Uncertainty-aware deep learning methods for robust diabetic retinopathy classification

HIGHLIGHTS

• who: Retinopathy Classification and colleagues from the Department of Computer Science, Aalto University, Aalto, Finland have published the research: Uncertainty-aware Deep Learning Methods for Robust Diabetic Retinopathy Classification, in the Journal: (JOURNAL)
• what: The authors derive a connection between entropy-based uncertainty measure and classifier risk from which the authors develop a novel uncertainty measure. The common aspect among the works focusing on robust methods is the use of uncertainty information to simulate a referral process, introduced by Leibig et_al . They observed that the approximate Bayesian methods outperformed the standard neural_network in all the . . .
  

   

  

  If you want to have access to all the content you need to log in!

  Thanks :)

  Username or E-mail:

  If you don't have an account, you can create one here.