Share Email Print
cover

Proceedings Paper

Compressive sensing in reflectance confocal microscopy of skin images: a preliminary comparative study
Author(s): Fernando X. Arias; Heidy Sierra; Milind Rajadhyaksha; Emmanuel Arzuaga
Format Member Price Non-Member Price
PDF $14.40 $18.00

Paper Abstract

Compressive Sensing (CS)-based technologies have shown potential to improve the efficiency of acquisition, manipulation, analysis and storage processes on signals and imagery with slight discernible loss in data performance. The CS framework relies on the reconstruction of signals that are presumed sparse in some domain, from a significantly small data collection of linear projections of the signal of interest. As a result, a solution to the underdetermined linear system resulting from this paradigm makes it possible to estimate the original signal with high accuracy. One common approach to solve the linear system is based on methods that minimize the L1-norm. Several fast algorithms have been developed for this purpose. This paper presents a study on the use of CS in high-resolution reflectance confocal microscopy (RCM) images of the skin. RCM offers a cell resolution level similar to that used in histology to identify cellular patterns for diagnosis of skin diseases. However, imaging of large areas (required for effective clinical evaluation) at such high-resolution can turn image capturing, processing and storage processes into a time consuming procedure, which may pose a limitation for use in clinical settings. We present an analysis on the compression ratio that may allow for a simpler capturing approach while reconstructing the required cellular resolution for clinical use. We provide a comparative study in compressive sensing and estimate its effectiveness in terms of compression ratio vs. image reconstruction accuracy. Preliminary results show that by using as little as 25% of the original number of samples, cellular resolution may be reconstructed with high accuracy.

Paper Details

Date Published: 9 March 2016
PDF: 11 pages
Proc. SPIE 9713, Three-Dimensional and Multidimensional Microscopy: Image Acquisition and Processing XXIII, 97130Z (9 March 2016); doi: 10.1117/12.2213685
Show Author Affiliations
Fernando X. Arias, Univ. of Puerto Rico Mayaguez (United States)
Heidy Sierra, Memorial Sloan Kettering Cancer Ctr. (United States)
Milind Rajadhyaksha, Memorial Sloan Kettering Cancer Ctr. (United States)
Emmanuel Arzuaga, Univ. of Puerto Rico Mayaguez (United States)


Published in SPIE Proceedings Vol. 9713:
Three-Dimensional and Multidimensional Microscopy: Image Acquisition and Processing XXIII
Thomas G. Brown; Carol J. Cogswell; Tony Wilson, Editor(s)

© SPIE. Terms of Use
Back to Top