A unified 3D framework for Organs at Risk Localization and Segmentation for Radiation Therapy Planning

TL;DR

This paper presents a unified 3D framework for localizing and segmenting organs at risk in CT scans, enhancing accuracy and efficiency in radiation therapy planning by exploiting 3D contextual information.

Contribution

The work introduces a comprehensive 3D pipeline that combines organ localization and segmentation, fully utilizing 3D context in medical imaging for the first time.

Findings

Achieved a Dice score of 0.9260 on the VISCERAL dataset.

Demonstrated effective multi-organ segmentation in varying CT scan fields.

Enabled accurate organ localization using a 3D regression network.

Abstract

Automatic localization and segmentation of organs-at-risk (OAR) in CT are essential pre-processing steps in medical image analysis tasks, such as radiation therapy planning. For instance, the segmentation of OAR surrounding tumors enables the maximization of radiation to the tumor area without compromising the healthy tissues. However, the current medical workflow requires manual delineation of OAR, which is prone to errors and is annotator-dependent. In this work, we aim to introduce a unified 3D pipeline for OAR localization-segmentation rather than novel localization or segmentation architectures. To the best of our knowledge, our proposed framework fully enables the exploitation of 3D context information inherent in medical imaging. In the first step, a 3D multi-variate regression network predicts organs' centroids and bounding boxes. Secondly, 3D organ-specific segmentation networks are leveraged to generate a multi-organ segmentation map. Our method achieved an overall Dice score of $0.9260\pm 0.18 \%$ on the VISCERAL dataset containing CT scans with varying fields of view and multiple organs.