Radiomic Feature Stability Analysis based on Probabilistic Segmentations

TL;DR

This study investigates the stability of radiomic features derived from probabilistic segmentations in lung cancer, revealing variability among features and proposing a more robust feature selection workflow to improve prognostic models.

Contribution

The paper introduces a novel approach using probabilistic segmentations to analyze radiomic feature stability and proposes a new feature selection workflow for robustness.

Findings

Statistics features are more robust to segmentation variability.

Gray-level size zone matrix features are less robust.

Segmentation variance affects prognostic model performance.

Abstract

Identifying image features that are robust with respect to segmentation variability and domain shift is a tough challenge in radiomics. So far, this problem has mainly been tackled in test-retest analyses. In this work we analyze radiomics feature stability based on probabilistic segmentations. Based on a public lung cancer dataset, we generate an arbitrary number of plausible segmentations using a Probabilistic U-Net. From these segmentations, we extract a high number of plausible feature vectors for each lung tumor and analyze feature variance with respect to the segmentations. Our results suggest that there are groups of radiomic features that are more (e.g. statistics features) and less (e.g. gray-level size zone matrix features) robust against segmentation variability. Finally, we demonstrate that segmentation variance impacts the performance of a prognostic lung cancer survival model and propose a new and potentially more robust radiomics feature selection workflow.