Towards Automatic Abdominal Multi-Organ Segmentation in Dual Energy CT using Cascaded 3D Fully Convolutional Network

TL;DR

This paper presents a cascaded 3D fully convolutional network approach for automatic multi-organ segmentation in dual energy CT scans, achieving high accuracy and demonstrating feasibility for clinical use.

Contribution

It introduces a novel cascaded 3D FCN method specifically adapted for DECT, trained on SECT data, and fine-tuned for DECT, advancing automatic multi-organ segmentation techniques.

Findings

Achieved average Dice coefficients of 93% for liver and 90% for spleen.

Demonstrated the method's feasibility and promising accuracy in DECT data.

Validated the approach with cross-validation on 42 clinical DECT scans.

Abstract

Automatic multi-organ segmentation of the dual energy computed tomography (DECT) data can be beneficial for biomedical research and clinical applications. However, it is a challenging task. Recent advances in deep learning showed the feasibility to use 3-D fully convolutional networks (FCN) for voxel-wise dense predictions in single energy computed tomography (SECT). In this paper, we proposed a 3D FCN based method for automatic multi-organ segmentation in DECT. The work was based on a cascaded FCN and a general model for the major organs trained on a large set of SECT data. We preprocessed the DECT data by using linear weighting and fine-tuned the model for the DECT data. The method was evaluated using 42 torso DECT data acquired with a clinical dual-source CT system. Four abdominal organs (liver, spleen, left and right kidneys) were evaluated. Cross-validation was tested. Effect of the weight on the accuracy was researched. In all the tests, we achieved an average Dice coefficient of 93% for the liver, 90% for the spleen, 91% for the right kidney and 89% for the left kidney, respectively. The results show our method is feasible and promising.