Publications by authors named "Deepthi Karkada"
Article Synopsis
- Machine learning can work well, but it often struggles to make accurate predictions on new data, which is called out-of-sample generalizability.
- To solve this problem, researchers are using a method called Federated ML that allows computers to share information about how well they're learning without actually sharing the data itself.
- In a big study with 71 locations around the world, scientists created a model to help detect brain tumors more accurately, showing a significant improvement compared to older methods and hoping to help with rare illnesses and data sharing in healthcare.
Brain extraction is an indispensable step in neuro-imaging with a direct impact on downstream analyses. Most such methods have been developed for non-pathologically affected brains, and hence tend to suffer in performance when applied on brains with pathologies, e.g.
View Article and Find Full Text PDF