Post-DAE: Anatomically Plausible Segmentation via Post-Processing with Denoising Autoencoders

TL;DR

Post-DAE is a post-processing technique using denoising autoencoders to enhance the anatomical plausibility of biomedical image segmentations, effectively correcting errors without additional computational burden.

Contribution

This paper introduces Post-DAE, a novel manifold learning-based post-processing method that improves segmentation plausibility by projecting masks onto learned anatomical manifolds.

Findings

Improves segmentation masks in chest X-ray and cardiac MRI.

Corrects noisy and erroneous segmentation outputs.

Operates independently of image modality and intensity information.

Abstract

We introduce Post-DAE, a post-processing method based on denoising autoencoders (DAE) to improve the anatomical plausibility of arbitrary biomedical image segmentation algorithms. Some of the most popular segmentation methods (e.g. based on convolutional neural networks or random forest classifiers) incorporate additional post-processing steps to ensure that the resulting masks fulfill expected connectivity constraints. These methods operate under the hypothesis that contiguous pixels with similar aspect should belong to the same class. Even if valid in general, this assumption does not consider more complex priors like topological restrictions or convexity, which cannot be easily incorporated into these methods. Post-DAE leverages the latest developments in manifold learning via denoising autoencoders. First, we learn a compact and non-linear embedding that represents the space of anatomically plausible segmentations. Then, given a segmentation mask obtained with an arbitrary method, we reconstruct its anatomically plausible version by projecting it onto the learnt manifold. The proposed method is trained using unpaired segmentation mask, what makes it independent of intensity information and image modality. We performed experiments in binary and multi-label segmentation of chest X-ray and cardiac magnetic resonance images. We show how erroneous and noisy segmentation masks can be improved using Post-DAE. With almost no additional computation cost, our method brings erroneous segmentations back to a feasible space.