respiratory2Dto3Dpaper

This repository contains links to relevant software from our paper.

Our paper proposes a new approach for reconstructing patient-specific lung and airway geometries from a 2D X-ray image. This relies on a statistical shape and appearance model (SSAM) from a set of 3D lung shapes and digitally reconstructed radiographs (based on the CT data). We have released a package pyssam to facilitate other SSAM studies.

2D-3D SSAM reconstruction code

We have provided the code for our 2D-3D shape reconstruction algorithm here or at doi: 10.5281/zenodo.10518240. We provide a static version of the code in this DOI. The linked repository also contains a version which relies on our pyssam package, which is available here. Note that pyssam is under active development.

CNN pytorch code

We have provided code and pre-trained models for lung and airway segmentations from CT images here or at doi: 10.5281/zenodo.10512412.

CFD solver source code

Our 3D-0D OpenFOAM solver for flow in patient airways is given here or at doi: 10.5281/zenodo.7448214.