Fast Myocardial Motion and Strain Estimation in 3D Cardiac Ultrasound with Sparse Demons

Oudom Somphone (1) , Mathieu de Craene (2) , Roberto Ardon (1) , Benoit Mory (1) , Pascal Allain (1) , Hang Gao (3) , Jan d'Hooge (3) , Stéphanie Marchesseau (4) , Maxime Sermesant (4) , Eric Saloux (5)

1 Medisys - MedisysResearch Lab
2 CISTIB - Center for Computational Imaging and Simulation Technologies in Biomedicine
3 KU Leuven - Catholic University of Leuven = Katholieke Universiteit Leuven
4 ASCLEPIOS - Analysis and Simulation of Biomedical Images
5 Service de cardiologie et de pathologie vasculaire [CHU Caen]

This article presents a new method for motion and strain estimation in 3D echocardiography, called Sparse Demons, along with quantitative and qualitative evaluations from a dataset of synthetic ultrasound sequences. Motion estimation is based on a fast demons-like algorithm focusing on myocardial tissue. Synthetic 3D ultrasound images were generated by combining a biomechanical model of the heart with a realistic ultrasound imaging model. Ischemic areas were defined in the mechanical model to investigate whether our algorithm can discriminate healthy from diseased segments.

Domaines: Imagerie médicale; Traitement des images [eess.IV]; Traitement du signal et de l'image [eess.SP]; Traitement du signal et de l'image [eess.SP]

Oudom Somphone, Mathieu de Craene, Roberto Ardon, Benoit Mory, Pascal Allain, et al.. Fast Myocardial Motion and Strain Estimation in 3D Cardiac Ultrasound with Sparse Demons. International Symposium on Biomedical Imaging (ISBI), 2013, Apr 2013, San Francisco, United States. pp.1182-1185, ⟨10.1109/ISBI.2013.6556691⟩. ⟨hal-00840038⟩