how to develop 2D images into 3D viewing.

hi all, I have code below. Anyone know how to develop it into viewing as 3D, let say I want add thickness of that images 2.332mm. So that I can develop it into 3D images viewing.

Because if still 2D images, it cannot develop as 3D viewing.

% For binary images (png format, each pixel just have value 1 and 0.)
clc
clear all
dataSetDir = fullfile('C:\Users\Akmal\Desktop\I-131 256 28.02.2020\I-131 SPECT NEMA VALIDATION 01112019 256X256 26.09.2021 petang');
imageDir = fullfile(dataSetDir,'bnwaftersegmentation');
imds = imageDatastore(imageDir);
for i = 1:41
    %     subplot(6,7,i)
    I = readimage(imds,i);
end

% For origional images 2D (png format, each pixel have pixel count.)
dataSetDir = fullfile('C:\Users\Akmal\Desktop\I-131 256 28.02.2020\I-131 SPECT NEMA VALIDATION 01112019 256X256 26.09.2021 petang');
imageDir = fullfile(dataSetDir,'Image');
imds = imageDatastore(imageDir);
for k = 1:41
    %     subplot(6,7,i)
    J = readimage(imds,k);
end

%for viewing as 3D images
p = patch(isosurface(J, I==1));
axis equal
set(p,'FaceColor','c','EdgeColor','none');
alpha(p,1)
hold

but have error

Error using isosurface (line 72)
Isovalue must be a scalar.

sorry I ask many times, because I'm newer in Matlab Language.

Answer 1

[yanqi liu]

sir，please check the follow code to get some information

for more information,you can read the book 《计算机视觉与深度学习实战》，Thank you！

% For binary images (png format, each pixel just have value 1 and 0.)
clc
clear all
dataSetDir = fullfile('C:\Users\Akmal\Desktop\I-131 256 28.02.2020\I-131 SPECT NEMA VALIDATION 01112019 256X256 26.09.2021 petang');
imageDir = fullfile(dataSetDir,'bnwaftersegmentation');
imds = imageDatastore(imageDir);
for i = 1:41
    %     subplot(6,7,i)
    I = readimage(imds,i);
    % binary
    Is{i} = logical(I);
end
% For origional images 2D (png format, each pixel have pixel count.)
dataSetDir = fullfile('C:\Users\Akmal\Desktop\I-131 256 28.02.2020\I-131 SPECT NEMA VALIDATION 01112019 256X256 26.09.2021 petang');
imageDir = fullfile(dataSetDir,'Image');
imds = imageDatastore(imageDir);
Js = [];
for k = 1:41
    %     subplot(6,7,i)
    J = readimage(imds,k);
    if ndims(J) == 3
        J = rgb2gray(J);
    end
    % gray image
    Js(:,:,k) = double(uint8(J).*uint8(Is{k}));
end
%for viewing as 3D images
Ds = smooth3(Js);
figure
hiso = patch(isosurface(Ds,5),...
    'FaceColor',[1,.75,.65],...
    'EdgeColor','none');
hcap = patch(isocaps(Js,5),...
    'FaceColor','interp',...
    'EdgeColor','none');
colormap copper
view(45,30)
axis tight
daspect([1,1,.8])
lightangle(45,30);
set(gcf,'Renderer','zbuffer'); lighting phong
isonormals(Ds,hiso)
set(hcap,'AmbientStrength',.6)
set(hiso,'SpecularColorReflectance',0,'SpecularExponent',50)

[mohd akmal masud]

its work liu!!! Thank you so much.

But how to calculate the total volume and count each volume??

I have the code below to calculate the total volume each and total count each volume. just one click at the volume that we want..but how to apply it. Can help me?

[ymax xmax zmax]=size(Js);
clc
disp(['-At this point, you can use all the figure tools to zoom, pan'])
disp([' and rotate your object as you wish.'])
disp([' '])
disp(['-To eliminate unnecessary object, please use data cursor tool '])
disp([' to select a point of ROI.'])
disp(['-Data tip of X, Y and Z appeared, right click and select '])
disp([' ''Export Cursor Data to Workspace''.'])
disp(['-Click OK on the pop-up window then click Enter on command '])
disp([' window. '])

ROI=1;
for a=1:ROI
    a;
    pause
    init(a,1:3)=round([cursor_info.Position(2) cursor_info.Position(1) cursor_info.Position(3)]);
end
%init=[round(38/75*ymax) round(36/71*xmax) round(80/200*zmax)] ;%[y x]
[points dim]=size(init);
count=1;

disp([' ']);
disp(['-Wait until volume marching in Figure 1 is finished. '])

% disp(['Exit loop by clicking on the figure and then press Enter.'])
while count<=points
    if 1<init(count,1) && init(count,1)<ymax && 1<init(count,2) ...
            && init(count,2)<xmax && 1<init(count,3) && init(count,3)<zmax
        if RZ01(init(count,1)+1,init(count,2),init(count,3))==1
            init(end+1,1)=init(count,1)+1;
            init(end,2)=init(count,2);
            init(end,3)=init(count,3);
            RZ01(init(count,1)+1,init(count,2),init(count,3))=2;
        end
        if RZ01(init(count,1)-1,init(count,2),init(count,3))==1
            init(end+1,1)=init(count,1)-1;
            init(end,2)=init(count,2);
            init(end,3)=init(count,3);
            RZ01(init(count,1)-1,init(count,2),init(count,3))=2;
        end
        if RZ01(init(count,1),init(count,2)+1,init(count,3))==1
            init(end+1,1)=init(count,1);
            init(end,2)=init(count,2)+1;
            init(end,3)=init(count,3);
            RZ01(init(count,1),init(count,2)+1,init(count,3))=2;
        end
        if RZ01(init(count,1),init(count,2)-1,init(count,3))==1
            init(end+1,1)=init(count,1);
            init(end,2)=init(count,2)-1;
            init(end,3)=init(count,3);
            RZ01(init(count,1),init(count,2)-1,init(count,3))=2;
        end
        if RZ01(init(count,1),init(count,2),init(count,3)+1)==1
            init(end+1,1)=init(count,1);
            init(end,2)=init(count,2);
            init(end,3)=init(count,3)+1;
            RZ01(init(count,1),init(count,2),init(count,3)+1)=2;
        end
        if RZ01(init(count,1),init(count,2),init(count,3)-1)==1
            init(end+1,1)=init(count,1);
            init(end,2)=init(count,2);
            init(end,3)=init(count,3)-1;
            RZ01(init(count,1),init(count,2),init(count,3)-1)=2;
        end
    end
    count=count+1;
    [points dim]=size(init);
    
