Abstract
We aimed to assess the reliability of an automatic solution for whole-heart segmentation of MRI images of patients with atrial fibrillation (AF). We propose a semi-interactive image segmentation algorithm based on region growing, GrowCut1, using novel neighborhood structures based on Cellular Automata. We complemented the algorithm with a global view of the signal using remote neighbors outside standard neighborhoods (Moore and Von Neumann). We validated the proposed semi-interactive algorithm on a clinical dataset from our AtFib study.
Citare
Marginean, R., Popa, L., Coman, M., Manole, S., Coman, V., Andreica, A., Diosan, L., Bálint, Z., Extended region growing algorithm for whole heart segmentation from cardiac MRI images, ESCR 2019, 2019, accepted