Digital Pathology of Cancer

Microscopic view of a tissue sample with densely packed cells, stained with hematoxylin and eosin. The nuclei are darkly visible, and the overall pattern suggests a cellular proliferation or tumor. The sample is presented at a high magnification.

Computational Grading of GI-NETs

We developed automated Ki-67 index estimation pipelines for gastrointestinal neuroendocrine tumors (GI-NETs) for defining the tumor grade using computational image analysis and deep learning tools. Our pipelines can improve accuracy and potentially save a significant amount of time if implemented into clinical practice.