CN108876772A - A kind of Lung Cancer Images diagnostic system and method based on big data - Google Patents
A kind of Lung Cancer Images diagnostic system and method based on big data Download PDFInfo
- Publication number
- CN108876772A CN108876772A CN201810567040.2A CN201810567040A CN108876772A CN 108876772 A CN108876772 A CN 108876772A CN 201810567040 A CN201810567040 A CN 201810567040A CN 108876772 A CN108876772 A CN 108876772A
- Authority
- CN
- China
- Prior art keywords
- image
- module
- lung cancer
- big data
- profile
- 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
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
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/12—Edge-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/136—Segmentation; Edge detection involving thresholding
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/155—Segmentation; Edge detection involving morphological operators
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/194—Segmentation; Edge detection involving foreground-background segmentation
-
- 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
- G06T2207/30061—Lung
-
- 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
- G06T2207/30096—Tumor; Lesion
Abstract
The invention belongs to medical diagnosis technical fields, disclose a kind of Lung Cancer Images diagnostic system and method based on big data, image capture module carries out image segmentation using grey relevant dynamic matrix, and defect mending is carried out to bianry image with Mathematical Morphology Method, profile information is stored by chain code following, have the image outline of Single pixel edge to extract, obtains user image data information；The gas of user's exhalation is acquired by exhalation test module；The temperature data of user is detected by temperature check module.Lung Cancer Images diagnostic method strong operability based on big data of the invention, image diagnosing method easily carry out.It is widely used in practice, can be applied to other field image real time transfer.Image diagnosing method strong robustness of the invention；It can be improved the accuracy of prediction early stage cell lung cancer；Diagnostic data analytical calculation speed can be greatly improved by cloud service module simultaneously, improve diagnosis efficiency.
Description
Technical field
The invention belongs to medical diagnosis technical field more particularly to a kind of Lung Cancer Images diagnostic system based on big data and
Method.
Background technique
Currently, the prior art commonly used in the trade is such：
Lung cancer is that morbidity and mortality growth is most fast, to one of population health and the maximum malignant tumour of life threat.
Many countries all report that the morbidity and mortality of lung cancer obviously increase in the past 50 years, and male lung cancer morbidity and mortality are equal
First of all malignant tumours is accounted for, women disease incidence accounts for second, and the death rate accounts for second.The cause of disease of lung cancer is still endless so far
Complete clear, great mass of data shows that long-term a large amount of smokings have very close relationship with lung cancer.Existing research has shown that：
The probability that long-term a large amount of smokers suffer from lung cancer is 10~20 times of non-smoker, and the age for starting smoking is smaller, suffers from the several of lung cancer
Rate is higher.In addition, smoking not only directly affects my health, adverse effect also is generated to the health of surrounding population, is led
Involuntary smoker's lung cancer illness rate is caused to obviously increase.The disease incidence of city dweller's lung cancer is higher than rural area, this may be with urban atmosphere
Pollution is related with carcinogen is contained in flue dust.Therefore non-smoking should be advocated, and reinforces city environmental hygiene work.However,
Existing pulmonary cancer diagnosis image data inaccuracy；Data of lung cancer processing speed is slow simultaneously, and diagnosis efficiency is low.
Image processing techniques based on computer vision is to merge photoelectronics, computer technology, figure based on optics
As the complex art of the more modern sciences such as processing technique.It obtains its geometrical characteristic ginseng by handling the image of collected object
Number, and the extraction of objects' contour then becomes the principal element for influencing measurement accuracy.Therefore, in survey based on computer vision
In amount system, it is very important to guarantee that measurement accuracy chooses suitable contour extraction method.
The basic skills of contours extract is edge detection method, i.e., by means of airspace differential operator carry out, by by template with
Image convolution is completed.
Classical edge detection method is Local Operator method, such as gradient operator, Sobel operator, Roberts operator, Canny
Operator etc., this method has many advantages, such as to realize that simple, arithmetic speed is fast, but there are following disadvantages：The edge that detected can not
Guarantee single pixel wide, often occurs isolated or be only the continuous edge of point segment.Therefore, it is necessary to carry out micronization processes, together
When to try to connect interrupted edge pixel, contours extract could be completed in this way.Obviously, this treatment process is excessively multiple
Miscellaneous, the contour accuracy of extraction not can guarantee.In some cases, due to the influence of noise, or even the profile of image can not be extracted.
In conclusion problem of the existing technology is：
Existing pulmonary cancer diagnosis image data inaccuracy；Data of lung cancer processing speed is slow simultaneously, and diagnosis efficiency is low.
The existing Lung Cancer Images diagnostic method poor operability based on big data, image diagnosing method are complicated.In reality
Middle application is restricted.
There are anti-interferences in conventional edge detection method it is weak, precision is low the features such as, be not able to satisfy engineering in medicine image skill
The actual needs of art.
Summary of the invention
In view of the problems of the existing technology, the Lung Cancer Images diagnostic system that the present invention provides a kind of based on big data and
Method.
