biostudies-literatureUnknown52(4)Pauthe AFrench Society of RadiologyCentre National d'Etudes SpatialesSociété Française de Radiologie<h4>Background</h4>Deep learning image reconstruction (DLIR) algorithms allow strong noise reduction while preserving noise texture, which may potentially improve hypervascular focal liver lesions.<h4>Purpose</h4>To assess the impact of DLIR on image quality (IQ) and detectability of simulated hypervascular hepatocellular carcinoma (HCC) in fast kV-switching dual-energy CT (DECT).<h4>Methods</h4>An anthropomorphic phantom of a standard patient morphology (body mass index of 23 kg m^-2) with customized liver, including mimickers of hypervascular lesions in both late arterial phase (AP) and portal venous phase (PVP) enhancement, was scanned on a DECT. Virtual monoenergetic images were reconstructed from raw data at four energy levels (40/50/60/70 keV) using filtered back-projectionMedical physics2257-2268https://www.ebi.ac.uk/biostudies/studies/S-EPMC11972042biostudies-literatureImpact of deep learning reconstructions on image quality and liver lesion detectability in dual-energy CT: An anthropomorphic phantom study.PMC11972042Campagnolo LLuciani AMilliner MPauthe AMule SPasquier HfalseImpact of deep learning reconstructions on image quality and liver lesion detectability in dual-energy CT: An anthropomorphic phantom study.<h4>Background</h4>Deep learning image reconstruction (DLIR) algorithms allow strong noise reduction while preserving noise texture, which may potentially improve hypervascular focal liver lesions.<h4>Purpose</h4>To assess the impact of DLIR on image quality (IQ) and detectability of simulated hypervascular hepatocellular carcinoma (HCC) in fast kV-switching dual-energy CT (DECT).<h4>Methods</h4>An anthropomorphic phantom of a standard patient morphology (body mass index of 23 kg m^-2) with customized liver, including mimickers of hypervascular lesions in both late arterial phase (AP) and portal venous phase (PVP) enhancement, was scanned on a DECT. Virtual monoenergetic images were reconstructed from raw data at four energy levels (40/50/60/70 keV) using filtered back-projection2025-01-01T00:00:00Z2025 Apr2025-07-08T03:13:45.173Z2025-07-08T03:13:45.173ZS-EPMC119720423988775010.1002/mp.17651