Application of simultaneous uncertainty quantification and segmentation for oropharyngeal cancer use-case with Bayesian deep learning

Abstract

Background Radiotherapy is a core treatment modality for oropharyngeal cancer (OPC), where the primary gross tumor volume (GTVp) is manually segmented with high interobserver variability. This calls for reliable and trustworthy automated tools in clinician workflow. Therefore, accurate uncertainty quantification and its downstream utilization is critical. Methods Here we propose uncertainty-aware deep learning for OPC GTVp segmentation, and illustrate the utility of uncertainty in multiple applications. We examine two Bayesian deep learning (BDL) models and eight uncertainty measures, and utilize a large multi-institute dataset of 292 PET/CT scans to systematically analyze our approach. Results We show that our uncertainty-based approach accurately predicts the quality of the deep learning segmentation in 86.6% of cases, identifies low performance cases for semi-automated correction, and visualizes regions of the scans where the segmentations likely fail. Conclusions Our BDL-based analysis provides a first-step towards more widespread implementation of uncertainty quantification in OPC GTVp segmentation.

Enhancing the reliability of deep learning-based head and neck tumour segmentation using uncertainty estimation with multi-modal images

“…Although MC Dropout only requires training a model once, at least 10 samples are needed for adequate calibration, and a high dropout rate may negatively impact segmentation accuracy. Deep Ensemble was anticipated to enhance both the model's overall robustness and its uncertainty estimation capabilities (Fort et al 2019, Sahlsten et al 2024. The Ensemble method outperforms MC Dropout and the Baseline in terms of segmentation accuracy and calibration improvement, however, it requires a minimum of five training and inference iterations.…”