Radiotherapy Target Contouring with Convolutional Gated Graph Neural Network

TL;DR

This paper introduces a convolutional gated graph neural network for efficient and interactive radiotherapy target contouring in medical imaging, outperforming traditional volumetric segmentation methods.

Contribution

The paper proposes a novel convolutional recurrent Gated Graph Propagator that models slice-wise interactions and allows for interactive refinement, improving segmentation accuracy in radiotherapy planning.

Findings

State-of-the-art segmentation results on esophageal cancer dataset

Enhanced performance with interactive slice editing

Efficient modeling of volumetric data using graph neural networks

Abstract

Tomography medical imaging is essential in the clinical workflow of modern cancer radiotherapy. Radiation oncologists identify cancerous tissues, applying delineation on treatment regions throughout all image slices. This kind of task is often formulated as a volumetric segmentation task by means of 3D convolutional networks with considerable computational cost. Instead, inspired by the treating methodology of considering meaningful information across slices, we used Gated Graph Neural Network to frame this problem more efficiently. More specifically, we propose convolutional recurrent Gated Graph Propagator (GGP) to propagate high-level information through image slices, with learnable adjacency weighted matrix. Furthermore, as physicians often investigate a few specific slices to refine their decision, we model this slice-wise interaction procedure to further improve our segmentation result. This can be set by editing any slice effortlessly as updating predictions of other slices using GGP. To evaluate our method, we collect an Esophageal Cancer Radiotherapy Target Treatment Contouring dataset of 81 patients which includes tomography images with radiotherapy target. On this dataset, our convolutional graph network produces state-of-the-art results and outperforms the baselines. With the addition of interactive setting, performance is improved even further. Our method has the potential to be easily applied to diverse kinds of medical tasks with volumetric images. Incorporating both the ability to make a feasible prediction and to consider the human interactive input, the proposed method is suitable for clinical scenarios.