On 3D tracking for echocardiographic classification of acute myocardial infarction

Accurate characterization of acute myocardial infarction (AMI) is crucial for the management of the patient. In this work we present a method for 3-dimensional (3D) velocity estimation and object tracking in sequences of ultrasound volume scans. Velocity estimation is based on motion estimation in 3D with maximum likelihood criteria. Tracking is carried out using a data association method based on location and velocity. The proposed algorithm has been tested on various data sources and on in-vivo 3D volume scan of the left ventricle, providing very useful results for the characterization of AMIs.