CN101564323B - Auxiliary equipment for diagnosing galactophore nidus based on galactophore X-ray photograph - Google Patents

Auxiliary equipment for diagnosing galactophore nidus based on galactophore X-ray photograph Download PDF

Info

Publication number
CN101564323B
CN101564323B CN2009100617759A CN200910061775A CN101564323B CN 101564323 B CN101564323 B CN 101564323B CN 2009100617759 A CN2009100617759 A CN 2009100617759A CN 200910061775 A CN200910061775 A CN 200910061775A CN 101564323 B CN101564323 B CN 101564323B
Authority
CN
China
Prior art keywords
image
image data
mammary
data
local
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.)
Expired - Fee Related
Application number
CN2009100617759A
Other languages
Chinese (zh)
Other versions
CN101564323A (en
Inventor
宋恩民
刘宏
潘宁
姜娈
朱传聪
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Huazhong University of Science and Technology
Original Assignee
Huazhong University of Science and Technology
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Huazhong University of Science and Technology filed Critical Huazhong University of Science and Technology
Priority to CN2009100617759A priority Critical patent/CN101564323B/en
Publication of CN101564323A publication Critical patent/CN101564323A/en
Application granted granted Critical
Publication of CN101564323B publication Critical patent/CN101564323B/en
Expired - Fee Related legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Landscapes

  • Apparatus For Radiation Diagnosis (AREA)

Abstract

The invention discloses auxiliary equipment for diagnosing galactophore nidus based on a galactophore X-ray photograph, which can locally digitize the galactophore X-ray photograph and conduct image processing, analysis and auxiliary diagnosis to the image obtained. The equipment comprises an image collecting device, an image processing device and an input-and-output device, wherein the image collecting device is used for photographing image data from the galactophore X-ray photograph or a display showing the galactophore X-ray photograph; the image processing device firstly conducts the required processing of removing flicker, adjusting direction, reducing resolution, regulating spherical screen distortion, adjusting gray value distribution, obtaining resolution parameter and the like to the image data photographed, then conducts nidus determination, feature extraction and micro-calcification nidus test to the local image data processed, and outputs the results to the display. The equipment can assist a clinician to effectively observe the local image of a galactophore X-ray photograph and conduct auxiliary diagnosis and analysis to the local image data of the galactophore.