It is described based on big data the invention is realized in this way a kind of Lung Cancer Images diagnostic method based on big data
Lung Cancer Images diagnostic method includes the following steps：
Image capture module using grey relevant dynamic matrix carry out image segmentation, and with Mathematical Morphology Method to bianry image into
Row defect mending stores profile information by chain code following, have the image outline of Single pixel edge to extract, obtains user
Image data information；The gas of user's exhalation is acquired by exhalation test module；The body of user is detected by temperature check module
Warm data；
Central control module dispatches gas componant detection module and the collected user's gas of air bag is pumped into reaction gas chamber, makes
Gas can react in reaction gas chamber with preparatory porphyrin sensors, detect the pernicious gas ingredient of exhalation；
Screening module detects the content of YKL-40 and NSE in blood sample, analyzes and determines image lesion risk；Pass through pathology
Analysis module carries out pathological analysis to the image data of detection；
Big data computing resource is concentrated to carry out data detection signal s (t) by cloud service moduleNonlinear transformation processing, wherein
A indicates the amplitude of signal, and a (m) indicates that the symbol of signal, p (t) indicate shaping function, fcIndicate the carrier frequency of signal,Indicate the phase of signal；Detection data information is shown by display module.
Further, grey relevant dynamic matrix carry out image segmentation method in, if the gray scale interval of image f (x, y) be [Zmin,
Zmax], a threshold value Z is set in the sectiont, and Zmin<Zt<Zmax, all gray values in image is enabled to be less than or equal to ZtPicture
The new gray value of element is all 0, is greater than ZtThe new gray value of pixel be all 1, by such Threshold segmentation construct one it is defeated
Bianry image ft (x, y) out：
Wherein, gray threshold Zt；
It specifically includes：
1) the maximum gradation value Z of image is found outmaxWith minimum gradation value Zmin, enable threshold value ZtInitial value be (Zmax+Zmin)/
2；
2) according to threshold value ZtTarget and background is divided the image into, finds out the average gray value Z of the two respectivelyOAnd ZB；
3) new threshold value Z is found outt+1=(ZO+ZB)/2；
If 4) Zt=Zt+1, then gained is threshold value；Otherwise Z is enabledt=Zt+1It goes to step and 2) continues to iterate to calculate.
Further, include with the method that Mathematical Morphology Method carries out defect mending to bianry image：
If A is the bianry image of input, Morphological scale-space is carried out to image using structural element B；
A is expanded by B, is expressed as A ⊕ B, is:A ⊕ B=x | x=a+b, to certain a ∈ A and b ∈ B }；
A is corroded by B, is expressed as A Θ B, is:A Θ B=x | (x+b) ∈ A, to each b ∈ B }；
A makees opening operation by B, is expressed as A Ο B, is：A Ο B=(A Θ B) ⊕ B；
A makees closed operation by B, is expressed as AB, is:AB=(A ⊕ B) Θ B.
Further, the method for chain code following storage profile information includes：
The first step, the profile point originated using line sweep technique, recording the coordinate is Start-X and Start-Y,
And using the starting point as current point, turn second step；If cannot get profile point after scanning, then turn the 4th step；
Second step is stopped by the sequential scan of directional chain-code 8 neighborhoods adjacent with current point if encountering profile point immediately
It tracks profile and records traced into directional chain-code value, turn third step；If not encountering profile point, then "-" Contour extraction is set
Sweep starting point is set to Start-X and Start-Y, turns the first step by end mark；
Third step fills the profile point scanned with background color, and current point is set at the profile point traced into, turns second
Step；
4th step, setting all Contour extractions with "-" mark terminates；Obtain the chain code sequence List of all profile informations；Chain
The format of code sequence includes the coordinate of starting point and the value of directional chain-code, uses profile end mark between differently contoured in sequence
It separates；Recording mode includes:It is profile coordinate starting point first to a closed outline, followed by directional chain-code sequence, followed by wheel
Wide end mark；Due to the self-enclosed tracking using profile, therefore the value in sequence is recorded according to the sequence of lines of outline,
It is provided convenience for the subsequent processing of profile.Chain code following method is able to achieve a secondary tracking you can get it all profiles；
The method for have the image outline of Single pixel edge to extract includes：
1) predicted value of objective contour is calculated with Sobel operator or based on Color Space Clustering method：
2) image object profile is considered as to the collection { dl being made of N number of unit line segmenti}1,2 ..., N, for i=1,2 ..., N；
3) in C0In find and dliCorresponding position, according to C0The tangent line of middle corresponding position is as dliSampled reference value,
Generate primary collection；
4) constantly assemble to known optimum solution direction according to state transition model guidance particle, standard particle is avoided to filter
The method degenerated during wave realizes particle state transfer, and calculates the corresponding profile point set of each particle；
5) particle weights are calculated according to the observation model of foundation；
6) the parameter dl obtained with the weighted average calculation current iteration of particle collectionj (i)=(kj (i), bj (i))；
If 7)Wherein, ε=0.5 is taken, then is obtained with the weighting of particle collection
Averagely it is used as dliOtherwise 4) estimation of parameter is gone to step.
