Machine learning powered Mueller matrix microscope for collagen and elastin visualization in the mouse cervix

POLARIZED LIGHT AND OPTICAL ANGULAR MOMENTUM FOR BIOMEDICAL DIAGNOSTICS 2022(2022)

Cited 2|Views2
No score
Abstract
There are 15 million infants born prematurely each year worldwide. Of these, about 1 million will die of complications from reduced gestation (37 weeks and less) before the age of five. Cervical remodeling, which is the transformation of the cervix from a firm structure to a soft one, is essential for both term and preterm birth (PTB). Monitoring the uterine cervix remodeling and particularly the arrangement of the cervix primary structural components (elastin and collagen) is of great interest to researchers studying PTB. We have utilized a Self-validating Mueller Matrix Micro-Mesoscope (SAMMM) with convolutional neural networks (CNN) and K-nearest neighbor (K-NN) for classification of elastin and collagen fibers in the mouse cervix. In this work, we proposed that an independent polarized microscope can be used for collagen and elastin classification leveraging the previously developed classifier. The Mueller matrix and decomposition parameters of depolarization, retardance and diattenuation obtained with this system are fed to the previously developed classifier. Excised cervical tissues (50 mu m thickness) were used in this study including samples obtained at different gestation days.
More
Translated text
Key words
Mueller matrix, collagen, elastin, machine learning, preterm birth, convolutional networks
AI Read Science
Must-Reading Tree
Example
Generate MRT to find the research sequence of this paper
Chat Paper
Summary is being generated by the instructions you defined