to lumbar vertebrae in CT, obtains lumbar vertebrae position in the training stage
The positional parameter set is predicted DRR image to be measured by positional parameter and convolutional neural networks, is obtained in forecast period
The lumbar vertebrae location information of DRR image to be measured, finally, the lumbar vertebrae location information in DRR image to be measured is mapped to vertebra CT to be measured
In stereo-picture, the automatic positioning of lumbar vertebrae in CT image is realized；That is, the embodiment of the present invention is by combining DRR and CNN technology,
It is 2D DRR image by CT human body 3D Volume reconstruction, then the positioning of DRR lumbar vertebrae is carried out by CNN, is finally positioned by 2D DRR
Coordinate is mapped to CT 3D volume data coordinate, realizes precisely quickly CT 3D volume data lumbar vertebrae positioning, it is automatic to improve lumbar vertebrae
Robustness, real-time and the accuracy of positioning.
As shown in figure 8, can wrap the embodiment of the invention also provides the device that another kind is automatically positioned lumbar vertebrae in CT
It includes：One or more processors 101, one or more input equipments 102, one or more output equipments 103 and memory
104, above-mentioned processor 101, input equipment 102, output equipment 103 and memory 104 are connected with each other by bus 105.Storage
Device 104 is for storing computer program, and the computer program includes program instruction, and the processor 101 is configured for adjusting
With the aforementioned method being automatically positioned to lumbar vertebrae in CT of described program instruction execution.
It should be appreciated that in embodiments of the present invention, alleged processor 101 can be central processing unit (Central
Processing Unit, CPU), which can also be other general processors, digital signal processor (Digital
Signal Processor, DSP), specific integrated circuit (Application Specific Integrated Circuit,
ASIC), ready-made programmable gate array (Field-Programmable Gate Array, FPGA) or other programmable logic
Device, discrete gate or transistor logic, discrete hardware components etc..General processor can be microprocessor or this at
Reason device is also possible to any conventional processor etc..
Input equipment 102 may include keyboard etc., and output equipment 103 may include display (LCD etc.), loudspeaker etc..
The memory 104 may include read-only memory and random access memory, and to processor 101 provide instruction and
Data.The a part of of memory 104 can also include nonvolatile RAM.For example, memory 104 can also be deposited
Store up the information of device type.
In the specific implementation, processor 101 described in the embodiment of the present invention, input equipment 102, output equipment 103 can
Execute the method provided in an embodiment of the present invention being automatically positioned to lumbar vertebrae in CT.
Implement the device being automatically positioned provided by the embodiment of the present invention to lumbar vertebrae in CT, obtains lumbar vertebrae position in the training stage
The positional parameter set is predicted DRR image to be measured by positional parameter and convolutional neural networks, is obtained in forecast period
The lumbar vertebrae location information of DRR image to be measured, finally, the lumbar vertebrae location information in DRR image to be measured is mapped to vertebra CT to be measured
In stereo-picture, the automatic positioning of lumbar vertebrae in CT image is realized；That is, the embodiment of the present invention is by combining DRR and CNN technology,
It is 2D DRR image by CT human body 3D Volume reconstruction, then the positioning of DRR lumbar vertebrae is carried out by CNN, is finally positioned by 2D DRR
Coordinate is mapped to CT 3D volume data coordinate, realizes precisely quickly CT 3D volume data lumbar vertebrae positioning, it is automatic to improve lumbar vertebrae
Robustness, real-time and the accuracy of positioning.
Correspondingly, the embodiment of the invention provides a kind of computer readable storage medium, the computer-readable storage mediums
Matter is stored with computer program, and the computer program includes program instruction, realization when described program instruction is executed by processor：
The method that lumbar vertebrae in CT is automatically positioned.
Those of ordinary skill in the art may be aware that list described in conjunction with the examples disclosed in the embodiments of the present disclosure
Member and algorithm steps, can be realized with electronic hardware, computer software, or a combination of the two, in order to clearly demonstrate hardware
With the interchangeability of software, each exemplary composition and step are generally described according to function in the above description.This
A little functions are implemented in hardware or software actually, the specific application and design constraint depending on technical solution.Specially
Industry technical staff can use different methods to achieve the described function each specific application, but this realization is not
It is considered as beyond the scope of this invention.