Further, screening module screening method includes：
Firstly, detecting the content of YKL-40 and NSE in blood sample；
Then, if the content of NSE is in its critical value range, and the content of YKL-40 is greater than its critical value range, then just
Step judges blood sample, and there are biggish cytopathy image risks；
If the content of NSE is greater than its critical value range, and the content of YKL-40 is equal to or less than its critical value range, then just
Step judges blood sample, and there are lesser cytopathy image risks.
Another object of the present invention is to provide described in a kind of realize based on the Lung Cancer Images diagnostic method of big data
Calculation machine program.
Another object of the present invention is to provide a kind of letters of the Lung Cancer Images diagnostic method described in realize based on big data
Cease data processing terminal.
Another object of the present invention is to provide a kind of computer readable storage mediums, including instruction, when it is in computer
When upper operation, so that computer executes the Lung Cancer Images diagnostic method based on big data.
Another object of the present invention is to provide a kind of bases of the Lung Cancer Images diagnostic method described in realize based on big data
In the Lung Cancer Images diagnostic system of big data, including：
Image capture module, exhalation test module, temperature check module, central control module, gas componant detection module,
Screening module, pathological analysis module, cloud service module, display module；
Image capture module is connect with central control module, for acquiring user image data information by camera；
Exhalation test module, connect with central control module, for acquiring the gas of user's exhalation by air bag；
Temperature check module, connect with central control module, for detecting the temperature data of user by temperature sensor；
Central control module detects mould with image capture module, exhalation test module, temperature check module, gas componant
Block, screening module, pathological analysis module, cloud service module, display module connection, work normally for controlling modules；
Gas componant detection module, connect with central control module, for the collected user's gas of air bag to be pumped into instead
Gas chamber is answered, allows gas to react in reaction gas chamber with preparatory porphyrin sensors, detects the pernicious gas ingredient of exhalation；
Screening module, connect with central control module, judges lesion risk for extracting user's analysis of blood；
Pathological analysis module, connect with central control module, for carrying out pathological analysis to the data of detection；
Cloud service module, connect with central control module, for concentrating big data computing resource to inspection by Cloud Server
Measured data is handled；
Display module is connect with central control module, for showing detection data information.
Another object of the present invention is to provide the Lung Cancer Images diagnostic systems based on big data described in a kind of be equipped with
Lung Cancer Images diagnostic device.
Advantages of the present invention and good effect are：
The present invention uses keratanase albumen (YKL-40) association neuron specificity olefinic alcohol enzyme by screening module
(NSE) two joint inspections carry out the screening of early stage cell lung cancer.Since the remolding sensitivity neuron of keratanase albumen (YKL-40) is special
Specific enolase (NSE) is high, in this way, can be improved the accuracy of prediction early stage cell lung cancer；Pass through cloud simultaneously
Service module can greatly improve diagnostic data analytical calculation speed, improve diagnosis efficiency.
Lung Cancer Images diagnostic method strong operability based on big data of the invention, image diagnosing method easily carry out.?
It is widely used in practice, can be applied to other field image real time transfer.Image diagnosing method strong robustness of the invention.
The characteristics of image-pickup method combination computer vision measurement technology of image capture module of the present invention, it is proposed that one
The practical contour extraction method of kind.This method carries out image segmentation using grey relevant dynamic matrix, and with Mathematical Morphology Method to two
It is worth image and carries out defect mending, profile information is stored by chain code following, the image outline with Single pixel edge is realized and mentions
It takes.Give the principle and implementation method of key technology.Experiment shows, compared with classical edge detection method, the method tool
There is the features such as strong interference immunity, precision is high, is able to satisfy the actual needs of engineering in medicine.
The profile that the present invention extracts has the characteristics that continuous, precision is high, single pixel wide, with classical edge detection method phase
Than the contour extraction method strong interference immunity that the present invention provides, robustness is good, can satisfy and is measured based on computer visual image
The requirement of technology.
The defects of there may be broken string, pothole, burrs in the bianry image that the present invention is obtained through Threshold segmentation, especially works as figure
There are in the case where noise as in, if directly carrying out contours extract using with defective image, will obtain including breakpoint, hair
The profile of the defects of thorn causes difficulty to being further processed for profile, or even influences measurement accuracy, it is therefore necessary to take measures
Eliminate these defects.
Mathematical morphology is a kind of non-linear filtering method, and expansion therein and erosion operation have very intuitive geometry back
Scape can make image thicken or be thinned in a certain direction, and direction depends on selected structural element.Based on this spy
Property, the present invention carries out defect mending to bianry image using Mathematical Morphology Method, to eliminate the defects of image and noise.
The present invention concentrates big data computing resource to carry out data detection signal s (t) by cloud service moduleNonlinear transformation processing, can be obtained accurate detection data, need to provide for the later period
Condition.
Detailed description of the invention
Fig. 1 is that the present invention implements the method for lung cancer diagnosis flow chart based on big data provided.
Fig. 2 is that the present invention implements the Lung Cancer Images diagnostic system structural block diagram based on big data provided.
In figure：1, image capture module；2, exhalation test module；3, temperature check module；4, central control module；5, gas
Body composition detection module；6, screening module；7, pathological analysis module；8, cloud service module；9, display module.
