Medical Segmentation Decathlon

[junxnone]

Reference

• Medical Segmentation Decathlon
  • amazonaws
  • google drive
  • 2/2019 A large annotated medical image dataset for the development and evaluation of segmentation algorithms
  • 如何读取NIFTI格式图像（.nii文件）

Brief

• NIFTI - Neuroimaging Informatics Technology Initiative
• Analyze Format - 早期用于 MRI, 逐渐被 NIFTI 替代
  • .img - 图像 data
  • .hdr - 元数据

Tasks

Num	Tasks	Description
01	BrainTumour	4 维 = 3维 + Time(T1 / T1-Gd / T2 / T2-FLAIR)

Read image

import nibabel as nib
from nilearn import image
import matplotlib.pyplot as plt
%matplotlib inline

img_path = 'Task01_BrainTumour/imagesTr/BRATS_001.nii.gz'
label_path = 'Task01_BrainTumour/labelsTr/BRATS_001.nii.gz'

img = nib.load(img_path).get_fdata()
label = nib.load(label_path).get_fdata()

print(f'image shape {img.shape}')
print(f'label shape {label.shape}')

plt.figure(figsize=(10,10))
plt.subplot(121)
plt.imshow(img[:,:,60,2],cmap='gray')
plt.title('image')

plt.subplot(122)
plt.imshow(label[:,:,60], cmap='gray')
plt.title('label')

import itk
from itkwidgets import view
import itkwidgets
nii = nib.load(img_path).get_fdata()[:,:,:,0]
viewer = view(nii,cmap=itkwidgets.cm.grayscale)
viewer

[junxnone]

898 #591