Dresden 2026 – scientific programme

Parts | Days | Selection | Search | Updates | Downloads | Help

BP: Fachverband Biologische Physik

BP 14: Poster Session II

BP 14.35: Poster

Tuesday, March 10, 2026, 18:00–21:00, P2

Red Blood Cell shape classification using curvature flow and spherical harmonics — •Nikolas Lerch, Felix Maurer, Diana Örüm, and Thomas John — Universität des Saarlandes

The morphological characterization of erythrocytes deformed due to pathology or disease has traditionally relied on manual assessment, which introduces subjective sources of error and hampers the differentiation of closely related morphologies such as echinocytes and acanthocytes. We present an automated method for the objective classification of deformed erythrocytes. The approach employs a bijective mapping of surface curvature onto a reference sphere via curvature flow and decomposes the resulting features into spherical harmonics. The resulting spectral signatures enable the assignment of cells to morphologically distinct clusters. The method is quantitative and reproducible, explicitly identifies transient shapes and transition regions between morphologies, recognizing them as such and distinguishing them from clearly defined class expressions, and thereby opens new perspectives for computer-assisted hematology.

Keywords: Red Blood Cells; Curvature Flow; Spherical Harmonics; Clustering