GTC 2020: Deep Learning-Based Subcellular Phenotyping of Cell Edge Dynamics Reveals Fine-Grained Drug Responses
After clicking “Watch Now” you will be prompted to login or join.
Click “Watch Now” to login or join the NVIDIA Developer Program.
Deep Learning-Based Subcellular Phenotyping of Cell Edge Dynamics Reveals Fine-Grained Drug Responses
Kwonmoo Lee, Worcester Polytechnic Institute
We'll highlight how unsupervised deep learning can reveal subcellular drug responses hidden in the heterogeneity in live cell images. We'll present a feature-learning method for time-series data, which combines long short-term memory autoencoder and the prior information from traditional machine-learning analysis. We'll discuss how feature learning can be used to identify drug-related rare phenotypes and accelerate drug discovery. You need to have basic knowledge about autoencoder, clustering, microscopy, and cell biology.