Detnet: Deep Neural Network For Particle Detection In Fluorescence Microscopy Images

Automatic detection of subcellular structures displayed as small spots in fluorescence microscopy images is an important task to determine quantitative information on cellular processes. We propose a new approach for particle detection, which uses deep learning and is based on a domain adapted Deconvolution Network. Compared to standard deep neural network architectures, the number of parameters is significantly reduced. We benchmarked our approach on the ISBI Particle Tracking Challenge data and live cell fluorescence microscopy data of hepatitis C virus proteins. It turned out that our approach yields high detection and localization performance for particles of different shapes as well as for different signal-to-noise ratios.