25 April 2018 High throughput calculation of local spatial autocorrelation length for label-free diagnosis of tissue biopsy
Author Affiliations +
Proceedings Volume 10503, Quantitative Phase Imaging IV; 105032B (2018) https://doi.org/10.1117/12.2293991
Event: SPIE BiOS, 2018, San Francisco, California, United States
Abstract
Quantitative phase imaging (QPI) can access quantitative information on thickness and/or refractive index changes of weakly absorbing and scattering objects, which normally require staining prior to observation. The quantitative phase image itself yields significant information for a medical diagnosis, particularly in cancer biopsies. Previously, several parameters such as a local standard deviation of refractive index have been utilized as a marker of diseases. We focus on the local spatial autocorrelation length, which is calculated at each point in the field of view. The local spatial autocorrelation length is defined as the standard deviation of the local spatial autocorrelation function and reveals the local and directional disorder information of tissues. However, generally, a direct calculation of the local spatial autocorrelation length take an immense amount of time. In this paper, we propose a high-throughput calculation procedure of a local spatial autocorrelation length, by exploiting frequency-domain calculations. After deriving a simple equation to calculate the local spatial autocorrelation length map in a short time, we perform label-free screening of benign and malignant breast tissue biopsies using this parameter as a marker.
© (2018) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Masanori Takabayashi, Hassaan Majeed, Andre Kajdacsy-Balla, Gabriel Popescu, "High throughput calculation of local spatial autocorrelation length for label-free diagnosis of tissue biopsy", Proc. SPIE 10503, Quantitative Phase Imaging IV, 105032B (25 April 2018); doi: 10.1117/12.2293991; https://doi.org/10.1117/12.2293991
PROCEEDINGS
1 PAGES


SHARE
Back to Top