    if round(count/20)==count/20
        for b=1:points
            c(init(b,1),init(b,2),init(b,3))=1;
            
        end
        
        clf
        c(end+1,:,:)=0;
        c(:,end+1,:)=0;
        c(:,:,end+1)=0;
        
        %                  plot3(0.5,0.5,0.5);
        p = patch(isosurface(smooth3(smooth3(c)),0.3));
        axis equal
        set(p,'FaceColor','m','EdgeColor','none');
        alpha(1)
        view(3)
        axis([min(init(:,2))-3 max(init(:,2))+3 min(init(:,1))-3 max(init(:,1))+3 min(init(:,3))-3 max(init(:,3))+3]);
        camlight('right')
        grid
        drawnow
    end
    
end

PixelCount=size(find(c==1))
pxlsz=info.SpatialResolution;
ROIvolume=gp*pxlsz.^2*PixelCount(1)
TotalCPS=0;
for b=1:points
    b
    c(init(b,1),init(b,2),init(b,3))=1;
    CPSlist(b)=spect(init(b,1),init(b,2),init(b,3));
    TotalCPS=TotalCPS+double(CPSlist(b));
end
TotalCPS
CPSlist

Volume=points*pxlsz^2*gp

[yanqi liu]

sir,may be use

>> Vertices=get(hiso, 'Vertices');
>> shp = alphaShape(Vertices);
>> totalvol = volume(shp)

[mohd akmal masud]

Hi sir, this is last once. I promise this is last. If I used 3D images , is it correct my coding?

% For binary images (png format, each pixel just have value 1 and 0.)
clc
clear all
dataSetDir = fullfile('C:\Users\Akmal\Desktop\I-131 256 28.02.2020\I-131 SPECT NEMA VALIDATION 01112019 256X256 26.09.2021 petang');
imageDir = fullfile(dataSetDir,'bnwaftersegmentation');
imds = imageDatastore(imageDir);
for i = 1:40
    %     subplot(6,7,i)
    I = readimage(imds,i);
    % binary
    Is{i} = logical(I);
end


% For 3D images spect
myFolder = ('C:\Users\Akmal\Desktop\I-131 256 28.02.2020\I-131 SPECT NEMA VALIDATION 01112019 256X256 26.09.2021 petang\dcmoriextract');
filePattern = fullfile(myFolder, '*.dcm'); % Change to whatever pattern you need.
theFiles = dir(filePattern);
for K = 1 : length(theFiles)
    baseFileName = theFiles(K).name;
    fullFileName = fullfile(theFiles(K).folder, baseFileName);
    fprintf(1, 'Now reading %s\n', fullFileName);
    % Now do whatever you want with this file name,
    % such as reading it in as an image array with imread()
    P(:,:,K) = dicomread(fullFileName);
end

P = (Is{K});

[yanqi liu]

I think should use isosurface to get 3D structure

Answer 2

[yanqi liu]

use matrix to smooth data

[mohd akmal masud]

Tq liu, but still have error

Error using smooth3 (line 51)
V must be a 3D array.

[mohd akmal masud]

Hi liu,

If i change my 2D images to 3D images, how to write the coordinate binary into 3D?

% For binary images (png format, each pixel just have value 1 and 0.)
clc
clear all
dataSetDir = fullfile('C:\Users\Akmal\Desktop\I-131 256 28.02.2020\I-131 SPECT NEMA VALIDATION 01112019 256X256 26.09.2021 petang');
imageDir = fullfile(dataSetDir,'bnwaftersegmentation');
imds = imageDatastore(imageDir);
for i = 1:41
    %     subplot(6,7,i)
    I = readimage(imds,i);
    % binary
    Is{i} = logical(I);
end


% For 3D images spect
myFolder = ('C:\Users\Akmal\Desktop\I-131 256 28.02.2020\I-131 SPECT NEMA VALIDATION 01112019 256X256 26.09.2021 petang\dcmoriextract');
filePattern = fullfile(myFolder, '*.dcm'); % Change to whatever pattern you need.
theFiles = dir(filePattern);
for K = 1 : length(theFiles)
    baseFileName = theFiles(K).name;
    fullFileName = fullfile(theFiles(K).folder, baseFileName);
    fprintf(1, 'Now reading %s\n', fullFileName);
    % Now do whatever you want with this file name,
    % such as reading it in as an image array with imread()
    P(:,:,K) = dicomread(fullFileName);
end

Is it like this?

  Js(:,:,k) = double(uint8(P).*uint8(Is{K}));
end
%for viewing as 3D images
Ds = smooth3(Js);

[yanqi liu]

yes，use logical image to filter target