Fully Automated Organ Segmentation in Male Pelvic CT Images

TL;DR

This paper introduces a fully automated deep learning workflow for segmenting prostate and surrounding organs in male pelvic CT images, achieving high accuracy in a clinical dataset.

Contribution

It develops a novel two-stage 2D and 3D deep learning architecture with residual networks for precise organ segmentation in pelvic CT scans.

Findings

Achieved mean Dice coefficients of 90% for prostate

Achieved mean Dice coefficients of 96% for left femoral head

Achieved mean Dice coefficients of 84% for rectum

Abstract

Accurate segmentation of prostate and surrounding organs at risk is important for prostate cancer radiotherapy treatment planning. We present a fully automated workflow for male pelvic CT image segmentation using deep learning. The architecture consists of a 2D localization network followed by a 3D segmentation network for volumetric segmentation of prostate, bladder, rectum, and femoral heads. We used a multi-channel 2D U-Net followed by a 3D U-Net with encoding arm modified with aggregated residual networks, known as ResNeXt. The models were trained and tested on a pelvic CT image dataset comprising 136 patients. Test results show that 3D U-Net based segmentation achieves mean (SD) Dice coefficient values of 90 (2.0)% ,96 (3.0)%, 95 (1.3)%, 95 (1.5)%, and 84 (3.7)% for prostate, left femoral head, right femoral head, bladder, and rectum, respectively, using the proposed fully automated segmentation method.