周捷博士学术报告

Automatic Analysis and Quantification of High-content Biological Images

摘要：

The recent availability of large amount of high-content biological images using advanced microscopic imaging has brought unprecedented opportunities for new scientific findings. Due to the high-volume and high-dimension of such data, automated quantification is the essential step in the process of discovery. Accordingly, current frontier in computational efforts goes one step beyond visualization and moves into the automatic analysis and quantification of such biological images. This talk will cover several related projects including a concurrent image regression model that infers the regulatory relationship using multi-dimensional gene expression, and quantification of neuronal cellular and sub-cellular structures (e.g. dendritic territory and synapse distribution during genetic screen). In addition, a portable and extensible framework that enables automatic annotation and classification of biological images based on pattern recognition will also be introduced.