Seminar: Automatic Image Analysis for High Content Screening
Farhad Mohammad Kazemi
Co-Supervisors: Drs. Wolfgang Banzhaf and Minglun Gong
Automatic Image Analysis for High Content Screening
Department of Computer Science
Friday, May 26, 2017, 11:00a.m., Room EN 2022
High content screening technologies deal with screening thousands of cells which can provide a number of parameters for each cell such as nuclear size, nuclear morphology, DNA replication, etc. The success of High Content Screening (HCS) systems is dependent on automated image analysis. This research focuses on the challenge of using information content that is as high as possible, by considering per-cell information and all the available features, to build a system for high content analysis. In this work, we developed a model to classify MOAs based on phenotypic screens of a number of cells in each image. We demonstrated that our model can predict the mechanisms of action (MOA) for a compendium of drugs that alter cells, without any segmentation and feature extraction steps.