Interactive Radiotherapy Target Delineation with 3D-Fused Context Propagation

TL;DR

This paper introduces a 3D-fused context propagation method that allows experts to efficiently refine CNN-based radiotherapy target predictions by editing only a few slices, improving accuracy without retraining models.

Contribution

The paper presents a novel interactive refinement technique that propagates user edits across the 3D volume using high-level features, without modifying existing CNN architectures.

Findings

Effective improvement of segmentation accuracy with minimal user edits

Applicable to different CNN architectures and datasets

Enhances clinical workflow for radiotherapy planning

Abstract

Gross tumor volume (GTV) delineation on tomography medical imaging is crucial for radiotherapy planning and cancer diagnosis. Convolutional neural networks (CNNs) has been predominated on automatic 3D medical segmentation tasks, including contouring the radiotherapy target given 3D CT volume. While CNNs may provide feasible outcome, in clinical scenario, double-check and prediction refinement by experts is still necessary because of CNNs' inconsistent performance on unexpected patient cases. To provide experts an efficient way to modify the CNN predictions without retrain the model, we propose 3D-fused context propagation, which propagates any edited slice to the whole 3D volume. By considering the high-level feature maps, the radiation oncologists would only required to edit few slices to guide the correction and refine the whole prediction volume. Specifically, we leverage the backpropagation for activation technique to convey the user editing information backwardly to the latent space and generate new prediction based on the updated and original feature. During the interaction, our proposed approach reuses the extant extracted features and does not alter the existing 3D CNN model architectures, avoiding the perturbation on other predictions. The proposed method is evaluated on two published radiotherapy target contouring datasets of nasopharyngeal and esophageal cancer. The experimental results demonstrate that our proposed method is able to further effectively improve the existing segmentation prediction from different model architectures given oncologists' interactive inputs.