In several embodiments provided herein, it should be understood that disclosed device and method can pass through it
Its mode is realized.For example, the apparatus embodiments described above are merely exemplary, for example, the division of the unit, only
Only a kind of logical function partition, there may be another division manner in actual implementation, such as multiple units or components can be tied
Another system is closed or is desirably integrated into, or some features can be ignored or not executed.In addition, shown or discussed phase
Mutually between coupling, direct-coupling or communication connection can be through some interfaces, the INDIRECT COUPLING or communication of device or unit
Connection is also possible to electricity, mechanical or other form connections.
The unit as illustrated by the separation member may or may not be physically separated, aobvious as unit
The component shown may or may not be physical unit, it can and it is in one place, or may be distributed over multiple
In network unit.Some or all of unit therein can be selected to realize the embodiment of the present invention according to the actual needs
Purpose.
It, can also be in addition, the functional units in various embodiments of the present invention may be integrated into one processing unit
It is that each unit physically exists alone, is also possible to two or more units and is integrated in one unit.It is above-mentioned integrated
Unit both can take the form of hardware realization, can also realize in the form of software functional units.
If the integrated unit is realized in the form of SFU software functional unit and sells or use as independent product
When, it can store in a computer readable storage medium.Based on this understanding, technical solution of the present invention is substantially
The all or part of the part that contributes to existing technology or the technical solution can be in the form of software products in other words
It embodies, which is stored in a storage medium, including some instructions are used so that a computer
Equipment (can be personal computer, server or the network equipment etc.) executes the complete of each embodiment the method for the present invention
Portion or part steps.And storage medium above-mentioned includes：USB flash disk, mobile hard disk, read-only memory (ROM, Read-Only
Memory), random access memory (RAM, Random Access Memory), magnetic or disk etc. are various can store journey
The medium of sequence code.
The above description is merely a specific embodiment, but scope of protection of the present invention is not limited thereto, any
Those familiar with the art in the technical scope disclosed by the present invention, can readily occur in various equivalent modifications or replace
It changes, these modifications or substitutions should be covered by the protection scope of the present invention.Therefore, protection scope of the present invention should be with right
It is required that protection scope subject to.
Claims (10)
1. the method that lumbar vertebrae is automatically positioned in a kind of couple of CT, which is characterized in that including following：
Training step：Initial vertebra CT stereo-picture is acquired, the initial vertebra CT stereo-picture is handled to obtain waist
The positional parameter of vertebra position；
Prediction steps：
Acquire vertebra CT stereo-picture to be measured；
Projection process is carried out to the vertebra CT stereo-picture to be measured using digital reconstruction irradiation image technology and geometrical relationship, is obtained
To DRR image to be measured；
The DRR image to be measured is predicted using convolutional neural networks and positional parameter, to obtain the DRR image to be measured
In lumbar vertebrae location information；
Using digital reconstruction irradiation image technology, the lumbar vertebrae location information in the DRR image to be measured is mapped to described to be measured
In vertebra CT stereo-picture, to obtain the lumbar vertebrae location information in vertebra CT stereo-picture to be measured.
2. the method being automatically positioned as described in claim 1 to lumbar vertebrae in CT, which is characterized in that vertical to the initial vertebra CT
Body image is handled to obtain the positional parameter of lumbar vertebrae position, is specifically included：
Projection process is carried out to the initial vertebra CT stereo-picture using digital reconstruction irradiation image technology and geometrical relationship, is obtained
To initial DRR image；
Lumbar vertebrae position in the initial DRR image is labeled, and records width, the height of every initial DRR image
Degree and the boundary coordinate up and down of lumbar vertebrae position, form training dataset；
The initial DRR image and training dataset are trained using convolutional neural networks, to obtain the positional parameter.
3. the method being automatically positioned as described in claim 1 to lumbar vertebrae in CT, which is characterized in that vertical to the initial vertebra CT
Body image is handled to obtain the positional parameter of lumbar vertebrae position, is specifically included：
Projection process is carried out to the initial vertebra CT stereo-picture using digital reconstruction irradiation image technology and geometrical relationship, is obtained
To initial DRR image, the initial DRR image is single channel image；
Single pass DRR image is converted into triple channel RGB image；
Lumbar vertebrae position in the initial DRR image is labeled, and records width, the height of every initial DRR image
Degree and the boundary coordinate up and down of lumbar vertebrae position, form training dataset；
By after conversion initial DRR image and training dataset input convolutional neural networks be trained, to obtain the positioning
Parameter.
