LoDoPaB-CT Challenge Set

Observation data for the LoDoPaB-CT challenge, which asks to reconstruct CT images of the human lung from (simulated) low photon count measurements.

The setting is identical to the one of the LoDoPaB-CT dataset (documented in this Data Descriptor article), which is supposed to be employed for training learned methods. This challenge set contains observations for a separate set of patients.

Python utilities for accessing this challenge set and creating the submission file are available at github.com/jleuschn/lodopab_challenge. The LoDoPaB-CT dataset for training can be accessed using the DIVal python library (github.com/jleuschn/dival).

Like for the LoDoPaB-CT dataset, reconstructions from the LIDC/IDRI dataset are used as a basis for this challenge set.

The ZIP file contains multiple HDF5 files. Each HDF5 file contains one HDF5 dataset named "data", that provides a number of samples (128 except for the last file). For example, the n-th observation sample is stored in the file "observation_challenge_%03d.hdf5" where "%03d" is floor(n / 128), at row (n mod 128) of "data".

For this challenge set no patient IDs are provided (in contrast to the fully public parts of the LoDoPaB-CT dataset), since the reconstruction algorithm should not rely on this information.

Acknowledgements

Johannes Leuschner, Maximilian Schmidt and Daniel Otero Baguer acknowledge the support by the Deutsche
Forschungsgemeinschaft (DFG) within the framework of GRK 2224/1 “π3: Parameter Identification – Analysis,
Algorithms, Applications”. We thank Simon Arridge, Ozan Öktem, Carola-Bibiane Schönlieb and Christian
Etmann for the fruitful discussion about the procedure, and Felix Lucka and Jonas Adler for their ideas and
helpful feedback on the simulation setup. The authors acknowledge the National Cancer Institute and the Foundation for the National Institutes of Health, and their critical role in the creation of the free publicly available LIDC/IDRI Database used in this study.