Currently, the cardiac dynamic of the entire cycle is not characterized and heart function is approximated by volumetric relationships in specific moments of the cycle. This work represents the left ventricular cardiac dynamics by intraventricular motion patterns in echocardiographic images. Briefly, after segmenting the ventricular cavity, intensity levels within this region of interest are spatially grouped by a classic superpixel (SP) clustering method, aiming to maximize the intergroup variance with a minimum number of groups. The distribution of these superpixels is then correlated with different cycle times, specifically diastole and systole. A total of 70 cases, half with Ejection Fraction (EF) less than 40 % and half with control EF (higher than 50 %), were processed and results demonstrated the number of groups is significantly smaller in cases with control EF, indicating that intraventricular patterns are correlated with the EF.