Zhihang Ren's Personal Homepage
VEATIC: Video-based Emotion and Affect Tracking in Context Dataset
BEAT: Berkeley Emotion and Affect Tracking Dataset
Description: SkinCON: Towards consensus for the uncertainty of skin cancer sub-typing through distribution regularized adaptive predictive sets (DRAPS)
skin cancer (258) uncertainty (65) skincon (1) predictive sets (1)
Deep learning has been widely utilized in medical diagnosis. Convolutional neural networks and transformers can achieve high predictive accuracy, which can be on par with or even exceed human performance. However, uncertainty quantification remains an unresolved issue, impeding the deployment of deep learning models in practical settings. Conformal analysis can, in principle, estimate the uncertainty of each diagnostic prediction, but doing so effectively requires extensive human annotations to characterize
We plot the response frequency of sample skin cancer images. We argue that the empirical response distribution may reveal the inherent skin cancer property.
User interface used for diagnostic annotation. A random skin cancer image was presented to users. Users are asked to classify which skin cancer type it is among the 8 classes.
Zhihang Ren's Personal Homepage
VEATIC: Video-based Emotion and Affect Tracking in Context Dataset
BEAT: Berkeley Emotion and Affect Tracking Dataset