Ching Lam Choi - Corona-Net

Discover CoronaNet, a deep learning model for automated CT diagnosis and segmentation of COVID-19 lung lesions, with potential applications in patient triaging, personalized treatment planning, and alleviating radiologist workload.

Key takeaways

CoronaNet is a model for automated CT diagnosis confirmation and segmentation of COVID-19 lung lesions
It utilizes a novel compound scaling method and the U-net architecture
The model was trained on a dataset of approximately 1,000 CT scans and achieved good results in binary and multi-class segmentation tasks
CoronaNet has potential applications in patient triaging and personalized treatment planning, and can help alleviate the workload of radiologists
The model was developed using PyTorch and other Python libraries, and the code is available on GitHub for further development and collaboration
Future work includes improving the model’s accuracy and generalizability, as well as exploring its applications in other medical imaging domains

Ching Lam Choi - Corona-Net

More talks