Discriminative dictionary learning for local LV wall motion classification in cardiac MRI

作者：

Highlights：

• Dictionary Learning for Left Ventricle (LV) motion classification in cardiac cine-MRI.

• Extraction of new spatio-temporal features without the need of LV delineation.

• Assessment of radial spatial-temporal profiles with echocardiography 2D STE.

• Best results on real data obtained by the DL techniques in the Wavelet domain.

• A best compromise in accuracy, complexity and time provided by the LC-KSVD2 algorithm.

摘要

•Dictionary Learning for Left Ventricle (LV) motion classification in cardiac cine-MRI.•Extraction of new spatio-temporal features without the need of LV delineation.•Assessment of radial spatial-temporal profiles with echocardiography 2D STE.•Best results on real data obtained by the DL techniques in the Wavelet domain.•A best compromise in accuracy, complexity and time provided by the LC-KSVD2 algorithm.

论文关键词：Dictionary learning,LV cardiac motion,Cardiac MRI,Spatio-temporal profiles,Sparse signal classification

论文评审过程：Received 25 July 2018, Revised 5 April 2019, Accepted 5 April 2019, Available online 6 April 2019, Version of Record 17 April 2019.

论文官网地址：https://doi.org/10.1016/j.eswa.2019.04.010