Efficient globally optimal segmentation of cells in fluorescence microscopy images using level sets and convex energy functionals

• Published in: Medical image analysis Vol. 16; no. 7; pp. 1436 - 1444
• Main Authors: Bergeest, Jan-Philip, Rohr, Karl
• Format: Journal Article
• Language: English
• Published: Netherlands Elsevier B.V 01.10.2012
• Subjects: Active contours; Algorithms; Artificial Intelligence; Cell segmentation; Cell Tracking - methods; Convex optimization; Image Enhancement - methods; Image Interpretation, Computer-Assisted - methods; Level sets; Microscopy, Fluorescence - methods; Pattern Recognition, Automated - methods; Reproducibility of Results; Sensitivity and Specificity; Level sets; Convex optimization; Cell segmentation; Active contours
• ISSN: 1361-8415; 1361-8423
• DOI: 10.1016/j.media.2012.05.012

[Display omitted] ► Cell segmentation using active contours, level sets, and convex energy functionals. ► Reformulation and combination of well-known non-convex energy functionals. ► Efficient numeric computation of the global solution using the Split Bregman method. ► Experimental evaluation based on different fluorescence microscopy images. ► The approach copes well with intensity inhomogeneities and cell clustering. In high-throughput applications, accurate and efficient segmentation of cells in fluorescence microscopy images is of central importance for the quantification of protein expression and the understanding of cell function. We propose an approach for segmenting cell nuclei which is based on active contours using level sets and convex energy functionals. Compared to previous work, our approach determines the global solution. Thus, the approach does not suffer from local minima and the segmentation result does not depend on the initialization. We consider three different well-known energy functionals for active contour-based segmentation and introduce convex formulations of these functionals. We also suggest a numeric approach for efficiently computing the solution. The performance of our approach has been evaluated using fluorescence microscopy images from different experiments comprising different cell types. We have also performed a quantitative comparison with previous segmentation approaches.