Multi-organ Nuclei Segmentation & Classification

Pathological image reading is time consuming. And we try to develop a method that can simultaneously segment and classify nuclei on the pathologicla image. Also due to the H&E staining, the color of each image varies which we found can significantly impact the classification performance. In addition, we come up with a U-Net based model with less parameters.

Here’s the code Github Repo for the project.