Description

Breast lesion auxiliary diagnosis equipment based on mammary X-ray photography
Technical field
The invention belongs to the area of computer aided observation analysis The Application of Technology field of medical image, be specifically related to a kind of breast lesion auxiliary diagnosis equipment based on mammary X-ray photography.
Background technology
Breast carcinoma is one of modal malignancy disease of middle and aged women, has occupied the first place of women's malignant tumor in China, and the trend that rises is year by year arranged.At present the prevention of breast carcinoma is not still had good plan, early diagnosis is to reduce mortality rate and improve the valid approach of breast carcinoma cure rate.Studies show that mammary X-ray photography (breast molybdenum target X ray technology) is a kind of method that early stage clinical asymptomatic breast carcinoma is checked of effectively carrying out.
Be subjected to the restriction of image-forming principle and some technical factors, focus based on mammary X-ray photography is diagnosed the part that comes with some shortcomings, these weak points comprise: the normal galactophore tissue of (1) mammary X-ray photography middle-high density can present the high brightness close with abnormal structure, can cause the focus in the mammary X-ray photography to be difficult to find that young Asia women its focus fine and close for mammary gland tissue then more is difficult for being found; (2) quality of some image is undesirable, and is dark partially or contrast is not strong as video imaging, or image greyscale distributes undesirablely, makes that some local details only are difficult for differentiating with naked eyes in the image; (3) brightness of the overseas large stretch of background area of region of interest may produce interference to reading image; (4) other light sources of readding in the sheet environment may have a negative impact to observing image; (5) some has the characteristics of image of diagnostic significance, as the image second order square, only is difficult for observing with naked eyes.Owing to there being above-mentioned reason, making the clinician carry out the breast tumor medical diagnosis on disease by the observation mammary X-ray photography has certain difficulty, causes failing to pinpoint a disease in diagnosis and mistaken diagnosis of focus easily.
At present, existing on the market breast tumor disease computer-aided diagnosis system, can only handle digitized video, for the patient who carries the mammary X-ray photography film, the system of this type is image data information from patient's mammary X-ray photography film directly, be difficult to it is carried out data analysis and auxiliary diagnosis, as: number of patent application is that 200710052747.1 " based on the breast carcinoma computer aided diagnosing method and the system thereof of mammary X-ray photography " discloses computer carries out the auxiliary diagnosis mammary gland tumor on the basis of mammary X-ray photography digital picture principle and process, but this system is based on digital picture, under the situation of having only the mammary X-ray film, can not directly carry out auxiliary diagnosis, and the disclosure system only carried out auxiliary diagnosis to mammary gland tumor, and the microcalciffcation kitchen range that breast carcinoma is characterized does not detect (number of the microcalciffcation point that occurs in 1 square centimeter of zone is used as the important indicator of judgement breast carcinoma usually).In addition, the image data that carries out auxiliary diagnosis if desired be not equipped with on the computer of assistant diagnosis system, image data must be transferred on this computer and just can handle so, some operations of this process need and processing time, this makes the use of at present existing breast carcinoma computer-aided diagnosis system be subjected to certain restriction.Number of patent application is 200720084171.2 " a kind of computer auxiliary medical image observing system " though disclose a kind of method of passing through the image data of camera collection medical image film, and this invention system can carry out the image that collects, and picture contrast is regulated and brightness of image strengthens, also can carry out angle to the area-of-interest of the picture that encloses on the image, length, the measurement of area etc., but this invention system does not have the function of carrying out auxiliary diagnosis at the breast carcinoma focus, and this invention system only is fit to images acquired information from the medical science image film simultaneously.
Summary of the invention
In order to address the above problem, breast lesion auxiliary diagnosis equipment based on mammary X-ray photography disclosed by the invention, the purpose of its design is to provide a kind of breast carcinoma focus auxiliary diagnosis equipment easy to use for the clinician, qualitative and analyze focus zone in the mammary X-ray photography quantitatively with assist clinicians, improve clinician's the diagnosis efficiency and the objectivity of diagnosis.
Breast lesion auxiliary diagnosis equipment based on mammary X-ray photography provided by the invention, it comprises image collecting device, image processing apparatus and input/output unit; It is characterized in that: image collecting device comprises housing, gathers cover, photographic head and three acceleration sensors, gathers the front end that cover is fixed on housing, and photographic head is fixed in the housing, and three acceleration sensors are fixed on the housing rear portion;
Image processing apparatus is the embedded system that is mounted with image capture module, picture pretreatment module, auxiliary diagnosis module and customer incident processing module;
The local mammo image data acquisition mode that image capture module is selected according to the user, from the local mammo image data that photographic head is taken, gather single width or several mammary gland X sheet partial image data, and according to the information calculations that three acceleration sensors provide publish picture the picture harvester deflection angle, according to this deflection angle the view data of gathering is carried out the image direction adjustment again, if from the indicator screen images acquired, also will carry out processing such as blink-off and resolution degraded to the mammary gland partial image data of gathering from indicator screen, the local mammo image data after will handling at last transfers to the image pretreatment module;
The picture pretreatment module is carried out the data pretreatment to receiving local mammo image data, and the data pretreatment comprises to be adjusted grey value profile and obtain resolution parameter, and the pretreated image data of data is sent to the auxiliary diagnosis module;
The auxiliary diagnosis module links to each other with input/output unit with the customer incident processing module; The auxiliary diagnosis module is carried out auxiliary diagnosis to the mammary X-ray photography partial image data, and sends the auxiliary diagnosis result to input/output unit and demonstration;
The customer incident processing module is used to receive the information of user by the input/output unit input, and this information comprises that function selecting and user enclose the regional location of picture; The customer incident processing module is determined the lesions position that the user selects according to the regional location that the user encloses picture.The customer incident processing module triggers image capture module, pretreatment module and auxiliary diagnosis module according to function select signal the lesions position that the user selects is handled accordingly.
Present device adopts Embedded System Structure, and the overall structure of equipment is simple, is easy to carry, and has taken into full account environment for use.The function of present device mainly comprises local data's collection and two parts of mammary gland topography auxiliary diagnosis of mammary X-ray photography, when the local data that carries out mammary X-ray photography gathers, for the intuitive and the accuracy of the picture that guarantees to photograph, present device can be carried out level correction to the image data that collects; When the mammary X-ray photography image on the display was gathered, present device can be removed the flicker striped in the image data, thereby had avoided the influence of flicker striped to testing result.
Present device is single from image collecting function partly, the function that not only has existing systems and had, as: number of patent application is " a kind of computer auxiliary medical image observing system " disclosed image data acquiring function based on the medical image film of 200720084171.2, but also have the more not available functions of existing systems, and as: the number of suspicious microcalciffcation kitchen range point in the unit of account area, acquisition of image data and the image data automatically photographic head gathered carry out functions such as horizontal direction adjustment from the display screen that shows mammary X-ray photography.
