Histogram-Based Optical Flow for Functional Imaging in Echocardiography

摘要

Echocardiographic imaging provides various challenges for medical image analysis due to the impact of physical effects in the process of data acquisition. The most significant difference to other medical data is its high level of speckle noise that makes the use of conventional algorithms difficult. Motion analysis on ultrasound (US) data is often referred to as 'Speckle Tracking' which plays an important role in diagnosis and monitoring of cardiovascular diseases and the identification of abnormal cardiac motion. In this paper we address the problem of speckle noise within US images for estimating optical flow. We demonstrate that methods which directly use image intensities are inferior to methods using local features within the US images. Based on this observation we propose an optical flow method which uses histograms as a local feature of US images and show that this approach is more robust under the presence of speckle noise than classical optical flow methods.