4. the method being automatically positioned as described in claim 1 to lumbar vertebrae in CT, which is characterized in that the DRR image packet to be measured
Two are included, wherein one corresponds to the coronal-plane of the CT image to be measured, another corresponds to the sagittal of the CT image to be measured
Face.
5. the device that lumbar vertebrae is automatically positioned in a kind of couple of CT, which is characterized in that including：
Training module is handled to obtain the initial vertebra CT stereo-picture for acquiring initial vertebra CT stereo-picture
To the positional parameter of lumbar vertebrae position；
Prediction module is used for：
Acquire vertebra CT stereo-picture to be measured；
Projection process is carried out to the vertebra CT stereo-picture to be measured using digital reconstruction irradiation image technology and geometrical relationship, is obtained
To DRR image to be measured；
The DRR image to be measured is predicted using convolutional neural networks and positional parameter, to obtain the DRR image to be measured
In lumbar vertebrae location information；
Using digital reconstruction irradiation image technology, the lumbar vertebrae location information in the DRR image to be measured is mapped to described to be measured
In vertebra CT stereo-picture, to obtain the lumbar vertebrae location information in vertebra CT stereo-picture to be measured.
6. the device being automatically positioned as claimed in claim 5 to lumbar vertebrae in CT, which is characterized in that the prediction module is specifically used
In：
Projection process is carried out to the initial vertebra CT stereo-picture using digital reconstruction irradiation image technology and geometrical relationship, is obtained
To initial DRR image；
Lumbar vertebrae position in the initial DRR image is labeled, and records width, the height of every initial DRR image
Degree and the boundary coordinate up and down of lumbar vertebrae position, form training dataset；
The initial DRR image and training dataset are trained using convolutional neural networks, to obtain the positional parameter.
7. the device being automatically positioned as claimed in claim 5 to lumbar vertebrae in CT, which is characterized in that the prediction module is specifically used
In：
Projection process is carried out to the initial vertebra CT stereo-picture using digital reconstruction irradiation image technology and geometrical relationship, is obtained
To initial DRR image, the initial DRR image is single channel image；
Single pass DRR image is converted into triple channel RGB image；
Lumbar vertebrae position in the initial DRR image is labeled, and records width, the height of every initial DRR image
Degree and the boundary coordinate up and down of lumbar vertebrae position, form training dataset；
By after conversion initial DRR image and training dataset input convolutional neural networks be trained, to obtain the positioning
Parameter.
8. the device being automatically positioned as claimed in claim 5 to lumbar vertebrae in CT, which is characterized in that the DRR image packet to be measured
Two are included, wherein one corresponds to the coronal-plane of the CT image to be measured, another corresponds to the sagittal of the CT image to be measured
Face.
9. in a kind of couple of CT lumbar vertebrae be automatically positioned device, which is characterized in that including processor, input equipment, output equipment and
Memory, the processor, input equipment, output equipment and memory are connected with each other, wherein the memory is based on storing
Calculation machine program, the computer program include program instruction, and the processor is configured for calling described program instruction, are executed
Method according to any of claims 1-4.
