Semi-overcomplete convolutional auto-encoder embedding as shape priors for deep vessel segmentation

TL;DR

This paper introduces a semi-overcomplete convolutional auto-encoder that embeds shape priors to improve deep vessel segmentation, especially for tiny structures, outperforming standard U-Net models on retinal and liver datasets.

Contribution

It proposes a novel semi-overcomplete auto-encoder architecture that enhances vessel segmentation by incorporating shape priors, addressing limitations of existing U-Net models.

Findings

Improved segmentation accuracy on retinal and liver vessel datasets.

Effective characterization of tiny vascular structures.

Outperforms U-Net without shape priors.

Abstract

The extraction of blood vessels has recently experienced a widespread interest in medical image analysis. Automatic vessel segmentation is highly desirable to guide clinicians in computer-assisted diagnosis, therapy or surgical planning. Despite a good ability to extract large anatomical structures, the capacity of U-Net inspired architectures to automatically delineate vascular systems remains a major issue, especially given the scarcity of existing datasets. In this paper, we present a novel approach that integrates into deep segmentation shape priors from a Semi-Overcomplete Convolutional Auto-Encoder (S-OCAE) embedding. Compared to standard Convolutional Auto-Encoders (CAE), it exploits an over-complete branch that projects data onto higher dimensions to better characterize tiny structures. Experiments on retinal and liver vessel extraction, respectively performed on publicly-available DRIVE and 3D-IRCADb datasets, highlight the effectiveness of our method compared to U-Net trained without and with shape priors from a traditional CAE.