Evolving Horizons in Radiotherapy Auto-Contouring: Distilling Insights, Embracing Data-Centric Frameworks, and Moving Beyond Geometric Quantification

TL;DR

This paper reviews recent advances in radiotherapy auto-contouring, highlighting the importance of high-quality data, the effectiveness of models with limited data, and the plateauing of performance, advocating for data-centric approaches.

Contribution

It synthesizes current insights and emphasizes the shift towards data-centric frameworks to enhance clinical adoption of auto-contouring in radiotherapy.

Findings

High-quality training data is crucial.

Auto-contouring performs well with limited data.

Performance improvements are plateauing.

Abstract

Deep learning has significantly advanced the potential for automated contouring in radiotherapy planning. In this manuscript, guided by contemporary literature, we underscore three key insights: (1) High-quality training data is essential for auto-contouring algorithms; (2) Auto-contouring models demonstrate commendable performance even with limited medical image data; (3) The quantitative performance of auto-contouring is reaching a plateau. Given these insights, we emphasize the need for the radiotherapy research community to embrace data-centric approaches to further foster clinical adoption of auto-contouring technologies.