Few-shot brain segmentation from weakly labeled data with deep heteroscedastic multi-task networks

by   Richard McKinley, et al.

In applications of supervised learning applied to medical image segmentation, the need for large amounts of labeled data typically goes unquestioned. In particular, in the case of brain anatomy segmentation, hundreds or thousands of weakly-labeled volumes are often used as training data. In this paper, we first observe that for many brain structures, a small number of training examples, (n=9), weakly labeled using Freesurfer 6.0, plus simple data augmentation, suffice as training data to achieve high performance, achieving an overall mean Dice coefficient of 0.84 ± 0.12 compared to Freesurfer over 28 brain structures in T1-weighted images of ≈ 4000 9-10 year-olds from the Adolescent Brain Cognitive Development study. We then examine two varieties of heteroscedastic network as a method for improving classification results. An existing proposal by Kendall and Gal, which uses Monte-Carlo inference to learn to predict the variance of each prediction, yields an overall mean Dice of 0.85 ± 0.14 and showed statistically significant improvements over 25 brain structures. Meanwhile a novel heteroscedastic network which directly learns the probability that an example has been mislabeled yielded an overall mean Dice of 0.87 ± 0.11 and showed statistically significant improvements over all but one of the brain structures considered. The loss function associated to this network can be interpreted as performing a form of learned label smoothing, where labels are only smoothed where they are judged to be uncertain.


page 3

page 8


Data augmentation using learned transforms for one-shot medical image segmentation

Biomedical image segmentation is an important task in many medical appli...

DeepAtlas: Joint Semi-Supervised Learning of Image Registration and Segmentation

Deep convolutional neural networks (CNNs) are state-of-the-art for seman...

Self-Supervised Generative Style Transfer for One-Shot Medical Image Segmentation

In medical image segmentation, supervised deep networks' success comes a...

Deep Label Fusion: A 3D End-to-End Hybrid Multi-Atlas Segmentation and Deep Learning Pipeline

Deep learning (DL) is the state-of-the-art methodology in various medica...

Brain MRI Segmentation using Template-Based Training and Visual Perception Augmentation

Deep learning models usually require sufficient training data to achieve...

CT organ segmentation using GPU data augmentation, unsupervised labels and IOU loss

Fully-convolutional neural networks have achieved superior performance i...

Multi-Model Medical Image Segmentation Using Multi-Stage Generative Adversarial Networks

Image segmentation is a challenging problem in medical applications. Med...

Please sign up or login with your details

Forgot password? Click here to reset