Identification and Quantification of Cardiovascular Structures From CCTA: An End-to-End, Rapid, Pixel-Wise, Deep-Learning Method.

Aged Computed Tomography Angiography Coronary Angiography Deep Learning Female Heart Diseases / diagnostic imaging Humans Male Middle Aged Multidetector Computed Tomography Predictive Value of Tests Prospective Studies Radiographic Image Interpretation, Computer-Assisted Registries Reproducibility of Results

coronary computed tomography angiography deep learning quantification

Journal

JACC. Cardiovascular imaging

ISSN: 1876-7591

Titre abrégé: JACC Cardiovasc Imaging

Pays: United States

ID NLM: 101467978

Informations de publication

Date de publication:
05 2020

Historique:

received: 30 05 2019

revised: 08 08 2019

accepted: 23 08 2019

pubmed: 15 10 2019

medline: 6 1 2021

entrez: 15 10 2019

Statut: ppublish

Résumé

This study designed and evaluated an end-to-end deep learning solution for cardiac segmentation and quantification. Segmentation of cardiac structures from coronary computed tomography angiography (CCTA) images is laborious. We designed an end-to-end deep-learning solution. Scans were obtained from multicenter registries of 166 patients who underwent clinically indicated CCTA. Left ventricular volume (LVV) and right ventricular volume (RVV), left atrial volume (LAV) and right atrial volume (RAV), and left ventricular myocardial mass (LVM) were manually annotated as ground truth. A U-Net-inspired, deep-learning model was trained, validated, and tested in a 70:20:10 split. Mean age was 61.1 ± 8.4 years, and 49% were women. A combined overall median Dice score of 0.9246 (interquartile range: 0.8870 to 0.9475) was achieved. The median Dice scores for LVV, RVV, LAV, RAV, and LVM were 0.938 (interquartile range: 0.887 to 0.958), 0.927 (interquartile range: 0.916 to 0.946), 0.934 (interquartile range: 0.899 to 0.950), 0.915 (interquartile range: 0.890 to 0.920), and 0.920 (interquartile range: 0.811 to 0.944), respectively. Model prediction correlated and agreed well with manual annotation for LVV (r = 0.98), RVV (r = 0.97), LAV (r = 0.78), RAV (r = 0.97), and LVM (r = 0.94) (p < 0.05 for all). Mean difference and limits of agreement for LVV, RVV, LAV, RAV, and LVM were 1.20 ml (95% CI: -7.12 to 9.51), -0.78 ml (95% CI: -10.08 to 8.52), -3.75 ml (95% CI: -21.53 to 14.03), 0.97 ml (95% CI: -6.14 to 8.09), and 6.41 g (95% CI: -8.71 to 21.52), respectively. A deep-learning model rapidly segmented and quantified cardiac structures. This was done with high accuracy on a pixel level, with good agreement with manual annotation, facilitating its expansion into areas of research and clinical import.

Sections du résumé

OBJECTIVES

This study designed and evaluated an end-to-end deep learning solution for cardiac segmentation and quantification.

BACKGROUND

Segmentation of cardiac structures from coronary computed tomography angiography (CCTA) images is laborious. We designed an end-to-end deep-learning solution.

METHODS

Scans were obtained from multicenter registries of 166 patients who underwent clinically indicated CCTA. Left ventricular volume (LVV) and right ventricular volume (RVV), left atrial volume (LAV) and right atrial volume (RAV), and left ventricular myocardial mass (LVM) were manually annotated as ground truth. A U-Net-inspired, deep-learning model was trained, validated, and tested in a 70:20:10 split.

RESULTS

Mean age was 61.1 ± 8.4 years, and 49% were women. A combined overall median Dice score of 0.9246 (interquartile range: 0.8870 to 0.9475) was achieved. The median Dice scores for LVV, RVV, LAV, RAV, and LVM were 0.938 (interquartile range: 0.887 to 0.958), 0.927 (interquartile range: 0.916 to 0.946), 0.934 (interquartile range: 0.899 to 0.950), 0.915 (interquartile range: 0.890 to 0.920), and 0.920 (interquartile range: 0.811 to 0.944), respectively. Model prediction correlated and agreed well with manual annotation for LVV (r = 0.98), RVV (r = 0.97), LAV (r = 0.78), RAV (r = 0.97), and LVM (r = 0.94) (p < 0.05 for all). Mean difference and limits of agreement for LVV, RVV, LAV, RAV, and LVM were 1.20 ml (95% CI: -7.12 to 9.51), -0.78 ml (95% CI: -10.08 to 8.52), -3.75 ml (95% CI: -21.53 to 14.03), 0.97 ml (95% CI: -6.14 to 8.09), and 6.41 g (95% CI: -8.71 to 21.52), respectively.

CONCLUSIONS

A deep-learning model rapidly segmented and quantified cardiac structures. This was done with high accuracy on a pixel level, with good agreement with manual annotation, facilitating its expansion into areas of research and clinical import.

Identifiants

DOI: 10.1016/j.jcmg.2019.08.025 PMID: 31607673

pubmed: 31607673

pii: S1936-878X(19)30873-3

doi: 10.1016/j.jcmg.2019.08.025

pii:

doi:

Types de publication

Journal Article Multicenter Study Observational Study Research Support, Non-U.S. Gov't Validation Study

Langues

eng

Sous-ensembles de citation

Pagination

1163-1171

Commentaires et corrections

Type : CommentIn