Artificial intelligence-based methods for quantitative phenotyping of neural rosettes from microscopy images

This project focuses on building novel image analysis pipelines for analyzing neural rosettes using multi-channel microscopy data. Approaches will include segmentation model development, automated quality control, instance-level morphometrics and spatial modelling of rosette substructures. The goal is to enable high-throughput reproducible quantification of early developmental changes in neurogenetic disorders. This project would suit students interested in artificial intelligence, biomedical imaging and computational biology.