Skip to content
Snippets Groups Projects
Commit 7ee8c4a3 authored by Paul McCarthy's avatar Paul McCarthy :mountain_bicyclist:
Browse files

nifti prac

parent ede2fe12
No related branches found
No related tags found
No related merge requests found
Hide whitespace changes
Inline Side-by-side
getting_started/05_nifti.ipynb
+ 26
− 19
View file @ 7ee8c4a3
%% Cell type:markdown id: tags: %% Cell type:markdown id: tags:
# NIfTI images and python # NIfTI images and python
The [`nibabel`](http://nipy.org/nibabel/) module is used to read and write NIfTI The [`nibabel`](http://nipy.org/nibabel/) module is used to read and write NIfTI
images and also some other medical imaging formats (e.g., ANALYZE, GIFTI, images and also some other medical imaging formats (e.g., ANALYZE, GIFTI,
MINC, MGH). `nibabel` is included within the FSL python environment. MINC, MGH). `nibabel` is included within the FSL python environment.
Building upon `nibabel`, the Building upon `nibabel`, the
[`fsl.data`](https://users.fmrib.ox.ac.uk/~paulmc/fsleyes/fslpy/latest/fsl.data.html#module-fsl.data) [`fslpy`](https://users.fmrib.ox.ac.uk/~paulmc/fsleyes/fslpy/latest/) library
package contains a number of FSL-specific classes and functions which you may contains a number of FSL-specific classes and functions which you may find
find useful. This is covered in a different practical useful. But let's start with `nibabel` - `fslpy` is introduced in a different
(`advanced_topics/08_fslpy.ipynb`). practical (`advanced_topics/08_fslpy.ipynb`).
## Contents ## Contents
* [Reading images](#reading-images) * [Reading images](#reading-images)
* [Header info](#header-info) * [Header info](#header-info)
* [Voxel sizes](#voxel-sizes) * [Voxel sizes](#voxel-sizes)
* [Coordinate orientations and mappings](#orientation-info) * [Coordinate orientations and mappings](#orientation-info)
* [Writing images](#writing-images) * [Writing images](#writing-images)
* [Exercise](#exercise) * [Exercise](#exercise)
--- ---
<a class="anchor" id="reading-images"></a> <a class="anchor" id="reading-images"></a>
## Reading images ## Reading images
It is easy to read an image: It is easy to read an image:
%% Cell type:code id: tags: %% Cell type:code id: tags:
``` ```
import numpy as np import numpy as np
import nibabel as nib import nibabel as nib
import os.path as op import os.path as op
filename = op.expandvars('${FSLDIR}/data/standard/MNI152_T1_1mm.nii.gz') filename = op.expandvars('${FSLDIR}/data/standard/MNI152_T1_1mm.nii.gz')
imobj = nib.load(filename, mmap=False) imobj = nib.load(filename, mmap=False)
# display header object # display header object
imhdr = imobj.header imhdr = imobj.header
# extract data (as an numpy array) # extract data (as a numpy array)
imdat = imobj.get_data().astype(float) imdat = imobj.get_fdata()
print(imdat.shape) print(imdat.shape)
``` ```
%% Cell type:markdown id: tags: %% Cell type:markdown id: tags:
> Make sure you use the full filename, including the .nii.gz extension. > Make sure you use the full filename, including the `.nii.gz` extension.
> `fslpy` provides FSL-like automatic file suffix detection though.
> We use the expandvars() function above to insert the FSLDIR > We use the `expandvars()` function above to insert the FSLDIR
> environmental variable into our string. This function is > environmental variable into our string. This function is
> discussed more fully in the file management practical. > discussed more fully in the file management practical.
Reading the data off the disk is not done until `get_data()` is called. Reading the data off the disk is not done until `get_fdata()` is called.
> Pitfall: > Pitfall:
> >
> The option `mmap=False`is necessary as turns off memory mapping, > The option `mmap=False` disables memory mapping, which would otherwise be
> which otherwise would be invoked for uncompressed NIfTI files but not for > invoked for uncompressed NIfTI files but not for compressed files. Since
> compressed files. Since some functionality behaves differently on memory > some functionality behaves differently on memory mapped objects, it is
> mapped objects, it is advisable to turn this off. > advisable to turn this off unless you specifically want it.
Once the data is read into a numpy array then it is easily manipulated. Once the data is read into a numpy array then it is easily manipulated.
> We recommend converting it to float at the start to avoid problems with > The `get_fdata` method will return floating point data, regardless of the
> integer arithmetic and overflow, though this is not compulsory. > underlying image data type. If you want the image data in the type that it
> is stored (e.g. integer ROI labels), then use
> `imdat = np.asanyarray(imobj.dataobj)` instead.
