Dense Volume-to-Volume Vascular Boundary Detection
This work addresses the problem of precise boundary detection in medical imaging for clinicians, though it appears incremental as it builds on existing 2D edge detection methods.
The paper tackles 3D vascular boundary detection in medical images by introducing I2I-3D, a 3D-CNN architecture, which outperforms state-of-the-art methods like structured forests 3D and HED-3D by a large margin, achieving efficient processing in about one minute per volume.
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.