In a word, present device can collect mammary gland partial image data clearly from mammary X-ray photography, and can carry out Flame Image Process and auxiliary diagnosis to the mammary gland partial image data of being gathered, for the clinician carries out the diagnostic message that the diagnosis of breast carcinoma focus provides usefulness, this system adopts embedded system structure simultaneously, makes that system structure is simple, be convenient to the doctor to portably use.
Description of drawings
Fig. 1 is the structural representation of present device;
Fig. 2 is the structural representation of the employed probe of the embodiment of the invention.
The specific embodiment
As shown in Figure 1, the breast lesion auxiliary diagnosis equipment based on the breast X-ray sheet of the present invention's proposition comprises image collecting device 1, image processing apparatus 2 and 3 three parts of input/output unit.
As shown in Figure 2, image collecting device 1 comprises housing 11, gathers cover 12, photographic head 13 and three acceleration sensors 14.Gather the front end that cover 12 is fixed on housing 11, photographic head 13 is fixed in the housing 11, and three acceleration sensors 14 are fixed on the rear portion of housing 11.
Image processing apparatus 2 is for being mounted with the embedded system of image capture module 21, picture pretreatment module 22, auxiliary diagnosis module 23 and customer incident processing module 24.Image capture module 21 links to each other with photographic head 13 and three acceleration sensors 14 by data wire respectively, and auxiliary diagnosis module 23 is connected with input/output unit 3 by data wire, and the customer incident processing module links by data wire and input/output unit 3.
When image collecting device 1 is attached on the mammary X-ray photography film surface or show on the indicator screen surface of mammary X-ray photography data image data, gathering cover 12 can keep photographic head 13 constant to the distance between film, gather cover 12 simultaneously and can block the exterior light photograph, make illumination that photographic head 13 receives mainly from the film viewer light source that sees through film, or the display of demonstration mammary X-ray photography data, reduced the interference of external light source.
Photographic head 13 is from the mammary X-ray photography film or show that the indicator screen of mammary X-ray photography gathers the area-of-interest view data, and the local mammo image data that will gather is transferred to image capture module 21.
Three acceleration sensors 14 can produce and export a pair of X axis corresponding with deflection angle and Y-axis to magnitude of voltage, and with output magnitude of voltage be transferred to image capture module 21, deflection angle during in order to computed image harvester 1 images acquired, promptly image collecting device 1 is with respect to the deflection angle of intrinsic horizontal direction.
Customer incident processing module 24 is used to receive the information of user by input/output unit 3 inputs, and this information comprises that function selecting and user enclose the regional location of picture.When input/output unit adopted touch screen, customer incident processing module 24 determined it is function select signal according to the screen coordinate signal, or screen regional area coordinate signal.If function select signal, customer incident processing module 24 can trigger image capture module 21, pretreatment module 22 and auxiliary diagnosis module 24 and select different data processing functions, as: local mammo image data acquisition mode is selected, the style of shooting of mammary X-ray photography is selected, analysis and diagnosis, the extraction of microcalciffcation kitchen range and the calculating of microcalciffcation kitchen range density etc. in focus zone.If screen regional area coordinate signal, customer incident processing module 24 can be determined the position of the focus that the user selects according to the regional area coordinate signal, or definite user encloses the regional location of picture on screen, and the regional location that the user encloses picture comprises: straight line, broken line and polygon closed area etc.The customer incident module can trigger each functional module further date processing is made in the position of focus, the zone that the user encloses picture according to function select signal simultaneously.
Image capture module 21 by input/output unit 3 selected local mammo image data acquisition modes, carries out the collection of partial image data in single width or several mammary X-ray photographies according to the user.Local mammo image data acquisition mode can be: image data from the mammary X-ray photography film, or from the indicator screen that shows mammary X-ray photography image data.If local mammo image data acquisition mode is an image data from the mammary X-ray photography film, then image capture module 21 is gathered the local mammary X-ray photography data of single width, if local mammo image data acquisition mode is an image data from the indicator screen that shows mammary X-ray photography, then image capture module 22 will several local mammary X-ray photography data of continuous acquisition.
Image capture module 21 carries out date processing to the data of being gathered simultaneously, and date processing comprises: image direction adjustment, resolution degraded, spherical screen distortion correction.If local mammo image data acquisition mode is an image data from the indicator screen that shows mammary X-ray photography, need that then several local mammary X-ray photography data of gathering are carried out black removal flicker striped and handle, the width of cloth that reduces does not have the local mammo image data of black flicker striped.Local mammo image data after image capture module 21 also will be handled transfers to image pretreatment module 22, carries out further date processing by 22 pairs of mammary X-ray photography partial image datas of image pretreatment module.
Image black removal flicker striped: when local mammo image data acquisition mode be from the indicator screen of demonstration mammary X-ray photography during image data, because the refreshing frequency of photographic head and the refreshing frequency of indicator screen are inconsistent, make the local galactophore image of gathering from indicator screen by photographic head have the striped that black glimmers, before view data is handled, the blackstreak in the picture must be removed.The method of black removal flicker striped can be: 1, gather one group of local mammo image data with black flicker striped on display screen; 2, by the position of black flicker striped and the width of blackstreak in the detected image, determine effective coverage and inactive area in the image, inactive area is the band of position at blackstreak place in the image, and the zone in the image beyond the blackstreak is the effective coverage; 3,, can replace with the meansigma methods of the effective coverage pixel value of all same positions in this set of image data sequence for the pixel value in the inactive area.
The position of blackstreak and the width of blackstreak can obtain in the following way: at first make up a mammary gland topography horizontal direction pixel value summing function.This argument of function is the height of image, and the dependent variable of function is under this picture altitude, on the level of function direction all pixel values and, be formulated as:
F ( y ) = Σ x = 1 w f ( x , y )
F in the following formula (y) is a mammary gland topography horizontal direction pixel value summing function, and (x y) is mammary gland topography function to f, and x, y are the image pixel coordinate, and w is the horizontal width of image.
Then, the horizontal direction pixel value summing function F of mammary gland topography (y) is asked the single order differential, maximum and minimum in obtaining between partial zones, maximum and the position between the minimum between partial zones are the position of blackstreak, and maximum and the distance between the minimum interior between partial zones are the width of blackstreak.
Image direction is adjusted: X axis that a pair of and deflection angle that three acceleration sensors 14 of image capture module 21 real-time receptions are exported is corresponding and Y-axis are to magnitude of voltage.When X axis magnitude of voltage and Y-axis when magnitude of voltage all is 0, show that three acceleration sensors 14 are parallel with horizontal plane, do not consider the direction adjustment in this case.Therefore, following narration image direction adjustment is all supposed to carry out under not parallel with the horizontal plane situation of three direction of principal axis pick offs 14.When the X axis magnitude of voltage be 0 and Y-axis when magnitude of voltage is not 0, three acceleration sensors, 14 residing directions at this moment are called intrinsic horizontal direction.When reality is used image collecting device 1 images acquired, the directions X and the Y direction magnitude of voltage of three-axis sensor 14 outputs during according to photographic images, calculate the direction of this moment three-axis sensor 14 and the deflection angle between the intrinsic horizontal direction with following formula:
Figure G2009100617759D00071