--- ---
<a class="anchor" id="header-info"></a> <a class="anchor" id="header-info"></a>
## Header info ## Header info
There are many methods available on the header object - for example, look at There are many methods available on the header object - for example, look at
`dir(imhdr)` or `help(imhdr)` or the [nibabel webpage about NIfTI `dir(imhdr)` or `help(imhdr)` or the [nibabel webpage about NIfTI
images](http://nipy.org/nibabel/nifti_images.html) images](http://nipy.org/nibabel/nifti_images.html)
<a class="anchor" id="voxel-sizes"></a> <a class="anchor" id="voxel-sizes"></a>
### Voxel sizes ### Voxel sizes
Dimensions of the voxels, in mm, can be found from: Dimensions of the voxels, in mm, can be found from:
%% Cell type:code id: tags: %% Cell type:code id: tags:
``` ```
voxsize = imhdr.get_zooms() voxsize = imhdr.get_zooms()
print(voxsize) print(voxsize)
``` ```
%% Cell type:markdown id: tags: %% Cell type:markdown id: tags:
<a class="anchor" id="orientation-info"></a> <a class="anchor" id="orientation-info"></a>
### Coordinate orientations and mappings ### Coordinate orientations and mappings
Information about the NIfTI qform and sform matrices can be extracted like this: Information about the NIfTI qform and sform matrices can be extracted like this:
%% Cell type:code id: tags: %% Cell type:code id: tags:
``` ```
sform = imhdr.get_sform() sform = imhdr.get_sform()
sformcode = imhdr['sform_code'] sformcode = imhdr['sform_code']
qform = imhdr.get_qform() qform = imhdr.get_qform()
qformcode = imhdr['qform_code'] qformcode = imhdr['qform_code']
print(qformcode) print(qformcode)
print(qform) print(qform)
``` ```
%% Cell type:markdown id: tags: %% Cell type:markdown id: tags:
You can also get both code and matrix together like this: You can also get both code and matrix together like this:
%% Cell type:code id: tags: %% Cell type:code id: tags:
``` ```
affine, code = imhdr.get_qform(coded=True) affine, code = imhdr.get_qform(coded=True)
print(affine, code) print(affine, code)
``` ```
%% Cell type:markdown id: tags: %% Cell type:markdown id: tags:
--- ---
<a class="anchor" id="writing-images"></a> <a class="anchor" id="writing-images"></a>
## Writing images ## Writing images
If you have created a modified image by making or modifying a numpy array then If you have created a modified image by making or modifying a numpy array then
you need to put this into a NIfTI image object in order to save it to a file. you need to put this into a NIfTI image object in order to save it to a file.
The easiest way to do this is to copy all the header info from an existing The easiest way to do this is to copy all the header info from an existing
image like this: image like this:
%% Cell type:code id: tags: %% Cell type:code id: tags:
``` ```
newdata = imdat * imdat newdata = imdat * imdat
newhdr = imhdr.copy() newhdr = imhdr.copy()
newobj = nib.nifti1.Nifti1Image(newdata, None, header=newhdr) newobj = nib.nifti1.Nifti1Image(newdata, None, header=newhdr)
nib.save(newobj, "mynewname.nii.gz") nib.save(newobj, "mynewname.nii.gz")
``` ```
%% Cell type:markdown id: tags: %% Cell type:markdown id: tags:
where `newdata` is the numpy array (the above is a random example only) and where `newdata` is the numpy array (the above is a random example only) and
`imhdr` is the existing image header (as above). `imhdr` is the existing image header (as above).
> It is possible to also just pass in an affine matrix rather than a > It is possible to also just pass in an affine matrix rather than a
> copied header, but we *strongly* recommend against this if you are > copied header, but we *strongly* recommend against this if you are
> processing an existing image as otherwise you run the risk of > processing an existing image as otherwise you run the risk of
> swapping the left-right orientation. Those that have used > swapping the left-right orientation. Those that have used
> `save_avw` in matlab may well have been bitten in this way in the > `save_avw` in matlab may well have been bitten in this way in the
> past. Therefore, copy a header from one of the input images > past. Therefore, copy a header from one of the input images
> whenever possible, and just use the affine matrix option if you are > whenever possible, and just use the affine matrix option if you are
> creating an entirely separate image, like a simulation. > creating an entirely separate image, like a simulation.
If the voxel size of the image is different, then extra modifications will be If the voxel size of the image is different, then extra modifications will be
required. Take a look at the `fslpy` practical for more advanced image required. Take a look at the `fslpy` practical for some extra image
manipulation options (`advanced_topics/08_fslpy.ipynb`). manipulation options, including cropping and resampling
(`advanced_topics/08_fslpy.ipynb`).
--- ---
<a class="anchor" id="exercises"></a> <a class="anchor" id="exercises"></a>
## Exercise ## Exercise
Write some code to read in a 4D fMRI image (you can find one Write some code to read in a 4D fMRI image (you can find one
[here](http://www.fmrib.ox.ac.uk/~mark/files/av.nii.gz) if you don't have one [here](http://www.fmrib.ox.ac.uk/~mark/files/av.nii.gz) if you don't have one
handy), calculate the tSNR and then save the 3D result. handy), calculate the tSNR and then save the 3D result.
> The tSNR of a time series signal is simply its mean divided by its standard
> deviation.
