Quantitative Evaluation of the Performance of a New Test Bolus–Based Computed Tomographic Angiography Contrast-Enhancement–Prediction Algorithm

    loading  Checking for direct PDF access through Ovid

Abstract

Objectives

The objective of this study was to assess the robustness of a novel test bolus (TB)–based computed tomographic angiography (CTA) contrast-enhancement–prediction (CEP) algorithm by retrospectively quantifying the systematic and random errors between the predicted and true enhancements.

Materials and Methods

All local institutional review boards approved this retrospective study, in which a total of 72 (3 × 24) anonymized cardiac CTA examinations were collected from 3 hospitals. All patients (46 men; median age, 62 years [range, 31–81 years]) underwent a TB scan and a cardiac CTA according to local scan and injection protocols. For each patient, a shorter TB signal and TB signals with lower temporal resolution were derived from the original TB signal. The CEP algorithm predicted the enhancement in the descending aorta (DAo) on the basis of the TB signals in the DAo, the injection protocols and kilovolt settings, as well as population-averaged blood circulation characteristics. The true enhancement was extracted with a region of interest along the DAo centerline. For each patient, the errors in timing and amplitude were calculated; differences between the hospitals were assessed using the 1-way analysis of variance (P < 0.05) and variations between the TB signals were assessed using the within-subject standard deviation.

Results

No significant differences were found between the 3 hospitals for any of the TB signals. With errors in the amplitude and timing of 0.3% ± 15.6% and −0.2 ± 2.0 seconds, respectively, no clinically relevant systematic errors existed. Shorter- and coarser-time–sampled TB signals introduced a within-subject standard deviation of 4.0% and 0.5 seconds, respectively.

Conclusions

This TB-based CEP algorithm has no systematic errors in the timing and amplitude of predicted enhancements and is robust against coarser-time–sampled and incomplete TB scans.

Related Topics

    loading  Loading Related Articles