Poisson inverse gradient approach to vascular myocyte detection and segmentation

Authors:
Scott T. Acton;Clare Yang;John A. Hossack;Brian R. Wamhoff
Affiliations:
Departments of Biomedical Engineering, University of Virginia, Charlottesville, VA and Departments of Electrical & Computer Engineering, University of Virginia, Charlottesville, VA;Departments of Electrical & Computer Engineering, University of Virginia, Charlottesville, VA;Departments of Biomedical Engineering, University of Virginia, Charlottesville, VA and Departments of Electrical & Computer Engineering, University of Virginia, Charlottesville, VA;Departments of Biomedical Engineering, University of Virginia, Charlottesville, VA and Departments of Medicine, University of Virginia, Charlottesville, VA
Venue:
ISBI'09 Proceedings of the Sixth IEEE international conference on Symposium on Biomedical Imaging: From Nano to Macro
Year:
2009

Citing 7
Cited 0

A Computational Approach to Edge Detection

IEEE Transactions on Pattern Analysis and Machine Intelligence
Iterative Methods for Sparse Linear Systems

Iterative Methods for Sparse Linear Systems
Parallel Distributed FFT-Based Solvers for 3-D Poisson Problems in Meso-Scale Atmospheric Simulations

International Journal of High Performance Computing Applications
Segmentation of external force field for automatic initialization and splitting of snakes

Pattern Recognition
Snakes, shapes, and gradient vector flow

IEEE Transactions on Image Processing
Active Contour External Force Using Vector Field Convolution for Image Segmentation

IEEE Transactions on Image Processing
Automatic Active Model Initialization via Poisson Inverse Gradient

IEEE Transactions on Image Processing

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper addresses the detection and segmentation of vascular myocytes. The detection and segmentation of these cells are critical to the investigation of atherosclerosis among other cardiovascular diseases. Our approach to detection is unique in that it attempts to compute the underlying external energy in an active contour model. Isolines in this computed external energy can be employed to localize boundaries of the cell nuclei. The process used to solve the inverse problem of obtaining the energy from the force vectors is called Poisson inverse gradient due to the Poisson-based solution. From the initial contours given by the isolines in the computed energy, parametric active contours are used to find the subtle cell boundaries. The results indicate that the Poisson inverse gradient improves detection accuracy and reduces false positives compared to existing morphological methods. Furthermore, the contourbased detection allows segmentation of the cell boundaries.