Arctan is an arctan function in the following formula.Direction when not having angular deviation between view data that image collecting device 1 is gathered and the raw image data is called the designated horizontal direction of image collecting device 1, and at this moment the angle between the direction of three acceleration sensors 14 and its intrinsic horizontal direction is called the initial deflection angle.The initial deflection angle can be dispatched from the factory preceding or the use initial stage is determined by the image of actual photographed one width of cloth zero deflection and stores at equipment, and this process is called the deflection angle calibration.For a stable image collecting device 1, this calibration process only need be done once.
When the image travel direction that is collected is adjusted, utilize the X axis of three acceleration sensors and Y-axis to magnitude of voltage earlier, the deflection angle of image collecting device 1 when computed image is gathered, obtain the poor of this deflection angle and initial deflection angle again, what obtain is exactly this image and is clapped the image anglec of rotation between the image, can calculate the rotation transformation matrix of images acquired with this image anglec of rotation, and according to this rotation transformation matrix, will be owing to the local mammary X-ray view data that deflection is arranged that the deflection of image collecting device 1 is gathered is carried out the deflection inverse transformation, in any case make image collecting device 1 deflection, image capture module 21 can both obtain the local mammary X-ray photography view data of a width of cloth zero deflection.
Spherical screen distortion correction: when photographic head has when gathering the mammary X-ray photography partial image data on the screen of certain sphere, because spherical screen makes the mammary X-ray photography partial image data content of being gathered can produce certain deformation.Do not have the partial image data of deformation in order to reduce a width of cloth, need carry out spherical screen deformation to partial image data and proofread and correct.Spherical screen deformation is proofreaied and correct and is comprised: spherical screen deformation calibration and two steps of spherical screen deformation inverse transformation.The concrete way of spherical screen deformation calibration is: the elastic material membrane that 1, will be printed on the alternate checker of dichromatism covers at the front end of gathering cover, makes photographic head can collect checker image on this thin film; 2, will gather cover and be attached on the spherical screen of the captured image data of wanting, and, make the thin film that covers gathering the cover front end that distortion take place, thereby camera collection be to the checker image with deformation because screen has certain radian; 3, utilize the relative position of the angle point of each grid in this chess and card graticule picture, and the relative position of the reality between the chessboard angle point (promptly not having the preceding position of distortion), obtain screen sphere deformation parameter, and this parameter is stored.By the sphere display screen of images acquired, this calibration process only need be done once for each.After this can utilize the deformation parameter that obtains in the calibration process, carry out spherical screen deformation by spherical screen deformation inverse transformation and proofread and correct, concrete way is: the deformation parameter that 1, utilizes calibration process to store calculates the deformation transformation matrix of spherical screen; 2, according to this deformation transformation matrix, to there be the local mammary X-ray view data of deformation to carry out the deformation inverse transformation owing to spherical screen is caused, and make image capture module 21 to obtain the local mammary X-ray photography view data that a width of cloth does not have deformation from spherical screen collection.
Resolution degraded: when on photographic head 13 is showing the indicator screen of mammary X-ray photography, gathering partial image data, might use more than a pixel and represent a pixel on the display screen.At this moment, will cause the image that the collects image resolution ratio more shown big than screen, the data that also include reflection indicator screen grid simultaneously in the local mammo image data of Cai Jiing, therefore, need carry out the resolution degraded to the mammary gland partial image data of gathering, make the resolution of the partial image data of degraded after the resolution consistent, remove the data of the indicator screen grid that contains in the partial image data simultaneously with the resolution of the partial image data of script.The specific practice of resolution degraded is: 1, the mammary gland partial image data with the screen collection duplicates portion, preserves as raw image data; 2, the mammary gland partial image data of gathering is done time domain and arrive frequency domain transform (as Fourier transformation, wavelet transformation etc.), obtain the spectral image of this image, and then remove in the image high-frequency data of reflection indicator screen grids of pixels by High frequency filter, after this do not contained the view data of screen raster again by the frequency domain to time domain conversion; 3, subtract each other with step 1 raw image data of preserving and the view data that does not contain the screen pixels grid that step 2 obtains, the view data after subtracting each other can highlight the data that reflect the indicator screen grid; 4, the distribution of record display device screen raster on image; 5, represent the data that this grid is interior with the brightest in each a grid pixel value, generate the new view data of a width of cloth, this view data is exactly the view data after the resolution degraded.
Picture pretreatment module 22 receives the mammary X-ray photography partial image data that image capture module 21 sends, and partial image data is carried out the data pretreatment, and the data pretreatment comprises: grey value profile adjustment, resolution parameter obtain.Image data after picture pretreatment module 22 will be handled sends to auxiliary diagnosis module 23, carries out auxiliary diagnosis by 23 pairs of mammary X-ray photography partial image datas of auxiliary diagnosis module.
Grey value profile is adjusted into: since the gray scale of different X line machine imaging may be different, be used to observe the X line take the photograph the film viewer brightness of sheet may skewness, be used to show that the X line takes the photograph the problems such as indicator screen brightness disunity of sheet, make that the grey value profile of the mammary X-ray photography partial image data that photographic head 13 is captured is inhomogeneous.Gather intensity profile uniform images data on the sheet for photographic head can be taken the photograph at mammary gland X, need carry out the grey value profile adjustment the view data of gathering.The grey value profile adjustment is that the statistic with the gradation of image value is distributed on each segment of gradation of image value by a certain percentage, and the way that the grey value profile adjustment is adopted can be: the gradation of image value is interval to stretch, or image greyscale histogram equalizing.By the grey value profile adjustment, make the distribution of the different mammary X-ray photography partial image data gray value that photographic head 13 collects remain unanimity.
Resolution parameter is retrieved as: for can be in the auxiliary diagnosis process geological information of calculating foci, as diameter, area etc., so need know the actual physical size of each pixel representative in the partial image data of collection.Resolution parameter obtains can be according to formula: the number of pixels that the actual physical size of each pixel representative=scale information/the regular length picture material is comprised calculates.
Image data and calculating parameter after the processing that auxiliary diagnosis module 23 reception pretreatment module 22 send carry out the analysis and the diagnosis of focus to mammary X-ray photography local data.The analysis and the diagnosis of focus comprise: the detection of the diagnosis of focus, microcalciffcation kitchen range, similar providing with reference to the case picture.
The judgement in focus zone can be " based on the breast carcinoma computer-aided diagnosis system of mammary X-ray photography " disclosed method of 200710052747.1 referring to number of patent application.The process of the judgement in the disclosed focus of above-mentioned patent zone is: 1, suspicious lump zone is cut apart accurately, and extracted the relevant serial eigenvalue of suspicious lump; 2, eigenvalue is imported grader, classification diagnosis is carried out in suspicious lump zone; 3, the segmentation result in the suspicious lump zone that computer is produced automatically is presented on the local mammary X-ray photography image of input, watches for the user.
