3D FCN Feature Driven Regression Forest-Based Pancreas Localization and Segmentation

TL;DR

This paper introduces an automated pancreas segmentation method from CT scans using 3D FCN features and regression forests for localization, improving handling of inter-patient anatomical variations.

Contribution

It proposes a novel combined approach of 3D FCN features and regression forests for accurate pancreas localization and segmentation from CT volumes.

Findings

Achieved 60.6% Jaccard index on 146 CT scans.

Attained 73.9% Dice overlap in pancreas segmentation.

Demonstrated robustness to inter-patient anatomical variations.

Abstract

This paper presents a fully automated atlas-based pancreas segmentation method from CT volumes utilizing 3D fully convolutional network (FCN) feature-based pancreas localization. Segmentation of the pancreas is difficult because it has larger inter-patient spatial variations than other organs. Previous pancreas segmentation methods failed to deal with such variations. We propose a fully automated pancreas segmentation method that contains novel localization and segmentation. Since the pancreas neighbors many other organs, its position and size are strongly related to the positions of the surrounding organs. We estimate the position and the size of the pancreas (localized) from global features by regression forests. As global features, we use intensity differences and 3D FCN deep learned features, which include automatically extracted essential features for segmentation. We chose 3D FCN features from a trained 3D U-Net, which is trained to perform multi-organ segmentation. The global features include both the pancreas and surrounding organ information. After localization, a patient-specific probabilistic atlas-based pancreas segmentation is performed. In evaluation results with 146 CT volumes, we achieved 60.6% of the Jaccard index and 73.9% of the Dice overlap.