Reproducibility of Computed Tomography perfusion parameters in hepatic multicentre study in patients with colorectal cancer

The Computed Tomography perfusion (CTp) is a promising tool in oncology to characterize tissue hemodynamics, but the difficulty achieve reproducible parameters several organs, with different methods, contributes hamper clinical translation of CTp. goal this study set up new approach aiming at achieving multicentre reproducibility blood flow (BF) values liver. 75 patients from two Centres (A and B) underwent an axial liver CTp, including arterial portal phases. A dedicated workflow addressing modelling computational aspects was implemented, novel two-stage strategy separate dual-input contributions hepatic signals, thus allowing compute independently both Maximum Slope (MS) Deconvolution (DV) on same contributing signals. 95% B showed excellent voxel-based Pearson correlation (ρ≥0.96) between MS DV BF values, very low coefficients variation (CV=0.11 worst case). good concordance confirmed for whole cohorts, single both, where R2=0.97, ρ≥0.97, ρs≥0.96, ICC≥0.78 CV=0.25 are values. Compared eighteen recent articles, these represent by far best outcomes. patient- cohort-based achieved BV confirms effectiveness presented. Our can be used improve other CTp studies, as well even questions, represents marked step forward towards standardization, favouring investigation imaging biomarkers.