A 3D Coarse-to-Fine Framework for Volumetric Medical Image Segmentation

TL;DR

This paper introduces a 3D coarse-to-fine neural network framework that significantly improves volumetric medical image segmentation by leveraging full 3D spatial information, outperforming previous methods on pancreas datasets.

Contribution

The paper presents a novel 3D-based coarse-to-fine framework that effectively addresses data scarcity and computational challenges in volumetric segmentation, achieving state-of-the-art results.

Findings

Outperforms 2D methods by a large margin

Achieves over 2% improvement in Dice coefficient on NIH dataset

Reaches nearly 70% Dice score, demonstrating clinical reliability

Abstract

In this paper, we adopt 3D Convolutional Neural Networks to segment volumetric medical images. Although deep neural networks have been proven to be very effective on many 2D vision tasks, it is still challenging to apply them to 3D tasks due to the limited amount of annotated 3D data and limited computational resources. We propose a novel 3D-based coarse-to-fine framework to effectively and efficiently tackle these challenges. The proposed 3D-based framework outperforms the 2D counterpart to a large margin since it can leverage the rich spatial infor- mation along all three axes. We conduct experiments on two datasets which include healthy and pathological pancreases respectively, and achieve the current state-of-the-art in terms of Dice-S{\o}rensen Coefficient (DSC). On the NIH pancreas segmentation dataset, we outperform the previous best by an average of over 2%, and the worst case is improved by 7% to reach almost 70%, which indicates the reliability of our framework in clinical applications.