10. a kind of computer readable storage medium, which is characterized in that the computer-readable recording medium storage has computer journey
Sequence, the computer program include program instruction, and described program instruction executes the processor such as
The described in any item methods of claim 1-4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810517798.5A CN108846830A (en) | 2018-05-25 | 2018-05-25 | The method, apparatus and storage medium be automatically positioned to lumbar vertebrae in CT |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810517798.5A CN108846830A (en) | 2018-05-25 | 2018-05-25 | The method, apparatus and storage medium be automatically positioned to lumbar vertebrae in CT |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108846830A true CN108846830A (en) | 2018-11-20 |
Family
ID=64213536
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810517798.5A Pending CN108846830A (en) | 2018-05-25 | 2018-05-25 | The method, apparatus and storage medium be automatically positioned to lumbar vertebrae in CT |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108846830A (en) |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109859213A (en) * | 2019-01-28 | 2019-06-07 | 艾瑞迈迪科技石家庄有限公司 | Bone critical point detection method and device in joint replacement surgery |
| CN109961436A (en) * | 2019-04-04 | 2019-07-02 | 北京大学口腔医学院 | A kind of median plane construction method based on artificial nerve network model |
| CN110490156A (en) * | 2019-08-23 | 2019-11-22 | 哈尔滨理工大学 | A kind of fast vehicle detection method based on convolutional neural networks |
| CN110852991A (en) * | 2019-10-10 | 2020-02-28 | 浙江大学 | Vertebra detection method based on combination of 3D and 2D |
| CN110992243A (en) * | 2019-10-29 | 2020-04-10 | 平安科技（深圳）有限公司 | Intervertebral disc section image construction method and device, computer equipment and storage medium |
| CN112825619A (en) * | 2019-09-20 | 2021-05-21 | 博医来股份公司 | Training machine learning algorithm using digitally reconstructed radiological images |
| CN113920178A (en) * | 2021-11-09 | 2022-01-11 | 广州柏视医疗科技有限公司 | Mark point-based multi-vision 2D-3D image registration method and system |
| CN114881930A (en) * | 2022-04-07 | 2022-08-09 | 重庆大学 | 3D target detection method, device, equipment and storage medium based on dimension reduction positioning |
| CN115147673A (en) * | 2022-05-10 | 2022-10-04 | 首都医科大学附属北京友谊医院 | Image equipment quality detection method, device, equipment and storage medium |
| CN118521532A (en) * | 2024-04-26 | 2024-08-20 | 中国人民解放军空军特色医学中心 | Detection method and device based on lumbar vertebra X-ray image |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101501704A (en) * | 2006-08-11 | 2009-08-05 | 艾可瑞公司 | Image segmentation for DRR generation and image registration |
| CN101843954A (en) * | 2009-03-27 | 2010-09-29 | 三菱电机株式会社 | Patient registration system |
| CN106875376A (en) * | 2016-12-29 | 2017-06-20 | 中国科学院自动化研究所 | The construction method and lumbar vertebrae method for registering of lumbar vertebrae registration prior model |
| CN107977971A (en) * | 2017-11-09 | 2018-05-01 | 哈尔滨理工大学 | The method of vertebra positioning based on convolutional neural networks |
| CN108038860A (en) * | 2017-11-30 | 2018-05-15 | 杭州电子科技大学 | Spine segmentation method based on the full convolutional neural networks of 3D |
-
2018
- 2018-05-25 CN CN201810517798.5A patent/CN108846830A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101501704A (en) * | 2006-08-11 | 2009-08-05 | 艾可瑞公司 | Image segmentation for DRR generation and image registration |
| CN101843954A (en) * | 2009-03-27 | 2010-09-29 | 三菱电机株式会社 | Patient registration system |
| CN106875376A (en) * | 2016-12-29 | 2017-06-20 | 中国科学院自动化研究所 | The construction method and lumbar vertebrae method for registering of lumbar vertebrae registration prior model |
| CN107977971A (en) * | 2017-11-09 | 2018-05-01 | 哈尔滨理工大学 | The method of vertebra positioning based on convolutional neural networks |
| CN108038860A (en) * | 2017-11-30 | 2018-05-15 | 杭州电子科技大学 | Spine segmentation method based on the full convolutional neural networks of 3D |
Non-Patent Citations (4)
| Title |
|---|
| RAINBOW_HEAVEN: ""CNN目标检测与分割（三）：SSD详解"", 《HTTP://WWW.360DOC.COM/CONTENT/17/1013/16/42392246_694639090.SHTML》 * |
| 夏黎明等: ""深度学习技术在医学影像领域的应用"", 《协和医学杂志》 * |
| 田娟秀等: ""医学图像分析深度学习方法研究与挑战"", 《自动化学报》 * |
| 陈智强等: ""基于机器学习和几何变换的实时2D/3D脊椎配准"", 《自动化学报》 * |
Cited By (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109859213A (en) * | 2019-01-28 | 2019-06-07 | 艾瑞迈迪科技石家庄有限公司 | Bone critical point detection method and device in joint replacement surgery |
