Modeling Uncertainty in Deep Learning-Based MRI Analysis

Home

Discover

Bookmarks

Keep Up With Latest Trending Papers. Computer Science, AI and more.Subscribe

CC BY

Source arXiv

Machine Learning

Image and Video Processing

Artificial Intelligence

Modeling Uncertainty in Deep Learning-Based MRI Analysis

Calibrated Diffusion Tensor Estimation

We propose a deep learning method to estimate the diffusion tensor and compute the estimation uncertainty of a model.

Data-dependent uncertainty is computed directly by the network and learned via lossattenuation.

We compare the new method with the standard least-squares tensor estimation and bootstrap-based uncertainty computation techniques.

Our experiments show that when the number of measurements is small the deep learning method is more accurate and its uncertainty predictions are better calibrated than the standard methods.

We show that the estimation uncertainties computed by the newmethod can highlight the model s biases, detect domain shift, and reflect the strength of noise in the measurements.

We also propose a new method for evaluating the quality of predicted uncertainties.

Authors

Davood Karimi, Simon K. Warfield, Ali Gholipour

◐ Latest Activity

◐ Recommended Follows