Datasets to support: Optimising lab X-ray computed tomography acquisition for high-throughput characterisation of microstructured samples

Reconstructed X-ray computed tomography data for lithium-sulfur (Li-S) and NMC532 samples. Data reconstructed using CIL [1] using FDK (Feldkamp-Davis-Kress) and CGLS (Conjugate gradient least squares) algorithms. See main text for acquisition details. 

All data are for the same region of interest (ROI) within the respective samples, cropped to 400x400 voxels in the plane of the electrode to ensure manageable data size. Data included here were used to calculate image quality metrics (root mean squared error, RMSE, structural similarity index, SSIM, and contrast to noise ratio, CNR) and 3d particle size distributions. 

Filenames include: exposure time (exp), reduction factor (RF), and for CGLS only iterations (iter). Reduction factor indicates number of projections used for reconstruction: for RF of n, every nth projection is used. All datasets acquired with 2401 projections. For example, RF6 corresponds to int(2401/6) = 400 projections.

Data are in the form of 3d tiff stacks which can be opened using:
- Thermofisher Avizo3d
- ImageJ
- Python:
	from tifffile import imread


Folder name					Sample	Recon method	Exposure times		Reduction factors	Number of iterations
LiS_CGLS_recon_400x400subvolumes		LiS	CGLS		15, 10, 5, 2.5		1, 2, 6, 12		30
LiS_FDK_recon_400x400subvolumes			LiS	FDK		15, 10, 5, 2.5		1, 2, 6, 12		N/A
NMC_CGLS_recon_400x400_aligned_subvolumes	NMC	CGLS		15, 2.5			1, 2, 6, 12		30
NMC_FDK_recon_400x400_aligned_subvolumes	NMC	FDK		15, 2.5			1, 2, 6, 12		N/A



[1] Jørgensen JS et al. 2021 Core Imaging Library Part I: a versatile python framework for tomographic imaging https://doi.org/10.1098/rsta.2020.0192 . Phil. Trans. R. Soc. A 20200192