Specific embodiment
In order to make the objectives, technical solutions, and advantages of the present invention clearer, with reference to embodiments, to the present invention
It is further elaborated.It should be appreciated that the specific embodiments described herein are merely illustrative of the present invention, it is not used to
Limit the present invention.
As shown in Figure 1, the method for lung cancer diagnosis provided in an embodiment of the present invention based on big data includes the following steps：
S101 acquires user image data information by image capture module；User is acquired by exhalation test module to exhale
Gas out；The temperature data of user is detected by temperature check module；
S102, central control module dispatch gas componant detection module and the collected user's gas of air bag are pumped into reaction gas
Room allows gas to react in reaction gas chamber with preparatory porphyrin sensors, detects the pernicious gas ingredient of exhalation；
S103 extracts user's analysis of blood by screening module and judges lesion risk；Pass through pathological analysis module pair
The data of detection carry out pathological analysis；
S104 concentrates big data computing resource to handle detection data by cloud service module；Pass through display module
Show detection data information.
As shown in Fig. 2, the Lung Cancer Images diagnostic system provided in an embodiment of the present invention based on big data includes：Image Acquisition
Module 1, exhalation test module 2, temperature check module 3, central control module 4, gas componant detection module 5, screening module 6,
Pathological analysis module 7, cloud service module 8, display module 9.
Image capture module 1 is connect with central control module 4, for acquiring user image data information by camera；
Exhalation test module 2 is connect with central control module 4, for acquiring the gas of user's exhalation by air bag；
Temperature check module 3 is connect with central control module 4, for detecting the body temperature number of user by temperature sensor
According to；
Central control module 4 is examined with image capture module 1, exhalation test module 2, temperature check module 3, gas componant
It surveys module 5, screening module 6, pathological analysis module 7, cloud service module 8, display module 9 to connect, for controlling modules just
Often work；
Gas componant detection module 5 is connect with central control module 4, for the collected user's gas of air bag to be pumped into
Gas chamber is reacted, allows gas to react in reaction gas chamber with preparatory porphyrin sensors, detects the pernicious gas ingredient of exhalation；
Screening module 6 is connect with central control module 4, judges lesion risk for extracting user's analysis of blood；
Pathological analysis module 7 is connect with central control module 4, for carrying out pathological analysis to the data of detection；
Cloud service module 8 is connect with central control module 4, for concentrating big data computing resource pair by Cloud Server
Detection data is handled；
Display module 9 is connect with central control module 4, for showing detection data information.
The invention will be further described combined with specific embodiments below.
Lung Cancer Images diagnostic method provided in an embodiment of the present invention based on big data, includes the following steps：
Image capture module using grey relevant dynamic matrix carry out image segmentation, and with Mathematical Morphology Method to bianry image into
Row defect mending stores profile information by chain code following, have the image outline of Single pixel edge to extract, obtains user
Image data information；The gas of user's exhalation is acquired by exhalation test module；The body of user is detected by temperature check module
Warm data；
Central control module dispatches gas componant detection module and the collected user's gas of air bag is pumped into reaction gas chamber, makes
Gas can react in reaction gas chamber with preparatory porphyrin sensors, detect the pernicious gas ingredient of exhalation；
Screening module detects the content of YKL-40 and NSE in blood sample, analyzes and determines image lesion risk；Pass through pathology
Analysis module carries out pathological analysis to the image data of detection；
Big data computing resource is concentrated to carry out data detection signal s (t) by cloud service moduleNonlinear transformation processing, whereinA
Indicate the amplitude of signal, a (m) indicates that the symbol of signal, p (t) indicate shaping function, fcIndicate the carrier frequency of signal,Indicate the phase of signal；Detection data information is shown by display module.
Grey relevant dynamic matrix carries out in the method for image segmentation, if the gray scale interval of image f (x, y) is [Zmin, Zmax],
A threshold value Z is set in the sectiont, and Zmin<Zt<Zmax, all gray values in image is enabled to be less than or equal to ZtPixel new ash
Angle value is all 0, is greater than ZtThe new gray value of pixel be all 1, the two-value of an output is constructed by such Threshold segmentation
Image ft (x, y)：
Wherein, gray threshold Zt；
It specifically includes：
1) the maximum gradation value Z of image is found outmaxWith minimum gradation value Zmin, enable threshold value ZtInitial value be (Zmax+Zmin)/
2；
2) according to threshold value ZtTarget and background is divided the image into, finds out the average gray value Z of the two respectivelyOAnd ZB；
3) new threshold value Z is found outt+1=(ZO+ZB)/2；
If 4) Zt=Zt+1, then gained is threshold value；Otherwise Z is enabledt=Zt+1It goes to step and 2) continues to iterate to calculate.
Further, include with the method that Mathematical Morphology Method carries out defect mending to bianry image：
If A is the bianry image of input, Morphological scale-space is carried out to image using structural element B；
A is expanded by B, is expressed as A ⊕ B, is:A ⊕ B=x | x=a+b, to certain a ∈ A and b ∈ B }；
A is corroded by B, is expressed as A Θ B, is:A Θ B=x | (x+b) ∈ A, to each b ∈ B }；
A makees opening operation by B, is expressed as A Ο B, is：A Ο B=(A Θ B) ⊕ B；
A makees closed operation by B, is expressed as AB, is:AB=(A ⊕ B) Θ B.
