Diagnostic Performance of CEM Quantitative Parameters for Breast Lesions and Concordance with MRI Enhancement Kinetics

Read the full article See related articles

Discuss this preprint

Start a discussion What are Sciety discussions?

Listed in

This article is not in any list yet, why not save it to one of your lists.
Log in to save this article

Abstract

Objective To evaluate the value of quantitative parameters in contrast-enhanced spectral mammography (CEM) for differentiating benign and malignant breast lesions and their concordance with MRI kinetic curve (TIC) types. Methods Fifty-five cases of pathologically confirmed breast lesions from March 2019 to November 2023 were retrospectively included. All of them underwent CEM and MRI before surgery, and the signal intensity of the lesions and surrounding tissues in the early cephalocaudal (CC) and delayed-phase oblique (MLO) positions on CEM subtraction maps were measured, and the relative enhancement gray values (CNR1, CNR2) and relative intensity gray difference (RSD) were calculated. The CEM enhancement curves were categorized into inflow, plateau, and outflow types according to the rate of change of RSD, and the MRI-TIC type was recorded. Non-parametric tests were used to compare the parameter differences between benign and malignant groups (17 cases of benign, 31 cases of invasive carcinoma, and 7 cases of non-invasive carcinoma), ROC curves were used to assess the diagnostic efficacy, and Kappa tests were used to analyze the consistency of the curves. Results CNR1 and CNR2 were significantly higher in invasive versus non-invasive cancers than in the benign group (both P<0.05), while RSD was lower than in the benign group (P<0.05). Two-by-two comparisons between groups: CNR1, CNR2 and RSD differed significantly between benign and invasive cancers (P<0.05); RSD differed significantly between benign and non-invasive cancers (P<0.05); and CNR2 differed significantly between invasive and non-invasive cancers (P<0.05).The AUCs for diagnosing malignant lesions with CNR1 and CNR2 were 0.816 and 0.757, respectively.CEM Intensification curve distribution differed significantly between benign and malignant (χ²=37.658, P<0.001), and its concordance rate with MRI-TIC type was 83.64% (Kappa=0.723). Conclusion CEM quantitative parameters (especially CNR) demonstrate high diagnostic value in differentiating benign and malignant breast lesions. CEM enhancement curves show strong concordance with MRI TIC, and enhancement intensity correlates with lesion type.

Article activity feed