/SparseViewCTRecon

Code for CS766 Computer Vision Course Project: Sparse View CT Reconstruction Based on a Dual-domain Convolutional Neural Network

Primary LanguagePython

Sparse-View CT Reconstruction

This repository contains code for reconstructing CT images from sparse-view projection data. The code is written in Python and uses the PyTorch and MONAI libraries. Check our final report and slides and webpage📜.

Usage:

Run the reconstruction script:

python train_grad.py -c config/hyperparam_grad.json

or

python train_sino.py -c config/hyperparam_sino.json

To evaluate the model:

python evaluate_grad.py -c config/hyperparam_grad.json

or

python evaluate_sino.py -c config/hyperparam_sino.json