Realtime automatic detection of heart failure in echocardiography

Assessment of heart function by echocardiography is challenging for non-experts. In a patient with dyspnea, assessment of mitral annular excursion (MAE) and velocities are important for the diagnosis of heart failure. By automatic measurement of these parameters, a preliminary diagnosis could be performed by the non-expert. We propose an automatic algorithm to localize the mitral annular points in the apical tissue Doppler image, and estimate MAE, as well as the three peak velocities early systolic, early diastolic and late diastolic. Results on estimation accuracy and classification of systolic function based on the estimated mitral annular excursion are presented.