Images are array for medical image processing

IMAGES
ARE ARRAY
[PART 2:
DICOM]
Nada Fitrieyatul Hikmah

Loading Images (DICOM)
• imageio : read and save images.

Loading Images
• Slice the array by specifying values along each available dimension.

Metadata
• Images are always
acquired in a specific
context. This
information is often
referred to as
metadata.
• Accessible in image
objects through the
meta dictionary
attribute.

Plotting Images
• Matplotlib's imshow()
function displays 2D image
data.
• Many colormaps available but
ofen shown in grayscale
( cmap='gray' )
• Axis ticks and labels are
often not useful for images

N-dimensional
images are stacks of
arrays

Loading volumes directly
Imageio.volread() :
• Read multi-dimensional data
directly.
• Assemble a volume from multiple
images

Sampling and field of view
• Sampling rate: physical
space covered by each
element.
• Field of view: physical
space covered along each
axis.

Plotting multiple images at once
• plt.subplots: creates
a figure canvas with
multiple
AxesSubplots
objects.

Modifying the aspect ratio
(1)
• Many datasets do not have equal sampling
rates across all dimensions. In these
cases, we will want to stretch the pixels
along one side to account for the
differences.
• Any two dimensions of an array can form
an image, and slicing along different axes
can provide a useful perspective.
• However, unequal sampling rates can
create distorted images.
• Changing the aspect ratio can address
this by increasing the width of one of the
dimensions.

Modifying the aspect ratio (2)
• This results in a properly proportioned image.
• Failing to adjust the aspect would have resulted
in a distorted image.

Handle DICOM Files (1)
References :
(1) Albertina, B., Watson, M.,Holback, C., Jarosz, R., Kirk, S., Lee, Y., …
Lemmerman, J. (2016). Radiology Data from The Cancer Genome Atlas Lung
Adenocarcinoma [TCGA-LUAD] collection. The Cancer Imaging
Archive. http://doi.org/10.7937/K9/TCIA.2016.JGNIHEP5
(2) Clark K, Vendt B, Smith K, Freymann J, Kirby J, Koppel P, Moore S, Phillips S,
Maffitt D, Pringle M, Tarbox L, Prior F. The Cancer Imaging Archive (TCIA):
Maintaining and Operating a Public Information Repository, Journal of Digital
Imaging, Volume 26, Number 6, December, 2013, pp 1045-1057.

Handle DICOM Files (2)
Download data:
 Log in to Kaggle
 Open link :
https://www.kaggle.com/kmader/siim-
medical-images?select=dicom_dir
Click on Download

Imports
 pathlib for easy path handling
 pydicom to handle dicom files
 matplotlib for visualization
 numpy to create the 3D container

Read a single dcm file
All information
about dicom file can
be accessed.

Show dicom image
Get information about shape of
image :

3D Data
 Data of full head MRI scan :
https://zenodo.org/record/16956#.
YFMM5PtKiV5
 Ref : Lionheart, W. R. B. (2015).
An MRI DICOM data set of the
head of a normal male human aged
52 [Data set].
Zenodo. http://doi.org/10.5281/zen
odo.16956

Make list of 3D Data (1)
Use the glob function to return all items
in a directory which correspond to the
provided pattern. As in this case, the
directory only contains the DICOM files
 return all files in it ("*")

It is possible that the DICOM files are not ordered according to their actual image
position  This can be verified by inspecting the SliceLocation.

It crucial to order them  use the "SliceLocation" attribute passed to
the sorted function to identify the 2D slice position and thus order the slices.

Store 3D data in a list
Extract the actual data (pixel_arrays) from Dicome files and store in a list:

Show 3D data
Some slices of the ordered 3D volume:

NIfTI
(NEUROIMAGING INFORMATICS TECHNOLOGY
INITIATIVE)

What is NIfTI?
 An open file format for storage of medical image data (historically used for
neuroimaging – hence the name, but not restricted to neuroimaging).
 Efficiently store medical image data together with necessary metadata.
 Mainly used in research settings.
 Not a clinical standard.

Structure
 Header : Containing information mainly about image geometry (resolution,
position, orientation).
 Body : Actual image pixel data (2D, 3D, 4D, …)
 In general, easier to handle than DICOM files.
 File extension usually “.nii” or “.nii-gz” (compressed version)

How to work with NIfTI files
 Python libraries : NiBabel, SimpleITK.
 Image viewer : 3D Slicer (www.slicer.org)
 Transform DICOM to NIfTI in Python  dicom2nifti

Convert DICOM to NIfTI
Convert DICOM to NIfTI using last DICOM data :
Generating file 201_t2w_tse.nii.gz in the path_to_nifti.
 Creating complete 3D MRI Scan.

Read NIfTI files (1)
 nibabel to handle nifti files
 matplotlib to plot the brain images
 load using nib.load(path)

Image Shape and Image Pixel Data
Obtaining image pixel data  The image pixel data can be extracted using the
get_fdata() function of the nifti object.

Images are array for medical image processing

More Related Content

Similar to Images are array for medical image processing (20)

More from Nada Hikmah (6)

Recently uploaded (20)

Images are array for medical image processing