Further, the method for chain code following storage profile information includes：
The first step, the profile point originated using line sweep technique, recording the coordinate is Start-X and Start-Y,
And using the starting point as current point, turn second step；If cannot get profile point after scanning, then turn the 4th step；
Second step is stopped by the sequential scan of directional chain-code 8 neighborhoods adjacent with current point if encountering profile point immediately
It tracks profile and records traced into directional chain-code value, turn third step；If not encountering profile point, then "-" Contour extraction is set
Sweep starting point is set to Start-X and Start-Y, turns the first step by end mark；
Third step fills the profile point scanned with background color, and current point is set at the profile point traced into, turns second
Step；
4th step, setting all Contour extractions with "-" mark terminates；Obtain the chain code sequence List of all profile informations；Chain
The format of code sequence includes the coordinate of starting point and the value of directional chain-code, uses profile end mark between differently contoured in sequence
It separates；Recording mode includes:It is profile coordinate starting point first to a closed outline, followed by directional chain-code sequence, followed by wheel
Wide end mark；Due to the self-enclosed tracking using profile, therefore the value in sequence is recorded according to the sequence of lines of outline,
It is provided convenience for the subsequent processing of profile.Chain code following method is able to achieve a secondary tracking you can get it all profiles；
The method for have the image outline of Single pixel edge to extract includes：
1) predicted value of objective contour is calculated with Sobel operator or based on Color Space Clustering method：
2) image object profile is considered as to the collection { dl being made of N number of unit line segmenti}1,2 ..., N, for i=1,2 ..., N；
3) in C0In find and dliCorresponding position, according to C0The tangent line of middle corresponding position is as dliSampled reference value,
Generate primary collection；
4) constantly assemble to known optimum solution direction according to state transition model guidance particle, standard particle is avoided to filter
The method degenerated during wave realizes particle state transfer, and calculates the corresponding profile point set of each particle；
5) particle weights are calculated according to the observation model of foundation；
6) the parameter dl obtained with the weighted average calculation current iteration of particle collectionj (i)=(kj (i), bj (i))；
If 7)Wherein, ε=0.5 is taken, then is obtained with the weighting of particle collection
Averagely it is used as dliOtherwise 4) estimation of parameter is gone to step.
Screening module screening method includes：
Firstly, detecting the content of YKL-40 and NSE in blood sample；
Then, if the content of NSE is in its critical value range, and the content of YKL-40 is greater than its critical value range, then just
Step judges blood sample, and there are biggish cytopathy image risks；
If the content of NSE is greater than its critical value range, and the content of YKL-40 is equal to or less than its critical value range, then just
Step judges blood sample, and there are lesser cytopathy image risks.
Sample provided by the invention is blood sample.
In the above-described embodiments, can come wholly or partly by software, hardware, firmware or any combination thereof real
It is existing.When using entirely or partly realizing in the form of a computer program product, the computer program product include one or
Multiple computer instructions.When loading on computers or executing the computer program instructions, entirely or partly generate according to
Process described in the embodiment of the present invention or function.The computer can be general purpose computer, special purpose computer, computer network
Network or other programmable devices.The computer instruction may be stored in a computer readable storage medium, or from one
Computer readable storage medium is transmitted to another computer readable storage medium, for example, the computer instruction can be from one
A web-site, computer, server or data center pass through wired (such as coaxial cable, optical fiber, Digital Subscriber Line (DSL)
Or wireless (such as infrared, wireless, microwave etc.) mode is carried out to another web-site, computer, server or data center
Transmission).The computer-readable storage medium can be any usable medium or include one that computer can access
The data storage devices such as a or multiple usable mediums integrated server, data center.The usable medium can be magnetic Jie
Matter, (for example, floppy disk, hard disk, tape), optical medium (for example, DVD) or semiconductor medium (such as solid state hard disk
SolidStateDisk (SSD)) etc..
The foregoing is merely illustrative of the preferred embodiments of the present invention, is not intended to limit the invention, all in essence of the invention
Made any modifications, equivalent replacements, and improvements etc., should all be included in the protection scope of the present invention within mind and principle.
Claims (10)
1. a kind of Lung Cancer Images diagnostic method based on big data, which is characterized in that the Lung Cancer Images based on big data are examined
Disconnected method includes the following steps：
Image capture module carries out image segmentation using grey relevant dynamic matrix, and is carried out to bianry image with Mathematical Morphology Method scarce
Repairing is fallen into, profile information is stored by chain code following, have the image outline of Single pixel edge to extract, obtains user images
Data information；The gas of user's exhalation is acquired by exhalation test module；The body temperature number of user is detected by temperature check module
According to；
Central control module dispatches gas componant detection module and the collected user's gas of air bag is pumped into reaction gas chamber, makes gas
It can be reacted in reaction gas chamber with preparatory porphyrin sensors, detect the pernicious gas ingredient of exhalation；
Screening module detects the content of YKL-40 and NSE in blood sample, analyzes and determines image lesion risk；Pass through pathological analysis
Module carries out pathological analysis to the image data of detection；
Big data computing resource is concentrated to carry out data detection signal s (t) by cloud service moduleNonlinear transformation processing, wherein
A indicates the amplitude of signal, and a (m) indicates that the symbol of signal, p (t) indicate shaping function, fcIndicate the carrier frequency of signal,Indicate the phase of signal；Detection data information is shown by display module.
