Personal tools
You are here: Home Research Publications Technical Reports 2008 cse-08-008


Application of Computational Symmetry to Histology Images

Brian Canada and Yanxi Liu

Penn State CSE Department Technical Report #008-01

Abstract—The larval zebrafish is an ideal organism for studying mutant phenotypes (observable traits) because of its small size and rapid, ex vivo development. Histology is one highly sensitive means for detecting and scoring zebrafish mutants, and while “high-throughput” methods have been developed for preparing digital “virtual slides” of zebrafish larval histology, problems of subjectivity and labor bottlenecks associated with scoring these virtual slides impede large-scale histological analysis from being widely adopted in zebrafish laboratories. Here, we demonstrate that novel computer vision techniques derived from and inspired by the current state-of-the-art algorithms for computational symmetry detection have the potential to improve the efficiency and accuracy of the histology image preparation and classification workflow, thereby bringing the overall process closer to being more fully automated and truly “high-throughput.”

Document Actions