Tissue perfusion diagnostic classification using a spatio-temporal analysis of contrast ultrasound image sequences

The analysis of tissue perfusion in myocardial contrast echocardiography (MCE) remains a qualitative process dependent on visual inspection by a clinician. Fully automatic techniques that can quantify tissue perfusion accurately has yet to be developed. In this paper, a novel spatio-temporal technique is described for segmenting the myocardium into differently perfused regions and obtaining quantitative perfusion indices, representing myocardial blood flow and blood flow reserve. Using these indices, Myocardial segments in 22 patients were classed as either healthy or diseased, and results compared to coronary angiogram analysis and an experienced clinician. The results show that the automatic method works as well as a human at detecting areas of ischaemia, but in addition localizes the spatial extent of each perfused region as well. To our knowledge this is the first reported spatio-temporal method developed and evaluated for MCE assessment.