Authors:

Sydney Cecelia Sobczak

Date Created:

Not specified

Course Title:

Professor:

Not specified

About Paper:

A key challenge in neuronal studies, particularly in the context of neurodegenerative disease research, lies in the accurate and consistent detection of neurons having particular characteristics, as viewed through a microscopic lens. While Fluorescence Lifetime Imaging (FLIM) allows for the visualization of fluorophore-tagged neurons, current methods for identifying and classifying these cells are often manual, time-consuming, and vulnerable to human error, thus potentially leading to biased study conclusions. To address this, we develop a classification method to detect neurons based on neuron an assessment of health using intensity and fluorescence lifetime data. Our FLIM system employs a pulsed laser and an intensified gated camera, allowing measurement of the temporal fluorescence response. Using this system, we implemented a code to obtain a large amount of fluorescence data. Next, to select neurons of interest, we use a classification approach based on K-means clustering, solidity analysis, and feature thresholding to distinguish and separate neurons based on their health. The K-means clustering successfully separates areas inside and outside the sample well, allowing us to identify the edges and boundaries of the sample. This methodology demonstrates promise for efficient neuron analysis. Future work will focus on incorporating a larger and diverse dataset to refine thresholds and explore shape-based features to improve detection capabilities and accuracy based on neuron morphology. Keywords: [no keywords provided]

Source:

Purdue University / 2025

Topics:

No topics listed

Co-authors:

Sydney Cecelia Sobczak