CN109785940A - A kind of method of medical image film typesetting - Google Patents

Abstract

The invention discloses a kind of methods of medical image film typesetting, comprising the following steps: obtains the image sequence information printed required for the image of various inspect-types；Obtain the ranks parameter for needing printing film；Input picture and the sequence information for reading image；Calculate the film number for needing to print；The image that each sequence needs to print is screened, to pick out the image on every row film；Film content is generated for printing.The present invention can be automatically selected to medical image and typesetting, is used for film printing for doctor or technician personnel, can fast and effeciently be handled amount of images is very big, significantly reduce the work load of staff.

Description

A kind of method of medical image film typesetting

Technical field

The present invention relates to a kind of methods of medical image film typesetting.

Background technique

Currently, the film type-setting mode that medical service organ is traditional, is all by doctor or technician's personnel's hand picking image After carry out typesetting, for printing.But all very greatly as CT and the amount of images of this kind of inspection of nuclear-magnetism, it therefore can be brought very to doctor Big work load.

Summary of the invention

The technical problem to be solved by the present invention is to provide a kind of methods of medical image film typesetting.

The technical scheme is that a kind of method of medical image film typesetting, comprising the following steps: 1) obtain various The image sequence information printed required for the image of inspect-type；2) the ranks parameter for needing printing film, respectively M are obtained Row, N column；3) input picture and the sequence information of image is read；4) the film number for needing to print is calculated, step 4a is divided into) and Step 4b): configuration condition 4a) is set, the image sequence number m for meeting print conditions is filtered out；4b) by m divided by M, if remainder It is 0, quotient is the number of film, and otherwise quotient adds 1 to be the number of film；5) image that each sequence needs to print is screened, from And the image on every row film is picked out, including step 5a) to step 5g): 5a) chain image all sequences in order；5b) count Whether the picture number for calculating the sequence is greater than the columns of film, is to go to step 5c), otherwise go to step 5d)；5c) pass through algorithm The similitude of different pictures is calculated, to select the maximum N of otherness images between image, and the N of corresponding filling to film is arranged In；It 5d) selects all images of the sequence and is filled into film respective column；5e) it is confirmed whether to have recycled completion institute orderly Column, are to go to step 5f), otherwise go to step 5g)；5f) complete film imposition layout；5g) go to step 5a), continuation follows in order Ring image all sequences；6) film content is generated for printing.

Further, in the present invention, the image obtained in step 1) is checked from CT examination, MR, DR is checked, CR is checked, MG is checked.

Further, in the present invention, step 4a) in configuration condition be white list, white list, which enumerates, needs the figure that prints As sequence names.

Further, in the present invention, white list supports fuzzy matching.

Further, in the present invention, step 5c) in the algorithm that uses for based on unsupervised machine learning algorithm, algorithm steps Suddenly include: A, input medical image, read the data of each image, do normalized；B, the size of initial K value is set, It is divided into N class；C, training is done to different classes of picture respectively, takes out different classes of one picture of the top respectively.

Compared with the prior art, the advantages of the invention are the method for medical image film typesetting provided by the invention can be right Medical image is automatically selected and typesetting, is used for film printing for doctor or technician personnel, can amount of images is very big It is fast and effeciently handled, significantly reduces the work load of staff.

Detailed description of the invention

The invention will be further described with reference to the accompanying drawings and embodiments:

Fig. 1 is the flow diagram of the method for the present invention.

Specific embodiment

Embodiment:

A kind of specific embodiment of the method for medical image film typesetting of the present invention, including following step are shown in conjunction with attached drawing It is rapid:

Step 1): obtaining the image sequence information printed required for the image of various inspect-types, and the image of acquisition is examined from CT It looks into, MR is checked, DR is checked, CR is checked, MG is checked.

Step 2: the ranks parameter that various inspect-type images need printing film, respectively M row, N column are obtained；

Step 3): input picture and the sequence information for reading image；

Step 4): the film number for needing to print is calculated, step 4a is divided into) and step 4b):

Step 4a): setting configuration condition filters out the image sequence number m for meeting print conditions, wherein configuration condition is white List, white list enumerates the image sequence title for needing to print, and white list supports fuzzy matching；

Sequence name is such as: T2WI_TRA, T1WI_TRA, FSFLAIR_TRA, lung, stan.

White list is to support fuzzy matching: such as T1WI, can filter out the image sequence of satisfactory picture to be printed Information.

