SUD: Supervision by Denoising for Medical Image Segmentation
release_zuxk2jcnnnat3jrbqxv5wutftu
by
Sean I. Young, Adrian V. Dalca, Enzo Ferrante, Polina Golland, Bruce Fischl, Juan Eugenio Iglesias
2022
Abstract
Training a fully convolutional network for semantic segmentation typically
requires a large, labeled dataset with little label noise if good
generalization is to be guaranteed. For many segmentation problems, however,
data with pixel- or voxel-level labeling accuracy are scarce due to the cost of
manual labeling. This problem is exacerbated in domains where manual annotation
is difficult, resulting in large amounts of variability in the labeling even
across domain experts. Therefore, training segmentation networks to generalize
better by learning from both labeled and unlabeled images (called
semi-supervised learning) is problem of both practical and theoretical
interest. However, traditional semi-supervised learning methods for
segmentation often necessitate hand-crafting a differentiable regularizer
specific to a given segmentation problem, which can be extremely
time-consuming. In this work, we propose "supervision by denoising" (SUD), a
framework that enables us to supervise segmentation models using their denoised
output as targets. SUD unifies temporal ensembling and spatial denoising
techniques under a spatio-temporal denoising framework and alternates denoising
and network weight update in an optimization framework for semi-supervision. We
validate SUD on three tasks-kidney and tumor (3D), and brain (3D) segmentation,
and cortical parcellation (2D)-demonstrating a significant improvement in the
Dice overlap and the Hausdorff distance of segmentations over supervised-only
and temporal ensemble baselines.
In text/plain
format
Archived Files and Locations
application/pdf 28.9 MB
file_tglretxffraalpz63p6ez4tjme
|
arxiv.org (repository) web.archive.org (webarchive) |
2202.02952v1
access all versions, variants, and formats of this works (eg, pre-prints)