Myocardial extracellular volume fraction radiomics analysis for differentiation of reversible versus irreversible myocardial damage and prediction of left ventricular adverse remodeling after ST-elevation myocardial infarction

Objectives Our study sought to explore the prognostic value of radiomic TA (texture analysis) on quantitative ECV (extracellular volume) fraction mapping to differentiate between reversible and irreversible myocardial damage and to predict left ventricular adverse remodeling in patients with reperfu...

Full description

Saved in:
Bibliographic Details
Published in:European radiology 2021, Vol.31 (1), p.504-514
Main Authors: Chen, Bing-Hua, An, Dong-Aolei, He, Jie, Wu, Chong-Wen, Yue, Ting, Wu, Rui, Shi, Ruo-Yang, Eteer, Khalid, Joseph, Bobby, Hu, Jiani, Xu, Jian-Rong, Wu, Lian-Ming, Pu, Jun
Format: Article
Language:English
Subjects:
Cardiac
Contrast Media
Damage
Diagnostic Radiology
Differentiation
Feature extraction
Gadolinium
Heart
Heart attacks
Humans
Imaging
Injuries
Injury analysis
Internal Medicine
Interventional Radiology
Magnetic Resonance Imaging, Cine
Male
Mapping
Medicine
Medicine & Public Health
Myocardial infarction
Myocardium
Neuroradiology
Parameters
Predictive Value of Tests
Prospective Studies
Radiology
Radiomics
Recovery of function
ST Elevation Myocardial Infarction - diagnostic imaging
Statistical analysis
Texture
Ultrasound
Ventricle
Ventricular Function, Left
Ventricular Remodeling
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Objectives Our study sought to explore the prognostic value of radiomic TA (texture analysis) on quantitative ECV (extracellular volume) fraction mapping to differentiate between reversible and irreversible myocardial damage and to predict left ventricular adverse remodeling in patients with reperfused STEMI (ST-elevation myocardial infarction). Methods This observational prospective cohort study identified 70 patients (62 ± 9 years, 62 men [85.70%]) with STEMI for TA who consecutively performed native and contrast T1 mapping. Texture features were extracted from each stack of ECV mapping based on ROI (region of interest) analysis. Results After texture feature selection and dimension reduction, five selected texture features were found to be statistically significant for differentiating the extent of myocardial injury. ROC (receiver operating characteristic) curve analysis for the differentiation of unsalvageable infarction and salvageable myocardium demonstrated a significantly higher AUC (area under the curve) (0.91 [95% CI, 0.86–0.96], p  
ISSN:0938-7994
1432-1084
DOI:10.1007/s00330-020-07117-9