Step 4b): by m divided by M, if remainder is 0, quotient is the number of film, and otherwise quotient adds 1 to be the number of film；

Step 5): the image that each sequence needs to print is screened, to pick out the image on every row film, including step 5a) To step 5g):

Step 5a): chain image all sequences in order；

Step 5b): whether the picture number for calculating the sequence is greater than the columns of film, is to go to step 5c), otherwise go to step 5d)；

Step 5c): using the similitude for calculating different pictures based on unsupervised machine learning algorithm, select difference between image Property maximum N images, the step of and corresponding filling is into the N column of film, algorithm includes:

A, medical image is inputted, the data of each image is read, does normalized；

B, the size for setting initial K value, is divided into N class；

C, training is done to different classes of picture respectively, takes out different classes of one picture of the top respectively.

Step 5d): it selects all images of the sequence and is filled into film respective column；

Step 5e): be confirmed whether to have recycled completion all sequences, be to go to step 5f), otherwise go to step 5g)；

Step 5f): complete film imposition layout；

Step 5g): go to step 5a), continue chain image all sequences in order；

Step 6): film content is generated for printing.

When the present embodiment specific works, by taking the film of 4 rows 3 column as an example, if the sequence number of input picture is 5, such as (1- 1 ..., 1-a), (2-1 ... 2-b), (3-1 ... 3-c), (4-1 ... 4-d), (5-1 ... 5-e), 5 divided by 4 remainder be 1, quotient It is 1, then the number of film is 2, if it divided by 4 remainder is 0, Shang Wei 3 that the sequence number of input picture, which is 12,12, then film Number is 3；Still by taking the sequence number of input picture is 5 as an example, (1-1 ..., 1-a) ~ (4-1 ... 4-d) it is sequentially filled first First to fourth row of film is opened, (5-1 ... 5-e) it is filled in the first row of second film, it needs to print choosing each sequence Image when, whether the picture number for calculating the sequence is greater than the columns of film, by taking first sequence as an example, if a be 2, it is small In the columns 3 of film, 2 images are stuffed entirely in respective column, if a is 6, are greater than the columns 3 of film, by using nothing Supervision machine learning algorithm chooses otherness maximum 3 from 6 images, and is filled into 3 column of film；Circulation completes institute After having sequence, film imposition layout is completed, generates film content for printing.

By using the method for medical image film typesetting of the invention, medical image can be automatically selected and be arranged Version is used for film printing for doctor or technician personnel, can fast and effeciently be handled amount of images is very big, be subtracted significantly The light work load of staff.

Certainly the above embodiments merely illustrate the technical concept and features of the present invention, and its object is to allow be familiar with technique People can understand the content of the present invention and implement it accordingly, it is not intended to limit the scope of the present invention.It is all according to this hair The modification that the Spirit Essence of bright main technical schemes is done, should be covered by the protection scope of the present invention.

Claims (5)

1. a kind of method of medical image film typesetting, which comprises the following steps:

1) image sequence information printed required for the image of various inspect-types is obtained；

2) the ranks parameter for needing printing film, respectively M row, N column are obtained；

3) input picture and the sequence information of image is read；

4) the film number for needing to print is calculated, step 4a is divided into) and step 4b):

Configuration condition 4a) is set, the image sequence number m for meeting print conditions is filtered out；

4b) by m divided by M, if remainder is 0, quotient is the number of film, and otherwise quotient adds 1 to be the number of film；

5) image that each sequence needs to print is screened, to pick out the image on every row film, including step 5a) to step 5g):

5a) chain image all sequences in order；

Whether the picture number for 5b) calculating the sequence is greater than the columns of film, is to go to step 5c), otherwise go to step 5d)；

The similitude of different pictures 5c) is calculated by algorithm, to select the maximum N of otherness images between image, and it is corresponding It fills into the N column of film；

It 5d) selects all images of the sequence and is filled into film respective column；

5e) it is confirmed whether to have recycled completion all sequences, is to go to step 5f), otherwise go to step 5g)；

5f) complete film imposition layout；

5g) go to step 5a), continue chain image all sequences in order；

6) film content is generated for printing.

2. a kind of method of medical image film typesetting according to claim 1, which is characterized in that obtained in the step 1) Image from CT examination, MR check, DR check, CR check, MG check.

3. a kind of method of medical image film typesetting according to claim 1, which is characterized in that the step 4a) in match Setting condition is white list, and the white list enumerates the image sequence title for needing to print.

4. a kind of method of medical image film typesetting according to claim 3, which is characterized in that the white list supports mould Paste matching.

5. a kind of method of medical image film typesetting according to claim 1, which is characterized in that the step 5c) in adopt Algorithm is based on unsupervised machine learning algorithm, and algorithm steps include:

A, medical image is inputted, the data of each image is read, does normalized；

B, the size for setting initial K value, is divided into N class；

C, training is done to different classes of picture respectively, takes out different classes of one picture of the top respectively.