PD-DWI is a physiologically-decomposed Diffusion-Weighted MRI machine-learning model for predicting response to neoadjuvant chemotherapy in invasive breast cancer.
PD-DWI was developed by TCML group as part of BMMR2 challenge using ACRIN-6698 dataset.
This work was accepted to MICCAI 2022 (to be held during Sept 18-22 in Singapore).
If you publish any work which uses this package, please cite the following publication:
M. Gilad, M. Freiman. PD-DWI: Predicting response to neoadjuvant chemotherapy in invasive breast cancer with Physiologically-Decomposed Diffusion-Weighted MRI machine-learning model. Medical Image Computing and Computer Assisted Intervention – MICCAI 2022 to be held during Sept 18-22 in Singapore.
A pre-print version of our paper is available at https://arxiv.org/abs/2206.05695
PD-DWI model can be installed directly from Github:
pip install git+https://github.com/TechnionComputationalMRILab/PD-DWI.git
PD-DWI can be used in a Python script or via command line.
To explore all CLI options and syntax requirements run pd-dwi --help in your terminal.
Please contact us on [email protected]