MedPath

Deep Learning Image Reconstruction for Abdominal CT of Hepatocellular Carcinoma Compared With 3-TESLA MRI

Completed
Conditions
Hepatocellular Carcinoma
Registration Number
NCT06037343
Lead Sponsor
Central Hospital, Nancy, France
Brief Summary

New algorithms for processing CT acquisitions, based on artificial intelligence, have been reported to improve acquisition quality. Thats' why it's possible to imagine that new scan post-processing algorithms enable better detection and characterization of hepatocellular carcinoma lesions than with standard reconstructions. DLIR reconstructions could even match with MRI detection.

The aim of the study is to compare the detection and characterization of hepatic lesions according to the LI-RADS classification in CT with DLIR artificial intelligence reconstruction, compared with ASIR-V reconstruction and the gold standard of MRI.

Detailed Description

Not available

Recruitment & Eligibility

Status
COMPLETED
Sex
All
Target Recruitment
50
Inclusion Criteria
  • undergoing CT and MRI scans in the same week, with protocols dedicated to the detection of HCC lesions
Exclusion Criteria
  • imaging with radiological artefact

Study & Design

Study Type
OBSERVATIONAL
Study Design
Not specified
Primary Outcome Measures
NameTimeMethod
Lesion size in mmThrough study completion, an average of 1 year
Hypervascular appearance of lesionThrough study completion an average of 1 year

Qualitative measurement: presence or absence of hypervascular lesion

Hypervascular capsuleThrough study completion an average of 1 year

Qualitative measurement : presence or absence of hypervascular capsule

Secondary Outcome Measures
NameTimeMethod
Infiltrative nature Classification of the hepatic lesion by LI-RADS in MRIThrough study completion an average of 1 year

Trial Locations

Locations (1)

CHRU de Nancy

🇫🇷

Nancy, France

© Copyright 2025. All Rights Reserved by MedPath