%% Cell type:code id: tags: %% Cell type:code id: tags:
``` ```
# Calculate tSNR # Calculate tSNR
``` ```
......
getting_started/05_nifti.md
+ 25
− 18
View file @ 7ee8c4a3
...@@ -6,10 +6,10 @@ MINC, MGH). `nibabel` is included within the FSL python environment. ...@@ -6,10 +6,10 @@ MINC, MGH). `nibabel` is included within the FSL python environment.
Building upon `nibabel`, the Building upon `nibabel`, the
[`fsl.data`](https://users.fmrib.ox.ac.uk/~paulmc/fsleyes/fslpy/latest/fsl.data.html#module-fsl.data) [`fslpy`](https://users.fmrib.ox.ac.uk/~paulmc/fsleyes/fslpy/latest/) library
package contains a number of FSL-specific classes and functions which you may contains a number of FSL-specific classes and functions which you may find
find useful. This is covered in a different practical useful. But let's start with `nibabel` - `fslpy` is introduced in a different
(`advanced_topics/08_fslpy.ipynb`). practical (`advanced_topics/08_fslpy.ipynb`).
## Contents ## Contents
...@@ -38,31 +38,33 @@ imobj = nib.load(filename, mmap=False) ...@@ -38,31 +38,33 @@ imobj = nib.load(filename, mmap=False)
# display header object # display header object
imhdr = imobj.header imhdr = imobj.header
# extract data (as an numpy array) # extract data (as a numpy array)
imdat = imobj.get_data().astype(float) imdat = imobj.get_fdata()
print(imdat.shape) print(imdat.shape)
``` ```
> Make sure you use the full filename, including the .nii.gz extension. > Make sure you use the full filename, including the `.nii.gz` extension.
> `fslpy` provides FSL-like automatic file suffix detection though.
> We use the expandvars() function above to insert the FSLDIR > We use the `expandvars()` function above to insert the FSLDIR
> environmental variable into our string. This function is > environmental variable into our string. This function is
> discussed more fully in the file management practical. > discussed more fully in the file management practical.
Reading the data off the disk is not done until `get_data()` is called. Reading the data off the disk is not done until `get_fdata()` is called.
> Pitfall: > Pitfall:
> >
> The option `mmap=False`is necessary as turns off memory mapping, > The option `mmap=False` disables memory mapping, which would otherwise be
> which otherwise would be invoked for uncompressed NIfTI files but not for > invoked for uncompressed NIfTI files but not for compressed files. Since
> compressed files. Since some functionality behaves differently on memory > some functionality behaves differently on memory mapped objects, it is
> mapped objects, it is advisable to turn this off. > advisable to turn this off unless you specifically want it.
Once the data is read into a numpy array then it is easily manipulated. Once the data is read into a numpy array then it is easily manipulated.
> We recommend converting it to float at the start to avoid problems with > The `get_fdata` method will return floating point data, regardless of the
> integer arithmetic and overflow, though this is not compulsory. > underlying image data type. If you want the image data in the type that it
> is stored (e.g. integer ROI labels), then use
> `imdat = np.asanyarray(imobj.dataobj)` instead.
--- ---
...@@ -109,6 +111,7 @@ print(affine, code) ...@@ -109,6 +111,7 @@ print(affine, code)
<a class="anchor" id="writing-images"></a> <a class="anchor" id="writing-images"></a>
## Writing images ## Writing images
If you have created a modified image by making or modifying a numpy array then If you have created a modified image by making or modifying a numpy array then
you need to put this into a NIfTI image object in order to save it to a file. you need to put this into a NIfTI image object in order to save it to a file.
The easiest way to do this is to copy all the header info from an existing The easiest way to do this is to copy all the header info from an existing
...@@ -134,8 +137,9 @@ where `newdata` is the numpy array (the above is a random example only) and ...@@ -134,8 +137,9 @@ where `newdata` is the numpy array (the above is a random example only) and
> creating an entirely separate image, like a simulation. > creating an entirely separate image, like a simulation.
If the voxel size of the image is different, then extra modifications will be If the voxel size of the image is different, then extra modifications will be
required. Take a look at the `fslpy` practical for more advanced image required. Take a look at the `fslpy` practical for some extra image
manipulation options (`advanced_topics/08_fslpy.ipynb`). manipulation options, including cropping and resampling
(`advanced_topics/08_fslpy.ipynb`).
--- ---
...@@ -148,6 +152,9 @@ Write some code to read in a 4D fMRI image (you can find one ...@@ -148,6 +152,9 @@ Write some code to read in a 4D fMRI image (you can find one
[here](http://www.fmrib.ox.ac.uk/~mark/files/av.nii.gz) if you don't have one [here](http://www.fmrib.ox.ac.uk/~mark/files/av.nii.gz) if you don't have one
handy), calculate the tSNR and then save the 3D result. handy), calculate the tSNR and then save the 3D result.
> The tSNR of a time series signal is simply its mean divided by its standard
> deviation.
``` ```
# Calculate tSNR # Calculate tSNR
``` ```
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment