Dense Volume-to-Volume Vascular Boundary Detection

TL;DR

This paper introduces I2I-3D, a novel 3D CNN architecture for accurate, efficient boundary detection in volumetric medical data, outperforming previous methods and extending edge detection techniques to 3D.

Contribution

The paper presents I2I-3D, a new deep learning framework for 3D boundary detection, and introduces HED-3D, extending 2D edge detection to volumetric data.

Findings

Outperforms state-of-the-art 3D boundary detection methods

Achieves precise voxel-level boundary localization

Processes a 512x512x512 volume in about one minute

Abstract

In this work, we present a novel 3D-Convolutional Neural Network (CNN) architecture called I2I-3D that predicts boundary location in volumetric data. Our fine-to-fine, deeply supervised framework addresses three critical issues to 3D boundary detection: (1) efficient, holistic, end-to-end volumetric label training and prediction (2) precise voxel-level prediction to capture fine scale structures prevalent in medical data and (3) directed multi-scale, multi-level feature learning. We evaluate our approach on a dataset consisting of 93 medical image volumes with a wide variety of anatomical regions and vascular structures. In the process, we also introduce HED-3D, a 3D extension of the state-of-the-art 2D edge detector (HED). We show that our deep learning approach out-performs, the current state-of-the-art in 3D vascular boundary detection (structured forests 3D), by a large margin, as well as HED applied to slices, and HED-3D while successfully localizing fine structures. With our approach, boundary detection takes about one minute on a typical 512x512x512 volume.