Research Project Title:
Image Segmentation of Digitized Histopathology Slides
abstract:Digital histopathology, the machine-aided processing of histopathology samples, promises life-saving advances in the accuracy, efficiency, and scope of medical diagnostics. One of the field's key goals is automated "image segmentation" of cellular entities -- information critical to most cancer grading schemes. This project aims to modify a pre-existing semantic segmentation network (Ademxapp model A1 pre-trained on the Cityscapes Dataset) to create a model capable of segmentation across multiple types of cellular entities; the model will be trained, tested, and evaluated on the PAIP2019 dataset. Finally, the network will be fine-tuned for segmentation related to histological grading of hepatocellular carcinoma; its resulting performance will be compared to that of pathologists.
"I am participating in SuperUROP because I want to engage in a long-term research project centered on machine vision-aided medical diagnostics and gain more exposure to multi-disciplinary collaborations within the healthcare R&D sector. I hope to use my previous UROP work in digital histopathology as a foundation for my project, which I dream might one day play some small, but genuine, role in helping another human being."