Speaker: Nailah Rawnaq and Pritom Roy, University of Arkansas | Department of Mathematical Sciences
Date: Wednesday, October 4, 2023, 4:40 – 5:00 PM
Location: Science Engineering Hall (SCEN 408)
Title: Performance analysis of U-Net over the different number of initial channels
Abstract:
Medical image segmentation is the process of identifying the objects of interest accurately within an image for diagnostic and treatment planning purposes. One of the well-known image segmentation algorithms is the Convolutional Neural Networks (CNNs)-based segmentation algorithm U-Net that consists of encoder-decoder architecture with four different stages in the network. We study the effect of the number of input channels on the performance of U-Net architecture with the goal of finding the most efficient way of analyzing the electron microscopy dataset for mitochondria separation in the region of the brain.
Recent Comments