The testing process of microcalciffcation kitchen range is: 1, the application choice filtering image strengthens the outstanding suspicious calcification kitchen range point of algorithm; 2, in image, be the subparticle shape, compare characteristic with neighborhood pixels according to calcification kitchen range point, cut apart suspicious calcification kitchen range point with high gray value; 3, extract morphological feature, gray feature and the textural characteristics of calcification kitchen range spot correlation, and the number of suspicious calcification kitchen range point in the unit of account area; 4, the calcification kitchen range point that detects is presented on the input/output unit 3, watches for the user.The concrete implementation detail of each step is as follows:
1, the algorithm for image enhancement that strengthens filtering based on selectivity is that people such as Qiang Li by Chicago University proposed (specifically referring to Qiang Li in 2003, Shusuke Sone, Kunio Doi.Selectiveenhancement filters for nodules, vessels, and airway walls in two-andthree-dimensional CT scans.Medical Physics 2003.30 (8): 2040-2051).This algorithm is the tuberosity that is applied to detecting in the lung CT image when proposing, and obtains experimental result preferably.Consider that tuberosity in the lung CT image and the microcalciffcation kitchen range point in the mammary X-ray photography have very big similarity, as them all is irregular spot distribution, brightness is higher, be submerged in easily and be difficult for resolution etc. in the background tissues, therefore the present invention will use this method and find out the dotted region with above character on the mammary X-ray photography image, promptly comprise the area-of-interest of suspicious calcification kitchen range point;
2, before realizing the cutting apart of calcification kitchen range point, the present invention at first uses based on the morphological method of Top-hat operator area-of-interest is carried out pretreatment, it is a kind of high-pass filtering operator, by selecting suitable structural elements, interested Weak target is extracted from complicated background, and then use threshold segmentation method based on maximum entropy (specifically referring to Pun T.A new method for gray levelpicture thresholding using the entropy of the histogram.Signal Process 1980; 2:233-237.) calcification kitchen range point is split from area-of-interest;
Recognizability, reliability and independency principle during 3, according to feature selection, the calcification kitchen range point of having cut apart is calculated its serial associated eigenvalue, as form characteristic of field, gray feature and textural characteristics, and calculate the number of suspicious calcification kitchen range point in the unit are according to the number that is partitioned into calcification kitchen range point in the area-of-interest;
4, the segmentation result of the calcification kitchen range point that detects and the relevant feature parameters value of calcification kitchen range point are presented on the input/output unit 3, the confession user watches.
Similar can be " a kind of mammary X-ray photography searching computer method and system " disclosed method of 200810197093.6 referring to number of patent application with reference to providing of case picture.The disclosed similar process that provides with reference to the case picture of above-mentioned patent is: 1, area-of-interest is cut apart and extracted the relevant iconology feature of several regions, the user can select the iconology feature and the relevant weight of input feature vector importance of its concern; 2, (characteristic vector is similar more to utilize the characteristic similarity measurement criterion of weighting, then image is similar more), (comprise the reference database: the region of interest area image from the mammary X-ray photography area-of-interest computer picture reference database of setting up the known diagnosis result, the corresponding sufferer essential information of area-of-interest file, the iconology feature log file that the area-of-interest that corresponding pathologic finding of area-of-interest and diagnostic result message file, computer extract is relevant) retrieves the some area-of-interests similar in to area-of-interest to be checked; 3, retrieval is obtained similar case picture and corresponding pathologic finding thereof and diagnostic result information and be presented on the input/output unit 3, for reference.
Input/output unit 3 is used to receive the mammary X-ray partial image data processing of image processing apparatus transmission and the result of auxiliary diagnosis, and shows in the mode of numeral and figure.Simultaneously input/output unit 3 can also receive the user and touch the screen position coordinate signal that is produced on display screen, and the position coordinates signal is sent to customer incident processing module 24 in the image processing apparatus 2.Input/output unit 3 can adopt display screen with touch function or other any one have a device of input and Presentation Function.
When native system used, the user at first selected local mammo image data acquisition mode by input/output unit 3, and local mammo image data acquisition mode can be: 1, directly gather from mammary X-ray photography; 2, gather from the indicator screen that shows mammary X-ray photography.After determining local mammo image data acquisition mode, image collecting device is placed on the surface of mammary X-ray photography film or display digit mammary X-ray photography indicator screen.Image capture module 21 can carry out single width or the collection of several local mammary X-ray photography data continuously according to local mammo image data acquisition mode.If local mammo image data acquisition mode is directly to gather from the mammary X-ray photography film, then take the single image data, if local mammo image data acquisition mode is directly to gather from the indicator screen that shows mammary X-ray photography, then continuous acquisition multiple image data.Image capture module 21 is handled in real time to the mammary gland local data that collects simultaneously, and data processing function comprises: image direction adjustment, resolution degraded, spherical screen distortion correction, blink-off are handled.After date processing was finished, the image data after image capture module 21 will be handled was transferred to image pretreatment module 22.The view data that image pretreatment module 22 receives after handling is carried out the data pretreatment to local mammo image data, and the data preprocessing function comprises: grey value profile adjustment, resolution parameter obtain.After the data pretreatment was finished, the local mammo image data after image pretreatment module 22 will be handled was transferred to image auxiliary diagnosis module 23.The view data that auxiliary diagnosis module 23 receives after handling, according to the auxiliary diagnosis function of user by input/output unit 3 selections, local mammo image data is carried out auxiliary diagnosis, the auxiliary diagnosis function comprises: the detection of the diagnosis of focus, microcalciffcation kitchen range, similar with reference to providing of case picture etc., auxiliary diagnosis module 23 is presented at the auxiliary diagnosis result on the input/output unit 3 simultaneously, watches for the user.
After input/output unit 3 showed the auxiliary diagnosis result, the user can also be by the lesions position in input/output unit 3 further definite local galactophore images or the regional location of circle picture.After determining lesions position and processing region, the user still can do further processing to area-of-interest by the auxiliary diagnosis function of selecting.
Embodiment:
Below enforcement of the present invention is described in further detail.This embodiment function comprises: from mammary X-ray photography film photographs local data, or from showing the display screen photographs local data of mammary X-ray photography, and, then the image data after handling is carried out the detection and similar the providing of breast lesion diagnosis, microcalciffcation kitchen range with reference to the case picture to processing such as picture carry out that image direction adjustment, resolution degraded, spherical screen distortion correction, the processing of screen blink-off, grey value profile adjustment, resolution parameter are obtained.The user can only select partial function.
Shown in figure (2), utilize metal or plastics will gather cover and be processed into front and back two parts, the front is a horn mouth, the back is a cylindrical drum, the face that two parts link to each other has a circular hole, photographic head 13 fixed placement are in cylindrical drum inside, and the adjustable focus camera lens exposes outside horn mouth from the circular hole on the face that links to each other.Photographic head is connected with image processing apparatus by data wire.The cylindrical drum rear portion is equipped with one three and quickens sensing chips 13, and chip is connected with image capture module 21 in the harvester 2 by holding wire.
Utilize box body of metal or plastic production, the inner image processing apparatus 2 of installing of box body, input/output unit 3 is installed in the top of box body.Image processing apparatus 2 is for being mounted with the embedded system of image capture module 21, picture pretreatment module 22, auxiliary diagnosis module 23 and customer incident processing module 24.Input/output unit 3 adopts the liquid crystal display screen with touch function.
