Catalina Correa, Javier RodrÃÆÃÂguez, Signed Prieto, Luisa ÃÆÃÂlvarez, BenjamÃÆÃÂn Ospino, ÃÆÃÂngela MunÃÆÃÂ©var, Pedro Bernal, Jessica Mora and Sarith Vitery
Fractal geometry allows the unbiased characterizati on of objects irregularity. Fractal dimension has been applied to different diagnostic images in orde r to differentiate normality from disease. In this paper, 7 normal and 7 abnormal erythrocytes from pa tients and transfusion bags were analyzed. Software was developed in order to evaluate fractal dimensions and the number of spaces covered by the edge of the red blood cells in the 5x5 pixel gr id in the Box Counting space. The surface occupied by the image in pixels was also analyzed in order to es tablish differences between normality and disease. A methodology of clinical application to diagnose e rythrocyte morphophysiology was developed based on the ratio between the number of surface pixels a nd the number of spaces touched by the edge in the 5x5 grid. Values equal or greater than 174.7 repres ent normal erythrocytes, and values of 170.84 or below represent the abnormal ones. The developed me thodology shows that the morphology and physiology of red cells have a geometric self-organ ization characterized by a mathematical proportion based on a simultaneous use of fractal and euclidia n measures. This allows differentiating normality and abnormality for any particular case in an objec tive way.
Share this article