Software MARTA: Improvements in Cell Segmentation and CX43 Lateralization Assessment

Cardiomyocytes’ geometry and connexin 43 (CX43) amount and distribution are structural features that play a pivotal role in electrical conduction of the heart. Their quantitative assessment is of high interest in the study of arrhythmias, but it is usually hampered by the lack of automatic tools. MARTA (Myocyte Automatic Retrieval and Tissue Analyzer) is a software algorithm which can automatically segment cardiomyocytes in fluorescent microscopy images of cardiac tissue and evaluate the expression of CX43 and its degree of lateralization. This study aims to analyze one type of myocardial tissue section of rat treated in the original software’s study, improve the algorithm's segmentation performance, and extend its evaluation to new images of rat cardiac tissue. Besides, a new metric to evaluate the CX43 lateralization is presented. The metric is based on the dispersion of the lateralization distribution instead of the mean value. The results suggest that the presented method doubles the precision of the original algorithm’s segmentation, with a slight increase in sensitivity. However, further evaluation of the proposed segmentation algorithm and metric are required to assess the actual performance of the methodology.