What does this research mean for the field?

A deep learning-based method trained on highly variable, multivendor, and multicenter data can achieve fully automated and accurate quantification of left ventricular function from cine MR images. Novelty: ClaimNovelty.METHODOLOGICAL. Consensus alignment: ConsensusAlignment.NEUTRAL.

October 9, 2018RadiologyOpen Access

Deep Learning–based Method for Fully Automatic Quantification of Left Ventricle Function from Cine MR Images: A Multivendor, Multicenter Study

View Full Paper

Ask AI

Bookmark

Why the study?

Does a deep learning-based method trained on a highly variable dataset accurately quantify left ventricular function from cine MR images compared to manual expert annotation?

Population

596 patients with cine MRI data sets obtained from three major MRI vendors in four medical centers from 2008…

Comparison

Fully automated quantification of left… vs Manual annotations from three experienced…

Design

Cohort

Authors

QTQian TaoSoochow University WYWenjun YanChina Meteorological Administration YWYuanyuan WangNational University of Defense Technology

Discussion

Member takes

Implication

Supports automated LV assessment in research; extends DL validation but leaves open prospective clinical trials.

Structured PICO

Does a deep learning-based method trained on a highly variable dataset accurately quantify left ventricular function from cine MR images compared to manual expert annotation?

Population

596 patients with cine MRI data sets obtained from three major MRI vendors in four medical centers from 2008 to 2016 (400 for training CNN3, 196 for independent testing).

Intervention

Fully automated quantification of left ventricular function using a deep learning-based method (CNN3 with U-NET architecture trained on multivendor, multicenter data).

Comparator

Manual annotations from three experienced observers.

Outcome

LV detection accuracy, LV segmentation accuracy, and LV functional parameter accuracy.surrogate

A deep learning model trained on a diverse, multivendor, multicenter dataset can accurately and automatically quantify left ventricular function from cine MRI, matching expert manual annotations.

Cite This Study

Tao et al. (2018) studied this question.

synapsesocial.com/papers/6a198f7c3f3ec013f0dec136 https://doi.org/10.1148/radiol.2018180513

Topics

Cardiac MRI

View Full Paper

Ask AI

Bookmark

Deep Learning–based Method for Fully Automatic Quantification of Left Ventricle Function from Cine MR Images: A Multivendor, Multicenter Study

Discussion

Member takes

Implication

Structured PICO

Cite This Study

Topics

Deep Learning–based Method for Fully Automatic Quantification of Left Ventricle Function from Cine MR Images: A Multivendor, Multicenter Study

Discussion

Member takes

Implication

Structured PICO

Cite This Study

Also Consider

Topics

Also Consider