Meeting Banner
Abstract #3762

Semi-Automatic Detection of Cerebral Microbleeds on Clinical 3.0T T2*-Weighted Images Using the Radial Symmetry Transform

Hugo J. Kuijf1, Manon Brundel2, Max A. Viergever1, Geert Jan Biessels2, Koen L. Vincken1

1Image Sciences Institute, University Medical Center Utrecht, Utrecht, Netherlands; 2Department of Neurology, Rudolf Magnus Institute of NeuroScience, UMC Utrecht

The current standard for cerebral microbleed detection is visual rating, which is laborious and has limited reproducibility. Semi-automated methods for detection of microbleeds on clinical images have been suggested before, but suffer from a large number of false positives. In this study, we present a method for microbleed detection based on the radial symmetry transform. This results in a high sensitivity of 75% and a limited number of false positives, requiring just two minutes of rater time to censor them. Furthermore, the method outperforms other known methods in terms of specificity.

Abstract PDF E-Poster Video

Keywords

added among anatomical anisotropic applied automated automatic better binary bleeds bound brain carried caused censor cerebral checks clinic clinical clinically comp computation confirmed consisted core created dark date decrease demanding detected detecting detection determined disagreed displayed dual elongated evaluated exceed experienced extra false fewer field final find five full gradient gray human included independently institute intensity involving lancet likely limited list locally losing manual marked mars mask mass medical memory middle minute minutes much need negatives neurology normalized objects original outperforms owing paper patient patients pattern percentiles positives potential press principle procedure process processing projection published quality radial rapidly rater raters rating ratter referred registered remove require required requiring rigidly rising scale sciences scored scoring seems semi sensitivity several significantly since slab slice slightly specificity spots stage stages structures symmetry table terms third thresholding trans transform true truth turbo typical upper validated vascular version vessels visual visually went white workstation written