How to calculate left ventricle ejection fraction?

Hi all

I would like to calculated left ventricle ejection fraction LVEF of the heart from an image set. I have a series of planner images (128x128) (n =16) for one patient (derived from SPECT modality). LVEF is calculated according to the following equation:

LVEF = (Left ventricle end diastolic counts - left ventricle end systolic counts)/ Left ventricle end diastolic counts

Note:

Left ventricle end diastolic counts: is the image of the highest count in the series.

Left ventricle end systolic counts: is the image of the lowest count in the series.

In order to show count from left ventricle only, i drawn ROI around left ventricle and masked all other parts, so that the left ventricle only is visible. I called this serires LVimages.

can you help me on how to do this calculation? note that my image series is called LVimages.

[darova]

Can you show some images?

[Ahmad Alenezi]

Hi darova,

they are 128x128 images and number of images is 16. They are 16uint.

best,

Ahmad

[darova]

Can you show one example image and what you want to do with it?

[Ahmad Alenezi]

[Ahmad Alenezi]

Hi darova,

This is one frame of the series, the frames are 16.

I need to find the frame (among the 16 frames) that have the maximum counts and the frame that have the minimum count.

Best,

Ahmad

[darova]

the maximum counts and the frame that have the minimum count.

what does it mean? Intensity value? Can you draw?

[Ahmad Alenezi]

This image is a part of a nuclear medicine scan to evalute left ventricle ejection fraction. the scan is a dynamic test where singals (gamma rays) in frames differ over time. The Left ventricle (LV) is located within the ROI in this image. Each frame (like this one below) in this series correspond to a cardiac phase that have specific number of signals, which is diffrent than other frames in the same series. In order to calculate the Left ventricle ejection fraction, we have to know which frame that contains the highest number of signals (end diastole) and the frame that contains the lowest number of signals (end systole) to apply the above equation. Below is the region of Left ventricle:

[darova]

FOr eahc image you have 128x128 pixels with some values. What do you want to do with pixels? How to find green area?

[Ahmad Alenezi]

I want to know the number of pixels and pixel values within ROIs in all images.

Answer 1

[darova]

There is an example with using impoly

I = imread('img1.png');
imshow(I)
h = impoly;                 % draw polygon
msk = h.createMask;         % get pixels inside polygon (binary mask)
number = sum(msk(:));       % number of pixels inside polygon
    % create (M x N x 3) mask and get values
msk3 = uint8( repmat(msk,[1 1 3]) );
I1 = msk3 .* I;
figure
imshow(I1)