Assessment of Human Red Blood Cell Aggregation Using Image Processing and Wavelets

Aggregation of red blood cells which manifests in the form of rouleaux is a major determinant of flow behavior of blood in both micro and macro circulation. Any alteration in the normal behavior is found to be a critical factor in plugging of arterioles and venules forming irreversible clumps. In this work, the aggregation behavior of red blood cells is analyzed using Wavelet transforms. The artificially induced red cell aggregate images obtained from normal adult volunteers are used for the study. An arbitrary aggregation size index is derived using four different wavelet functions. The results demonstrate the ability of wavelets to precisely differentiate variations in degree of aggregation and their intercellular bonding. Also the aggregation size indices are found to be similar and consistent for a given aggregate for all wavelet functions. As the association and dissociation of red cells and the bonding strength during cluster formation are direct indicators of altered flow behavior of red cells in micro-vessels and large arteries such analysis seems to be clinically relevant. The methodology, algorithm and observations based on Wavelet transforms are discussed in detail.