Cutting white or peach balloon are made into a transparent rubber-like thin film, and stamp chequered with black and white checker on this thin film.When equipment carries out the spherical screen distortion correction, this elastic film is covered at the front end of gathering cover, and to make the checker of seal be regular.
This work process based on the breast lesion auxiliary diagnosis equipment of breast X-ray sheet is:
(1) acquisition mode of the local mammo image data of selection on input/output unit 3.The acquisition mode of local mammo image data can be: 1, directly gather from the mammary X-ray photography film; 2, gather from the indicator screen that shows mammary X-ray photography.
(2) image collecting device is placed on the surface of mammary X-ray photography film or display digit mammary X-ray photography indicator screen.
(3) after cover to be collected was placed and stablized, triggering collection module 21 was gathered single width or multiple image data.If style of shooting is: from the mammary X-ray photography film, directly gather, then gather single width mammary gland partial image data, if style of shooting is: gather from the indicator screen that shows mammary X-ray photography, then gather several mammary gland partial image datas.
(4) if style of shooting is to gather the mammary gland partial image data from the indicator screen that shows mammary X-ray photography, then acquisition module need carry out the calibration of screen sphere to the indicator screen with certain sphere, for each by the sphere display screen of images acquired, this calibration process only need be done once, and its step is as follows:
(4.1) will make the elastic film that is printed on the black and white checker and cover, and will gather to cover and be attached on the spherical screen of the captured image data of wanting at the front end of gathering cover.
(4.2) the checker image with deformation is gathered in repeating step (2)-(3).
(4.3) utilize the relative position of the angle point of each grid in this chess and card graticule picture and the preceding relative position of angle point deformation of each grid, obtain screen sphere deformation parameter, and this parameter is stored.
(5) for the partial image data that collects from screen with certain sphere, then utilize the deformation parameter that obtains in the calibration process, carry out spherical screen deformation by spherical screen deformation inverse transformation and proofread and correct, make image capture module 21 to obtain the local mammary X-ray photography view data that a width of cloth does not have deformation from spherical screen collection.
(6) if style of shooting is to gather the mammary gland partial image data from the display that shows mammary X-ray photography, then acquisition module needs one group of local mammary gland data of taking are carried out the blink-off processing, the width of cloth that reduces does not have the local mammo image data of black flicker striped, and its step is as follows:
(6.1) from the image sequence of taking, choose the picture of some uniformly, each width of cloth image is determined the position of blackstreak that this image comprises and the width of blackstreak, and the position of blackstreak is an inactive area in the marking image, and the zone in the image beyond the blackstreak is the effective coverage.
(6.2), can replace with the meansigma methods of the effective coverage pixel value of all same positions in this set of image data sequence for the pixel value in the inactive area.
(7) three acceleration sensors of image capture module 21 receptions, 14 a pair of X axis corresponding with deflection angle of transmission and Y-axis are to magnitude of voltage, calculate direction when not having angular deviation between view data that image collecting device 1 is gathered and the raw image data according to this group magnitude of voltage, and the angle between this direction and the intrinsic horizontal direction stored, these angle number of degrees are called the initial deflection angle.The calculating of initial deflection angle and storing process are called the deflection angle calibration.For a stable image collecting device 1, this calibration process only need be done once.
(8) when the image travel direction that is collected was adjusted, its step was as follows:
(8.1) utilize the X axis of three acceleration sensors and Y-axis to magnitude of voltage, the deflection angle of image collecting device 1 when computed image is gathered.
(8.2) obtain the poor of this deflection angle and initial deflection angle again, what obtain is exactly this image and is clapped the image anglec of rotation between the image, can calculate the rotation transformation matrix of images acquired with this image anglec of rotation.
(8.3) according to the rotation transformation matrix, will be owing to the local mammary X-ray view data that deflection is arranged that the deflection of image collecting device 1 is gathered is carried out the deflection inverse transformation, in any case make image collecting device 1 deflection, image capture module 21 can both obtain the local mammary X-ray photography view data of a width of cloth zero deflection.
(9) when photographic head 13 is gathered partial image data on the indicator screen of demonstration mammary X-ray photography, need carry out the resolution degraded to the mammary gland partial image data of gathering, make the resolution of the partial image data that obtains after the degraded resolution consistent with the resolution of partial image data originally.The concrete steps of resolution degraded are as follows:
(9.1) the mammary gland partial image data with the screen collection duplicates portion, preserves as raw image data;
(9.2) the mammary gland partial image data of gathering is done the spectral image that Fourier transformation obtains this image, and then remove in the image high-frequency data of reflection indicator screen grids of pixels by High frequency filter, after this do not contained the view data of screen raster again by inversefouriertransform;
(9.3) subtract each other with step 1 raw image data of preserving and the view data that does not contain the screen pixels grid that step 2 obtains, the view data after subtracting each other can highlight the data that reflect the indicator screen grid;
(9.4) distribution of record display device screen raster on image;
(9.5) represent the data that this grid is interior with the brightest in each a grid pixel value, generate the new view data of a width of cloth, this view data is exactly the view data after the resolution degraded.
(10) the mammary gland partial image data after 21 of image acquisition moulds will be handled is sent to that image pretreatment module 22 is done the grey value profile adjustment and resolution parameter obtains processing, and the mammary gland partial image data after will handling simultaneously sends input/output unit 3 to, and its process is as follows:
(10.1) the mammary gland partial image data of 22 pairs of acquisition modules processing of image pretreatment module, stretch by the gradation of image value is interval, or image greyscale histogram equalizing, the statistic of the gray value of image is distributed on each segment of gradation of image value by a certain percentage.
(10.2) image pretreatment module 22 is calculated the resolution parameter of mammary gland partial image data according to formula:
Figure G2009100617759D00151
(11) the mammary gland partial image data after image pretreatment module 22 will be handled is sent to 23 of auxiliary diagnosis moulds and is further analyzed and diagnoses, and the mammary gland partial image data after will handling simultaneously is sent to input/output unit 3, and its process is as follows:
(11.1) according to the prompting of input/output unit 3, select the mammary X-ray style of shooting, style of shooting is: left breast side position is taken, left breast axle position is taken, right breast side position is taken and right breast axle position is taken.
(11.2) according to the mammary X-ray photography style of shooting of selecting, input/output unit 3 display image pretreatment module 22 are handled data later.
(11.3) select the auxiliary diagnosis function by touching input/output unit 3, the auxiliary diagnosis function comprises: the detection of the judgement of focus, microcalciffcation kitchen range and similar providing with reference to the case picture.
(11.4) auxiliary diagnosis module 23 is according to the function of selecting, and the mammary gland partial image data is analyzed and diagnosed, and the result that will diagnose is sent to input/output unit 3, and the confession user watches.
(11.5) select the local location of focus and the position of microcalciffcation kitchen range by touching input/output unit 3, and engagement step (11.3-11.4) is further analyzed and diagnoses to determined regional area.
The above is preferred embodiment of the present invention, but the present invention should not be confined to the disclosed content of this embodiment and accompanying drawing.So everyly do not break away from the equivalence of finishing under the spirit disclosed in this invention or revise, all fall into the scope of protection of the invention.