2. the Lung Cancer Images diagnostic method based on big data as described in claim 1, which is characterized in that grey relevant dynamic matrix carries out
In the method for image segmentation, if the gray scale interval of image f (x, y) is [Zmin, Zmax], a threshold value is set in the section
Zt, and Zmin<Zt<Zmax, all gray values in image is enabled to be less than or equal to ZtThe new gray value of pixel be all 0, be greater than ZtPicture
The new gray value of element is all 1, and the bianry image ft (x, y) of an output is constructed by such Threshold segmentation：
Wherein, gray threshold Zt；
It specifically includes：
1) the maximum gradation value Z of image is found outmaxWith minimum gradation value Zmin, enable threshold value ZtInitial value be (Zmax+Zmin)/2；
2) according to threshold value ZtTarget and background is divided the image into, finds out the average gray value Z of the two respectivelyOAnd ZB；
3) new threshold value Z is found outt+1=(ZO+ZB)/2；
If 4) Zt=Zt+1, then gained is threshold value；Otherwise Z is enabledt=Zt+1It goes to step and 2) continues to iterate to calculate.
3. the Lung Cancer Images diagnostic method based on big data as described in claim 1, which is characterized in that with mathematical morphology side
Method to bianry image carry out defect mending method include：
If A is the bianry image of input, Morphological scale-space is carried out to image using structural element B；
A is expanded by B, is expressed as A ⊕ B, is:A ⊕ B=x | x=a+b, to certain a ∈ A and b ∈ B }；
A is corroded by B, is expressed as A Θ B, is:A Θ B=x | (x+b) ∈ A, to each b ∈ B }；
A makees opening operation by B, is expressed as A Ο B, is：A Ο B=(A Θ B) ⊕ B；
A makees closed operation by B, is expressed as AB, is:AB=(A ⊕ B) Θ B.
4. the Lung Cancer Images diagnostic method based on big data as described in claim 1, which is characterized in that chain code following storage wheel
The method of wide information includes：
The first step, the profile point originated using line sweep technique, recording the coordinate is Start-X and Start-Y, and with
The starting point is current point, turns second step；If cannot get profile point after scanning, then turn the 4th step；
Second step, if encountering profile point, stops tracking by the sequential scan of directional chain-code 8 neighborhoods adjacent with current point immediately
Profile and the traced into directional chain-code value of record, turn third step；If not encountering profile point, then setting "-" Contour extraction terminates
Mark, is set to Start-X and Start-Y for sweep starting point, turns the first step；
Third step fills the profile point scanned with background color, and current point is set at the profile point traced into, turns second step；
4th step, setting all Contour extractions with "-" mark terminates；Obtain the chain code sequence List of all profile informations；Chain code sequence
The format of column includes the coordinate of starting point and the value of directional chain-code, with profile end mark point between differently contoured in sequence
It opens；Recording mode includes:It is profile coordinate starting point first to a closed outline, followed by directional chain-code sequence, followed by profile
End mark；
The method for have the image outline of Single pixel edge to extract includes：
1) predicted value of objective contour is calculated with Sobel operator or based on Color Space Clustering method：
2) image object profile is considered as to the collection { dl being made of N number of unit line segmenti1,2 ..., N, for i=1,2 ..., N；
3) in C0In find and dliCorresponding position, according to C0The tangent line of middle corresponding position is as dliSampled reference value, generate
Primary collection；
4) constantly assemble to known optimum solution direction according to state transition model guidance particle, standard particle is avoided to filter
The method degenerated in journey realizes particle state transfer, and calculates the corresponding profile point set of each particle；
5) particle weights are calculated according to the observation model of foundation；
6) the parameter dl obtained with the weighted average calculation current iteration of particle collectionj (i)=(kj (i), bj (i))；
If 7) ‖ dlj (i)-dlj ( - i) 1‖ < ε, wherein take ε=0.5, then obtain using the weighted average of particle collection as dliParameter
4) estimation, otherwise goes to step.
5. the Lung Cancer Images diagnostic method based on big data as described in claim 1, which is characterized in that screening module screening side
Method includes：
Firstly, detecting the content of YKL-40 and NSE in blood sample；
Then, if the content of NSE is in its critical value range, and the content of YKL-40 is greater than its critical value range, then tentatively sentences
There are biggish cytopathy image risks for disconnected blood sample；
If the content of NSE is greater than its critical value range, and the content of YKL-40 is equal to or less than its critical value range, then tentatively sentences
There are lesser cytopathy image risks for disconnected blood sample.
6. a kind of computer journey for realizing the Lung Cancer Images diagnostic method described in Claims 1 to 5 any one based on big data
Sequence.
7. a kind of information data for realizing the Lung Cancer Images diagnostic method described in Claims 1 to 5 any one based on big data
Processing terminal.
