GitHub - piyushkansalongit/Medical-Image-Segmentation-: UNet is a state-of-the-art CNN based architecture typically used in segmentation tasks. I have used this for semantic segmentation of Anterior and Posterior Views of Human Spine. The catch here is that the architecture is trained on just 30-90 masks and can be as large as 30 million parameters and yet achieves an IoU >70.

piyushkansalongit / Medical-Image-Segmentation- Public

Notifications You must be signed in to change notification settings
Fork 2
Star 5

UNet is a state-of-the-art CNN based architecture typically used in segmentation tasks. I have used this for semantic segmentation of Anterior and Posterior Views of Human Spine. The catch here is that the architecture is trained on just 30-90 masks and can be as large as 30 million parameters and yet achieves an IoU >70.