Repository logo
 

Study of Morphology Based Cell Assay by Diffraction Imaging Flow Cytometry

Loading...
Thumbnail Image

Date

2015

Access

Authors

Jiang, Wenhuan

Journal Title

Journal ISSN

Volume Title

Publisher

East Carolina University

Abstract

Development of an accurate and label-free method for single cell assay attracts intensive research efforts for its importance to cell biology research and clinicG14G7al applications. Flow cytometry is one of the most widely used technologies for rapid assay of single cells but existing approaches provide very limited information on cell morphology and require the fluorescence staining. In this dissertation research, we focus our efforts on the quantitative analysis of cell morphology using confocal microscopy based three-dimensional (3D) reconstruction and the exploration of a new approach of flow cytometry through imaging of highly coherent scattered light. The goal of the dissertation research is to develop a new and morphology-based approach for rapid cell assay and phenotyping with the polarization diffraction imaging flow cytometry (p-DIFC) platform through investigation of the structure-function relations at the cell level. To achieve this goal, cross-polarized diffraction image pairs have been acquired from single cells excited by a linearly polarized laser beam. Image texture and intensity parameters are extracted with a gray level co-occurrence matrix (GLCM) algorithm to obtain a set of image parameters to quantify the diffraction patterns. An automated cell classification method has been developed using a Support Vector Machine (SVM) algorithm in the feature space formed by the training data of the cross-polarized diffraction image pairs. We have investigated different types of human lymphocytes and prostate epithelial cells with the confocal imaging and p-DIFC measurements and conducted cell morphology and classification studies. The analysis of 3D morphology parameters among the six types of cells provides, for the first time, the ability to quantitatively evaluate the morphologic differences among these phenotypes and to gain insights on the morphology-based classification. It has been further shown that the diffraction image parameters can be mapped into a high-dimensional feature space with the SVM algorithm to obtain the optimized model and yield accurate classifications between Jurkat T cells and Ramos B cells and between the normal and cancerous prostate epithelial cells. Based on these results we conclude that the p-DIFC method has significant potentials to be developed into a rapid and label-free method for cell assay and morphology-based classification to discriminate cells of high similarity in their morphology.

Description

Citation

DOI