Table 1.

Quantitative evaluation of hepatic steatosis using computed tomography

CT-based tools	Principles	Acquisition methods
Deep learning	Automated algorithms for liver segmentation and analysis	All voxels designated as liver by the segmentation algorithm were analyzed, and the mean and median HU were computed
Quantitative CT	Using a scanner with a five-rod calibration phantom with an aqueous K2HPO4 bone density standard placed beneath the participants	CTFF = (HUlean -HUliuer)/(HUlean—HUfat) HUliver is the measurement in Hounsfield units in the liver HUlean is the value in Hounsfield units for fat-free liver tissue HUfat is the value for 100% fat
Dual-energy CT	It provides information about tissue composition	VNC and iodine maps; TNC images; MMD algorithm
Deep learning	Automated algorithms for liver segmentation and analysis	All voxels designated as liver by the segmentation algorithm were analyzed, and the mean and median HU were computed
Photon-counting CT	It is able to detect and weight individual photons based on their energies	TNC and VNC images

CT Computed tomography, CTFF CT fat fraction, HU Hounsfield units, MMD Multi-material decomposition, TNC True non-contrast, VNC Virtual non-contrast