Claims (6)

1. breast lesion auxiliary diagnosis equipment based on mammary X-ray photography, it comprises image collecting device (1), image processing apparatus (2) and input/output unit (3); It is characterized in that: image collecting device (1) comprises housing (11), gathers cover (12), photographic head (13) and three acceleration sensors (14), gather the front end that cover (12) is fixed on housing (11), photographic head (13) is fixed in the housing (11), and three acceleration sensors (14) are fixed on the housing rear portion;
Image processing apparatus (2) is for being mounted with the embedded system of image capture module (21), image pretreatment module (22), auxiliary diagnosis module (23) and customer incident processing module (24);
The local mammo image data acquisition mode that image capture module (21) is selected according to the user, from the local mammo image data that photographic head (13) is taken, gather single width or several mammary X-ray photography partial image datas, and according to the information calculations that three acceleration sensors provide publish picture the picture harvester deflection angle, according to this deflection angle the view data of gathering is carried out the image direction adjustment again, if from the indicator screen images acquired, the mammary gland partial image data of gathering from indicator screen is carried out blink-off and resolution degradation treatment, and the local mammo image data after will handling at last transfers to image pretreatment module (22);
Image pretreatment module (22) is carried out the data pretreatment to receiving local mammo image data, and the data pretreatment comprises to be adjusted grey value profile and obtain resolution parameter, and the pretreated image data of data is sent to auxiliary diagnosis module (23);
Auxiliary diagnosis module (23) links to each other with input/output unit (3) with customer incident processing module (24); Auxiliary diagnosis module (23) is carried out auxiliary diagnosis to the mammary X-ray photography partial image data, and sends the auxiliary diagnosis result to input/output unit (3) and demonstration;
Customer incident processing module (24) is used to receive the information of user by input/output unit (3) input, and this information comprises that function selecting and user enclose the regional location of picture; Customer incident processing module (24) is determined the lesions position that the user selects according to the regional location that the user encloses picture; Customer incident processing module (24) triggers image capture module (21), image pretreatment module (22) and auxiliary diagnosis module (24) according to function select signal the lesions position that the user selects is handled accordingly.
2. breast lesion auxiliary diagnosis equipment according to claim 1, it is characterized in that: the X axis magnitude of voltage of setting three acceleration sensors (14) output is 0 and Y-axis is not 0 o'clock to magnitude of voltage, and at this moment the residing direction of three acceleration sensors (14) is called intrinsic horizontal direction; Direction when not having angular deviation between view data that image collecting device (1) is gathered and the raw image data is called the designated horizontal direction of image collecting device (1), and at this moment the angle between the direction of three acceleration sensors (14) and its intrinsic horizontal direction is called the initial deflection angle; Raw image data is meant mammary X-ray photography, perhaps shows the view data on the indicator screen of mammary X-ray photography; Image capture module (21) is adjusted image direction according to following process:
X axis that a pair of and deflection angle that A1 real-time three acceleration sensors of reception of step (14) are exported is corresponding and Y-axis are to magnitude of voltage;
The directions X and the Y direction magnitude of voltage of three-axis sensor (14) output when A2 goes on foot according to photographic images, adopt formula (I) to calculate the direction of this moment three-axis sensor (14) and the deflection angle between the intrinsic horizontal direction:
Figure FSB00000127658400021
A3 step utilizes the X axis of three acceleration sensors and Y-axis to magnitude of voltage, deflection angle when utilizing formula (I) computed image to be gathered between image collecting device (1) and the intrinsic horizontal direction, obtain the poor of this deflection angle and initial deflection angle again, obtain the image anglec of rotation between this image and the quilt bat image;
The A5 step utilizes this image anglec of rotation to calculate the rotation transformation matrix of images acquired, and according to this rotation transformation matrix, to obtain the local mammary X-ray photography view data of level owing to the X line image data that deflection is arranged that the deflection of image collecting device (1) is gathered are carried out the deflection inverse transformation.
3. breast lesion auxiliary diagnosis equipment according to claim 1 and 2 is characterized in that: image capture module (21) carries out the resolution degraded according to following process:
The B1 step will be duplicated portion from the mammary gland partial image data that display screen is gathered, and preserve as raw image data;
The B2 step is done time domain to frequency domain transform to the mammary gland partial image data of gathering, obtain the spectral image of this image, remove the high-frequency data of reflection indicator screen grids of pixels in the image then by High frequency filter, do not contained the view data of screen raster again by the frequency domain to time domain conversion;
B3 goes on foot the picture number that does not contain the screen pixels grid that obtains with the raw image data of B1 step preservation and B2 step and subtracts each other, and the view data that obtains highlights the data of reflection indicator screen grids of pixels;
The distribution of B4 step record display device screen pixels grid on image;
The B5 step is represented the data that this grid is interior with the brightest in each a grid pixel value, generates the new view data of a width of cloth, and this view data is exactly the view data after the resolution degraded.
4. breast lesion auxiliary diagnosis equipment according to claim 1 and 2 is characterized in that: auxiliary diagnosis module (23) is also carried out the microcalciffcation kitchen range to mammary X-ray photography local data and is detected, and detailed process is:
C1 step application choice filtering image strengthens the outstanding suspicious calcification kitchen range point of algorithm;
The C2 step is the subparticle shape, compares the characteristic with high gray value with neighborhood pixels according to calcification kitchen range point in image, cut apart suspicious calcification kitchen range point;
C3 step is extracted morphological feature, gray feature and the textural characteristics of calcification kitchen range spot correlation, and the number of suspicious calcification kitchen range point in the unit of account area;
The suspicious calcification kitchen range point that the C4 step will be detected is presented on the input/output unit.
5. breast lesion auxiliary diagnosis equipment according to claim 1 and 2, it is characterized in that: when photographic head (13) has when gathering the mammary X-ray photography partial image data on the screen of certain sphere, carry out spherical screen deformation calibration according to D1 step to the D3 step earlier, then the data of each collection carried out the spherical screen distortion correction according to D4 step to the D5 step:
The elastic material membrane that D1 step will be printed on the alternate checker of dichromatism covers at the front end of gathering cover, makes photographic head can collect checker image on this thin film;
The D2 step gathers cover and is attached on the spherical screen of the captured image data of wanting, and collects the checker image with deformation;
D3 goes on foot the relative position of the angle point of each grid in the checker image that utilizes this deformation and the preceding relative position of angle point deformation of each grid, obtains screen sphere deformation parameter, and this parameter is stored;
The D4 step utilizes the deformation parameter of calibration process storage to calculate the deformation transformation matrix of spherical screen;
The D5 step will have the local mammary X-ray view data of deformation to carry out the deformation inverse transformation owing to spherical screen is caused according to this deformation transformation matrix, make image capture module obtain the local mammary X-ray photography view data that a width of cloth does not have deformation.
6. breast lesion auxiliary diagnosis equipment according to claim 1 and 2, it is characterized in that: if local mammo image data acquisition mode is image data from the indicator screen that shows mammary X-ray photography, image capture module (21) carries out black removal flicker striped in the following manner to several local mammary X-ray photography data of gathering to be handled, and is reduced into the local mammo image data that a width of cloth does not have black flicker striped;
The E1 step is gathered one group of local mammo image data with black flicker striped on display screen;
The E2 step is determined the position of black flicker striped in the image and the width of blackstreak in the following manner:
At first make up a mammary gland topography horizontal direction pixel value summing function, this argument of function is the height of image, and the dependent variable of function is under this picture altitude, all pixel value summations on the level of function direction; Then, mammary gland topography horizontal direction pixel value summing function is asked the single order differential, and maximum and minimum in asking between partial zones, maximum and the position between the minimum between partial zones are the position of blackstreak, and maximum and the distance between the minimum interior between partial zones are the width of blackstreak;
Effective coverage and inactive area in the E3 step difference image, inactive area is the band of position at blackstreak place in the image, the zone in the image beyond the blackstreak is the effective coverage; For the pixel value in the inactive area, adopt the meansigma methods of the effective coverage pixel value of all same positions in this set of image data sequence to replace.
CN2009100617759A 2009-04-20 2009-04-20 Auxiliary equipment for diagnosing galactophore nidus based on galactophore X-ray photograph Expired - Fee Related CN101564323B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN2009100617759A CN101564323B (en) 2009-04-20 2009-04-20 Auxiliary equipment for diagnosing galactophore nidus based on galactophore X-ray photograph

