← Journal

My ProjectsAI/ML 1 min read

AI for Ranking Immunocytochemistry Images

Automatically ranking ICC images by how well antibodies bind to the nucleus, using deep learning features.

Analysing immunocytochemistry (ICC) images, figuring out how well antibodies attach to the nucleus of interest, is still mostly done by hand.

In this project, we proposed an automatic process that ranks ICC images by how strongly antibodies adhere to the nucleus. A reliable computer-aided system like this depends on a classifier built on meaningful features, so we evaluated how well deep learning features work for the task.

The results were published at the 4th International Conference on Pattern Recognition and Image Analysis (IPRIA) in 2019.