Automatic Pancreas Segmentation via Coarse Location and Ensemble Learning

Shangqing Liu,Xinrui Yuan,Yuhua Ai,Shujun Liang,Shaohua Feng,Yu Zhang,Runyue Hu

doi:10.1109/access.2019.2961125

Abstract

Automatic and reliable segmentation of the pancreas is an important but difficult task for various clinical applications, such as pancreatic cancer radiotherapy and computer-aided diagnosis (CAD). The main challenges for accurate CT pancreas segmentation lie in two aspects: (1) large shape variation across different patients, and (2) low contrast and blurring around the pancreas boundary. In this paper, we propose a two-stage, ensemble-based fully convolutional neural network (FCN) to solve the challenging pancreas segmentation problem in CT images. First, candidate region generation is performed by classifying patches generated by superpixels. Second, five FCNs based on the U-Net architecture are trained with different objective functions. For each network, 2.5D slices are used as the input to provide 3D image information complementarily without the need for computationally expensive 3D convolutions. Then, an ensemble model is utilized to combine the five output segmentation maps and achieve the final segmentation. The proposed method is extensively evaluated on a publicly available dataset of 82 manually segmented CT volumes via 4-fold cross-validation. Experimental results show its superior performance compared with several state-of-the-art methods with a Dice coefficient of 84.10±4.91% and Jaccard coefficient of 72.86±6.89%.

Highlights

The pancreas, as an important organ of the human body, has internal and external secretion functions and is susceptible to various diseases
To segment the pancreas accurately in CT images, we propose an ensemble-based multiloss fully convolutional neural network (FCN)
The key points of this novel approach includes three folds: (1) coarse pancreas location via superpixel over-segmentation and classification; (2) an ensemble model combined with five FCNs, which were trained with different objective functions; (3) 2.5D image input

Summary

Introduction

The pancreas, as an important organ of the human body, has internal and external secretion functions and is susceptible to various diseases. Pancreatic cancer, which is one of the most prevalent cancers in the world, is a devastating malignant disease with a median survival of 3–6 months and a 5-year survival rate of less than 5% [1]. Contrast-enhanced CT is the worldwide imaging modality of choice for pancreatic disease evaluation and may be the best modality of the resectability of pancreatic cancer [1]. The segmentation of pancreas in CT images can support clinical workflows, including pancreas cancer diagnosis, treatment planning, and surgical assistance, in multiple domains [2]. A robust, accurate, and automatic segmentation method for the pancreas is worth exploring

Objectives

Methods

Results

Conclusion