Detegmentation of Organs-at-Risk for Head and Neck Cancer

Shan Wang, Jax Luo, Xu Qiao, Nazim Haouchine, Scott Raymond

Keywords: Detegmentation, Detection, Segmentation, Foundational Model, Head and Neck Cancer

TL;DR: Detegmentation for autonomous box prompt generation in foundation segmentation models.

Abstract: Segmentation of organs-at-risk (OARs) is a critical step in radiation therapy planning for head and neck cancer (HNC). We present a practical Detegmentation framework that integrates a detection network to autonomously generate box prompts for training and testing a foundation segmentation model for OARs in HNC. Our method achieves state-of-the-art performance without clinician intervention, demonstrating its strong potential for clinical application.

Submission Number: 127