Project Description:
In collaboration with the Institute for Bioscience and Biotechnology Research, our project is centered on designing an advanced image processing system that utilizes a convolutional neural network, specifically a U-Net architecture, to analyze synthetic two-dimensional images of SARS-CoV-2 viruses. The primary objective is to accurately segment the virus body from the image background and pinpoint the distinctive spike proteins, which are crucial for understanding the virus’s infectious mechanisms. By extracting and quantifying detailed morphological features such as the viral body's radius and the number of spike proteins, the system aims to provide vital statistics that can inform virological studies. This proof-of-concept system not only shows promise in enhancing precision and speed of virological research but also lays the groundwork of identifying potential therapeutic targets.
