Description
The application of deep learning to medical diagnosis has been an active area of research in recent years. Convolutional neural networks (CNNs) have made the computational processing of medical images more achievable. CNNs automatically extract important features from images. These features are then processed by an Artificial Neural Network which produces the final classification or regression result. Deploying CNNs in a clinical setting motivates a measure of the certainty of its prediction. In this study, we evaluate the effectiveness of various uncertainty measures on binary classification by CNNs.
Learning Objective: After reading in this poster, the learner should be better able to explain the need for uncertainty quantification in deep learning image classification and explain multiple approach to computing uncertainty.
Authors:
Katherine Brown, Tennessee Tech University
Douglas Talbert (Presenter)
Tennessee Tech University
Presentation Materials: