TUNet: Incorporating segmentation maps to improve classification
This work addresses protein localization classification for biomedical research, but it appears incremental as it builds on existing U-Net architectures with minor modifications.
The authors tackled the problem of classifying protein localization in human cells from fluorescence microscopy images by introducing TUNet, a model that incorporates segmentation maps, and achieved competitive performance compared to models like GoogleNet and ResNet.
Determining the localization of specific protein in human cells is important for understanding cellular functions and biological processes of underlying diseases. Among imaging techniques, high-throughput fluorescence microscopy imaging is an efficient biotechnology to stain the protein of interest in a cell. In this work, we present a novel classification model Twin U-Net (TUNet) for processing and classifying the belonging of protein in the Atlas images. Several notable Deep Learning models including GoogleNet and Resnet have been employed for comparison. Results have shown that our system obtaining competitive performance.