Advanced Search

Browse by Discipline

Scientific Societies

E-print Alerts

Add E-prints

E-print Network

  Advanced Search  

Multi-Assignment Interacting Multiple Model for Tracking Microbubbles

Summary: Multi-Assignment Interacting Multiple Model
for Tracking Microbubbles
Bing Li, Peter Tay, and Scott T. Acton
{bingli, ptay, acton}@virginia.edu
Charles L. Brown Department of Electrical and Computer Engineering
Department of Biomedical Engineering
University of Virginia, Charlottesville, VA 22904 USA
Abstract -- A novel multi-target tracking (MTT) method is
developed in this paper, which is specifically designed to track
microbubbles in vitro. The microbubbles, which frequently
overlap and move erratically, are accurately tracked by a
tailored multi-assignment (MA) algorithm combined with the
interacting multiple model (IMM) estimator. The superiority of
the proposed modified multi-assignment interacting multiple
model (MA-IMM) tracking algorithm is demonstrated by way of
a comparison to three other approaches.
Molecular imaging is involved in a number of disease states
and is of great interest in various biomedical research areas
[1-2]. Molecular imaging is based on the detection of


Source: Acton, Scott - Department of Electrical and Computer Engineering, University of Virginia


Collections: Computer Technologies and Information Sciences