Segmentation of Lungs from Chest X-Ray Images

I need to segment the lungs from some Chest X-Ray images. I need the best code for this purpose if someone can provide that. Threshold or region based segmentation is much preferable.

Answer 1

[KALYAN ACHARJYA]

There is no best code for Segmentation of Lungs from Chest X-Ray Images. The code depends on datasets or simmilar data types. The objectice of the code have to write the code, which may sutable for universal all images having simmilar types, though quite challenging, hence write the code which is best suitable for some wll verified public images multiple databases.

The segmentation approach is depends on test images, it would be better to comment more, if you can share a sample image. If the the ROI is distinctly differ from the rest part of the image, do threshold (recommended too, if it helps), otherwise with go with region growing or other segmentation methods and see theresults, you will be know which methods works for those images. There are so many approaches, request to follow the Gonzalez image processing using MATLAB book (Image Segmentation) for basic foundation.

Good Wishes!

[Rida Memon]

I've used the chest xray images from the JSRT dataset. I am unable to attach the image of that type over here. Could you provide me any universally accepted code for lung segmentation?

Answer 2

[Image Analyst]

Go here to this link Vision Bibliography to find papers on the topic.

• 20.7.2 Lungs, and Lung Cancer Image Analysis
• 20.7.2.1 Emphysema, Lung Analysis
• 20.7.2.2 Airway Tree Structure
• 20.7.2.3 Bronchoscopy Systems, Bronchial Analysis
• 20.7.2.4 Thorax, Thoracic Analysis
• 20.7.2.5 Pulmonary Nodules, Lung Nodules
• 20.7.2.6 Tuberculosis Analysis, Tuberculosis Bacilli
• 20.7.2.7 Ribs, Chest X-Rays

Pick one and code it up.

[Rida Memon]

Kindly help me remove this error from my code. Where do i need to set the dimensions? Kindly explain

CODE:

fid = fopen('JPCLN001.img','r','b'); %image from JSRT dataset

oneSlice = fread(fid, [2048 2048], '*uint16','b');

img = mat2gray(oneSlice, [0,4096]);

img1 = imrotate(img,-90);

figure(101);

imshow(img1);

fclose(fid);

colormap(gray);

title('Grayscale X-Ray');

I=wiener2(img1, [5 5]);

figure(102);

subplot(2,1,1);

imshow(I);

subplot(2,1,2);

imhist(I, 256);

a_thresh = I >= 172; % set this threshold

[labelImage, numberOfBlobs] = bwlabel(a_thresh);

props = regionprops(a_thresh,'all');

sortedSolidity = sort([props.Solidity], 'descend');

SB = sortedSolidity(1);

if SB == 1 % SB only accept solidity == 1 filter out bones

binaryImage = imbinarize(I); figure(103);

imshow(binaryImage); colormap(gray);

SE = strel('square',3);

morphologicalGradient = imsubtract(imdilate(binaryImage, SE),imerode(binaryImage, SE));

mask = imbinarize(morphologicalGradient,0.03);

SE = strel('square',2);

mask = imclose(mask, SE);

mask = imfill(mask,'holes');

mask = bwareafilt(mask,2); % control number of area show

notMask = ~mask;

mask = mask | bwpropfilt(notMask,'Area',[-Inf, 5000 - eps(5000)]);

showMaskAsOverlay(0.5,mask,'r'); % you have to download app/function showMaskAsOverlay

BW2 = imfill(binaryImage,'holes');

new_image = BW2 ;

new_image(~mask) = 0; % invert background and holes

B=bwboundaries(new_image); % can only accept 2 dimensions

figure(104);

imshow(new_image);

hold on

visboundaries(B);

end

ERROR:

Warning: Image is too big to fit on screen; displaying at 33%

> In images.internal.initSize (line 71)

In imshow (line 309)

In a (line 6)

Index exceeds matrix dimensions.

Error in a (line 21)

SB = sortedSolidity(1);

[rifat hasan]

is your code ok ,right now?