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN2009100617759A CN101564323B (en) 2009-04-20 2009-04-20 Auxiliary equipment for diagnosing galactophore nidus based on galactophore X-ray photograph

Publications (2)

Publication Number Publication Date
CN101564323A CN101564323A (en) 2009-10-28
CN101564323B true CN101564323B (en) 2010-09-01

Family

ID=41280803

Family Applications (1)

Application Number Title Priority Date Filing Date
CN2009100617759A Expired - Fee Related CN101564323B (en) 2009-04-20 2009-04-20 Auxiliary equipment for diagnosing galactophore nidus based on galactophore X-ray photograph

Country Status (1)

Country Link
CN (1) CN101564323B (en)

Families Citing this family (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8571282B2 (en) * 2009-12-24 2013-10-29 Albert Davydov Method and apparatus for measuring spinal characteristics of a patient
RU2604698C2 (en) * 2011-03-16 2016-12-10 Конинклейке Филипс Н.В. Method and system for intelligent linking of medical data
CN105433965A (en) * 2014-08-14 2016-03-30 南京普爱射线影像设备有限公司 Image collecting and processing device used for X-ray machine
CN104217437A (en) * 2014-09-19 2014-12-17 西安电子科技大学 Lesion area partitioning method for prostate KVCT image
CN105956198B (en) * 2016-06-20 2019-04-26 东北大学 A kind of galactophore image searching system and method based on lesions position and content
CN109602437A (en) * 2018-12-06 2019-04-12 宁波耀通管阀科技有限公司 Controllable type nanometer focus CT machine
CN110367984B (en) * 2019-07-08 2022-11-08 中国人民解放军第四军医大学 Electrical impedance scanning imaging detection probe with gradient sensitivity

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1953708A (en) * 2004-05-14 2007-04-25 皇家飞利浦电子股份有限公司 System and method for diagnosing breast cancer
CN1988603A (en) * 2006-12-30 2007-06-27 华中科技大学 Computer auxiliary medical image observing system

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1953708A (en) * 2004-05-14 2007-04-25 皇家飞利浦电子股份有限公司 System and method for diagnosing breast cancer
CN1988603A (en) * 2006-12-30 2007-06-27 华中科技大学 Computer auxiliary medical image observing system

Also Published As

Publication number Publication date
CN101564323A (en) 2009-10-28

Similar Documents

Publication Publication Date Title
CN101564323B (en) Auxiliary equipment for diagnosing galactophore nidus based on galactophore X-ray photograph
CN102626305B (en) Image processing equipment and image processing method
CN101849836B (en) Mammary gland content rate estimating apparatus, method and recording medium
AU706993B2 (en) Computerized detection of masses and parenchymal distortions
US6246782B1 (en) System for automated detection of cancerous masses in mammograms
Giancardo et al. Textureless macula swelling detection with multiple retinal fundus images
EP2472473B1 (en) Image analysis device
Chan et al. Digitization requirements in mammography: Effects on computer‐aided detection of microcalcifications
JP3808123B2 (en) How to detect abnormal shadows
JP2001516604A (en) Method and apparatus for clinical histology
CN111091127A (en) Image detection method, network model training method and related device
US20080212864A1 (en) Method and Arrangement Relating to X-Ray Imaging
TW201347737A (en) Breast ultrasound scanning and diagnosis aid system
KR20130104553A (en) Apparatus and method for diagnosis of medical image in a frequency domain
EP2917894B1 (en) Skin image analysis
CN104812288A (en) Image processing device, image processing method, and image processing program
CN108888284A (en) Image adjusting method, device and equipment, storage medium
WO2019047365A1 (en) Medical cloud platform-based image big data analysis system and method
CN110288698B (en) Meniscus three-dimensional reconstruction system based on MRI
CN116630762B (en) Multi-mode medical image fusion method based on deep learning
EP3786881A1 (en) Image processing for stroke characterization
JP3758894B2 (en) Mammogram image diagnosis support device
WO2018176717A1 (en) Medical cloud platform-based breast screening image analysis system and method
CN108399354A (en) The method and apparatus of Computer Vision Recognition tumour
Makarovskikh et al. Automatic classification Infectious disease X-ray images based on Deep learning Algorithms

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
C14 Grant of patent or utility model
GR01 Patent grant
C17 Cessation of patent right
CF01 Termination of patent right due to non-payment of annual fee

Granted publication date: 20100901

Termination date: 20130420