8. a kind of computer readable storage medium, including instruction, when run on a computer, so that computer is executed as weighed
Benefit requires the Lung Cancer Images diagnostic method described in 1-5 any one based on big data.
9. a kind of Lung Cancer Images based on big data for realizing the Lung Cancer Images diagnostic method described in claim 1 based on big data
Diagnostic system, which is characterized in that the Lung Cancer Images diagnostic system based on big data includes：
Image capture module is connect with central control module, for acquiring user image data information by camera；
Exhalation test module, connect with central control module, for acquiring the gas of user's exhalation by air bag；
Temperature check module, connect with central control module, for detecting the temperature data of user by temperature sensor；
Central control module, with image capture module, exhalation test module, temperature check module, gas componant detection module, sieve
Module, pathological analysis module, cloud service module, display module connection are looked into, is worked normally for controlling modules；
Gas componant detection module, connect with central control module, for the collected user's gas of air bag to be pumped into reaction gas
Room allows gas to react in reaction gas chamber with preparatory porphyrin sensors, detects the pernicious gas ingredient of exhalation；
Screening module, connect with central control module, judges lesion risk for extracting user's analysis of blood；
Pathological analysis module, connect with central control module, for carrying out pathological analysis to the data of detection；
Cloud service module, connect with central control module, for concentrating big data computing resource to testing number by Cloud Server
According to being handled；
Display module is connect with central control module, for showing detection data information.
10. a kind of Lung Cancer Images diagnostic device for being equipped with the Lung Cancer Images diagnostic system described in claim 9 based on big data.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810567040.2A CN108876772B (en) | 2018-06-05 | 2018-06-05 | Lung cancer image diagnosis system and method based on big data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810567040.2A CN108876772B (en) | 2018-06-05 | 2018-06-05 | Lung cancer image diagnosis system and method based on big data |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108876772A true CN108876772A (en) | 2018-11-23 |
| CN108876772B CN108876772B (en) | 2021-10-12 |
Family
ID=64336173
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810567040.2A Active CN108876772B (en) | 2018-06-05 | 2018-06-05 | Lung cancer image diagnosis system and method based on big data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108876772B (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109724689A (en) * | 2019-02-21 | 2019-05-07 | 哈尔滨工程大学 | The measuring device and measuring method of Single bubble sound characteristics in a kind of water |
| CN110246584A (en) * | 2019-06-13 | 2019-09-17 | 吉林大学第一医院 | A kind of prediction and evaluation method and system of lung cancer sample |
| CN110533658A (en) * | 2019-09-02 | 2019-12-03 | 山东大学齐鲁医院 | Intelligent pulmonary emphysema diagnostic message processing system and method, information data processing terminal |
| CN114821114A (en) * | 2022-03-28 | 2022-07-29 | 南京业恒达智能系统股份有限公司 | Groove cutting robot image processing method based on visual system |
| CN114821114B (en) * | 2022-03-28 | 2024-04-30 | 南京业恒达智能系统有限公司 | Groove cutting robot image processing method based on vision system |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080051653A1 (en) * | 2006-08-23 | 2008-02-28 | Medison Co., Ltd. | System and method for image processing |
| CN101420991A (en) * | 2003-11-20 | 2009-04-29 | 血管技术国际股份公司 | Polymer compositions and methods for their use |
| CN103942803A (en) * | 2014-05-05 | 2014-07-23 | 北京理工大学 | SAR (Synthetic Aperture Radar) image based automatic water area detection method |
| CN106372390A (en) * | 2016-08-25 | 2017-02-01 | 姹ゅ钩 | Deep convolutional neural network-based lung cancer preventing self-service health cloud service system |
| CN106461664A (en) * | 2013-12-30 | 2017-02-22 | 斯克利普斯研究所 | Circulating tumor cell diagnostics for lung cancer |
| CN106456628A (en) * | 2014-03-19 | 2017-02-22 | 无限药品股份有限公司 | Heterocyclic compounds for use in the treatment of pi3k-gamma mediated disorders |
| CN107016665A (en) * | 2017-02-16 | 2017-08-04 | 浙江大学 | A kind of CT pulmonary nodule detection methods based on depth convolutional neural networks |
| CN107180426A (en) * | 2017-06-06 | 2017-09-19 | 西北工业大学 | Area of computer aided Lung neoplasm sorting technique based on transportable multiple-model integration |
| CN107767362A (en) * | 2017-09-01 | 2018-03-06 | 苏州侠洛信息科技有限公司 | A kind of early screening of lung cancer device based on deep learning |
| CN107765012A (en) * | 2016-08-16 | 2018-03-06 | 华明康生物科技（东莞）有限公司 | Early stage NSCLC screening method and kit |
-
2018
- 2018-06-05 CN CN201810567040.2A patent/CN108876772B/en active Active
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101420991A (en) * | 2003-11-20 | 2009-04-29 | 血管技术国际股份公司 | Polymer compositions and methods for their use |
