Image Action Set

This example is not available for the Lua programming language.

This section contains Python code. For more information about coding in Python, see Getting Started with SAS Viya for Python and SAS Viya: System Programming Guide.

The following Python code loads a 2-D CT image of human lungs (Armato et al. 2015a, 2015b; Armato et al. 2016; Clark et al. 2013), performs watershed segmentation, and overlays the segmentation on the CT image.^[9]Figure 5 and Figure 6 show the input and output images.

# Load a medical image
imdata = s.CASTable('imdata', replace=True)
s.image.loadimages(path='lung_for_annotation.nii',
                   casout=imdata,
                  )

# Compute the gradient magnitudes, copy the original image
cvars = ['vcimage']
cvarspgm = "length vcimage varchar(*); vcimage=_image_;"
imdata.append_computed_columns(cvars, cvarspgm)
pdata = s.CASTable('pdata', replace=True)
s.biomedimage.processbiomedimages(
    images=dict(table=imdata),
    steps=[
        dict(stepparameters=dict(
            steptype='gradient_magnitude',
            gradientmagnitudeparameters=dict(
                gradienttype='basic'))),
    ],
    copyvars={'vcimage'},
    casout=pdata)

# Segment using the watershed method, copy the original image
sdata = s.CASTable('sdata', replace=True)
s.biomedimage.segmentbiomedimages(
    images=dict(table=pdata),
    method=dict(
        name='watershed',
        watershedparameters=dict(
        level=0.4)),
    copyvars={'vcimage'},
    casout=sdata)

# Export the segmentation, copy the original image
sdata_exp1 = s.CASTable('sdata_exp1', replace=True)
s.biomedimage.processbiomedimages(
    images=dict(table=sdata),
    steps=[
        dict(stepparameters=dict(
            steptype='export'))
    ],
    copyvars={'vcimage'},
    casout=sdata_exp1)

# Export the original image, copy the segmentation
cvars = ['vcimage_seg']
cvarspgm = "length vcimage_seg varchar(*); vcimage_seg=_image_;"
sdata_exp1.append_computed_columns(cvars, cvarspgm)
sdata_exp2 = s.CASTable('sdata_exp2', replace=True)
s.biomedimage.processbiomedimages(
    images=dict(table=sdata_exp1, image='vcimage'),
    steps=[
        dict(stepparameters=dict(
            steptype='export'))
    ],
    copyvars={'vcimage_seg'},
    casout=sdata_exp2)

# Do annotation
adata = s.CASTable('adata', replace=True)
s.image.annotateimages(
    images=dict(table=sdata_exp2),
    annotations=[
        dict(annotationparameters=dict(
            annotationtype='segmentation',
            image='vcimage_seg',
            transparency=75,
            colormap='jet'))
    ],
    decode=True,
    casout=adata)

Figure 5: The 2-D CT Image of Human Lungs

Figure 6: The 2-D CT Image of Human Lungs Annotated with Its Watershed Segmentation

This example is not available for the R programming language.