| CN109961436A (en) * | 2019-04-04 | 2019-07-02 | 北京大学口腔医学院 | A kind of median plane construction method based on artificial nerve network model |
| CN109961436B (en) * | 2019-04-04 | 2021-05-18 | 北京大学口腔医学院 | Median sagittal plane construction method based on artificial neural network model |
| CN110490156A (en) * | 2019-08-23 | 2019-11-22 | 哈尔滨理工大学 | A kind of fast vehicle detection method based on convolutional neural networks |
| CN112825619A (en) * | 2019-09-20 | 2021-05-21 | 博医来股份公司 | Training machine learning algorithm using digitally reconstructed radiological images |
| CN110852991B (en) * | 2019-10-10 | 2022-08-09 | 浙江大学 | Vertebra detection method based on combination of 3D and 2D |
| CN110852991A (en) * | 2019-10-10 | 2020-02-28 | 浙江大学 | Vertebra detection method based on combination of 3D and 2D |
| CN110992243A (en) * | 2019-10-29 | 2020-04-10 | 平安科技（深圳）有限公司 | Intervertebral disc section image construction method and device, computer equipment and storage medium |
| CN110992243B (en) * | 2019-10-29 | 2023-12-22 | 平安科技（深圳）有限公司 | Intervertebral disc cross-section image construction method, device, computer equipment and storage medium |
| CN113920178B (en) * | 2021-11-09 | 2022-04-12 | 广州柏视医疗科技有限公司 | Mark point-based multi-vision 2D-3D image registration method and system |
| CN113920178A (en) * | 2021-11-09 | 2022-01-11 | 广州柏视医疗科技有限公司 | Mark point-based multi-vision 2D-3D image registration method and system |
| CN114881930A (en) * | 2022-04-07 | 2022-08-09 | 重庆大学 | 3D target detection method, device, equipment and storage medium based on dimension reduction positioning |
| CN114881930B (en) * | 2022-04-07 | 2023-08-18 | 重庆大学 | 3D target detection method, device, equipment and storage medium based on dimension reduction positioning |
| CN115147673A (en) * | 2022-05-10 | 2022-10-04 | 首都医科大学附属北京友谊医院 | Image equipment quality detection method, device, equipment and storage medium |
| CN115147673B (en) * | 2022-05-10 | 2023-04-07 | 首都医科大学附属北京友谊医院 | Image equipment quality detection method, device, equipment and storage medium |
| CN118521532A (en) * | 2024-04-26 | 2024-08-20 | 中国人民解放军空军特色医学中心 | Detection method and device based on lumbar vertebra X-ray image |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108846830A (en) | The method, apparatus and storage medium be automatically positioned to lumbar vertebrae in CT | |
| CN111724904B (en) | Multi-task progressive network for patient modeling for medical scanning | |
| US10751548B2 (en) | Automated image segmentation using DCNN such as for radiation therapy | |
| CN104346821B (en) | Automatic planning for medical imaging | |
| RU2436161C2 (en) | Recording images at deformation for image-based control beam therapy | |
| US10672128B2 (en) | Online learning enhanced atlas-based auto-segmentation | |
| CN113710159A (en) | System and method for automated and interactive analysis of bone scan images to detect metastasis | |
| US9600856B2 (en) | Hybrid point-based registration | |
| CN103295234B (en) | Based on the medical image segmentation system and method for deformation surface model | |
| CN103443824A (en) | Correlated image mapping pointer | |
| US11751947B2 (en) | Soft tissue tracking using physiologic volume rendering | |
| US11423554B2 (en) | Registering a two-dimensional image with a three-dimensional image | |
| CN105813567A (en) | Three dimensional (3D) pre-scan based volumetric image data processing | |
| CN105641814A (en) | Apparatus, method, and program for processing medical image, and radiotherapy apparatus | |
| CN104603840B (en) | Dosage deformation error calculation method and system | |
| CN110910342B (en) | Analysis of skeletal trauma by using deep learning | |
| KR20240013724A (en) | Artificial Intelligence Training Using a Multipulse X-ray Source Moving Tomosynthesis Imaging System | |
| CN114092475B (en) | Focal length determining method, image labeling method, device and computer equipment | |
| KR20140061380A (en) | Image processing method | |
| CN109414234A (en) | System and method for being projected from the 3D data set generation 2D being previously generated | |
| CN107945203A (en) | PET image processing method and processing device, electronic equipment, storage medium | |
| Paysan et al. | Deep learning methods for image guidance in radiation therapy | |
| CN116848549A (en) | Detection of image structures via dimension-reduction projection | |
| EP3777686B1 (en) | Medical image processing device, medical image processing method, and program | |
| Maffei et al. | SIS epidemiological model for adaptive RT: Forecasting the parotid glands shrinkage during tomotherapy treatment |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20181120 |
|
| RJ01 | Rejection of invention patent application after publication |