| US20080051653A1 (en) * | 2006-08-23 | 2008-02-28 | Medison Co., Ltd. | System and method for image processing |
| CN106461664A (en) * | 2013-12-30 | 2017-02-22 | 斯克利普斯研究所 | Circulating tumor cell diagnostics for lung cancer |
| CN106456628A (en) * | 2014-03-19 | 2017-02-22 | 无限药品股份有限公司 | Heterocyclic compounds for use in the treatment of pi3k-gamma mediated disorders |
| CN103942803A (en) * | 2014-05-05 | 2014-07-23 | 北京理工大学 | SAR (Synthetic Aperture Radar) image based automatic water area detection method |
| CN107765012A (en) * | 2016-08-16 | 2018-03-06 | 华明康生物科技（东莞）有限公司 | Early stage NSCLC screening method and kit |
| CN106372390A (en) * | 2016-08-25 | 2017-02-01 | 姹ゅ钩 | Deep convolutional neural network-based lung cancer preventing self-service health cloud service system |
| CN107016665A (en) * | 2017-02-16 | 2017-08-04 | 浙江大学 | A kind of CT pulmonary nodule detection methods based on depth convolutional neural networks |
| CN107180426A (en) * | 2017-06-06 | 2017-09-19 | 西北工业大学 | Area of computer aided Lung neoplasm sorting technique based on transportable multiple-model integration |
| CN107767362A (en) * | 2017-09-01 | 2018-03-06 | 苏州侠洛信息科技有限公司 | A kind of early screening of lung cancer device based on deep learning |
Non-Patent Citations (1)
| Title |
|---|
| 刘露: "基于CT图像肺癌定性诊断若干关键问题研究", 《中国博士学位论文全文数据库 信息科技辑》 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109724689A (en) * | 2019-02-21 | 2019-05-07 | 哈尔滨工程大学 | The measuring device and measuring method of Single bubble sound characteristics in a kind of water |
| CN109724689B (en) * | 2019-02-21 | 2021-11-30 | 哈尔滨工程大学 | Device and method for measuring acoustic characteristics of single bubble in water |
| CN110246584A (en) * | 2019-06-13 | 2019-09-17 | 吉林大学第一医院 | A kind of prediction and evaluation method and system of lung cancer sample |
| CN110533658A (en) * | 2019-09-02 | 2019-12-03 | 山东大学齐鲁医院 | Intelligent pulmonary emphysema diagnostic message processing system and method, information data processing terminal |
| CN114821114A (en) * | 2022-03-28 | 2022-07-29 | 南京业恒达智能系统股份有限公司 | Groove cutting robot image processing method based on visual system |
| CN114821114B (en) * | 2022-03-28 | 2024-04-30 | 南京业恒达智能系统有限公司 | Groove cutting robot image processing method based on vision system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108876772B (en) | 2021-10-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108764286B (en) | Classification and identification method of feature points in blood vessel image based on transfer learning | |
| CN108464840B (en) | Automatic detection method and system for breast lumps | |
| Narasimha-Iyer et al. | Robust detection and classification of longitudinal changes in color retinal fundus images for monitoring diabetic retinopathy | |
| EP2973397B1 (en) | Tissue object-based machine learning system for automated scoring of digital whole slides | |
| CN110772286B (en) | System for discernment liver focal lesion based on ultrasonic contrast | |
| CN113420826B (en) | Liver focus image processing system and image processing method | |
| CN109671053A (en) | A kind of gastric cancer image identification system, device and its application | |
| CN104363815A (en) | Image processing device, image processing method, and image processing program | |
| CN108876772A (en) | A kind of Lung Cancer Images diagnostic system and method based on big data | |
| US20190261880A1 (en) | Automatic extraction of disease-specific features from doppler images | |
| CN107543788A (en) | A kind of urine erythrocyte abnormal rate detection method and system | |
| CN114332910A (en) | Human body part segmentation method for similar feature calculation of far infrared image | |
| CN113706518A (en) | Femoral head necrosis stage analysis method and system based on artificial intelligence | |
| Yan et al. | HLDnet: Novel deep learning based artificial intelligence tool fuses acetic acid and Lugol’s iodine cervicograms for accurate pre-cancer screening | |
| CN108364289A (en) | IVOCT image vulnerable plaque automatic testing methods | |
| Atiyah et al. | Brain MRI Images Segmentation Based on U-Net Architecture | |
| CN106127738B (en) | agglutination test interpretation method | |
| CN115880266B (en) | Intestinal polyp detection system and method based on deep learning | |
| Meejaroen et al. | Detection of fibrosis in liver biopsy images by using Bayesian classifier | |
| Yadav et al. | Edge Detection of Images using Prewitt Algorithm Comparing with Sobel Algorithm to Improve Accuracy | |
| CN108416769A (en) | Based on pretreated IVOCT images vulnerable plaque automatic testing method | |
| Koukounis et al. | Retinal image registration based on multiscale products and optic disc detection | |
| Yang et al. | Analysis of mitochondrial shape dynamics using large deformation diffeomorphic metric curve matching | |
| Sherwani et al. | Automated segmentation of optic disc boundary and diameter calculation using fundus imagery | |
| Riana et al. | Comparison of Segmentation Analysis in Nucleus Detection with GLCM Features using Otsu and Polynomial Methods |
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 |