Instructions for installation and usage examples can be found below.

Install

It's highly recommended you install I2T2 and its dependencies in a virtual environment (conda or others), so that you don't interfere with system-wide python packages.

1. Installing via github download:

git clone https://github.com/UCSF-intelligent-imaging/I2T2/
cd I2T2
conda create -n I2T2 python=3.7
conda activate I2T2
pip install nbdev jupyter
pip install -e .
conda install -c conda-forge gdcm

1. Installing via pip (latest stable version)

conda create -n I2T2 python=3.7
conda activate I2T2
pip install nbdev jupyter
pip install I2T2
conda install -c conda-forge gdcm

Test your installation

conda activate I2T2
python
>>> import I2T2

NOTE:

I2T2 uses GDCM to uncompress some DICOMs.

Make sure to have GDCM installed in your conda environment via conda install -c conda-forge gdcm

conda activate I2T2
python
>>> import gdcm

Using I2T2

# example usage:
from I2T2.io import *
import matplotlib.pyplot as plt

data_path = '../data/knee/'
try:
    array = load_dcm(path_to_dicom_dir = data_path, sort_by_slice_location=True)
    plt.imshow(array[:,:,0])
    plt.show()
except:
    print('could not load array')    

could not load array