AUTHOR=Sepehri Shima , Tankyevych Olena , Iantsen Andrei , Visvikis Dimitris , Hatt Mathieu , Cheze Le Rest Catherine 

TITLE=Accurate Tumor Delineation vs. Rough Volume of Interest Analysis for 18F-FDG PET/CT Radiomics-Based Prognostic Modeling inNon-Small Cell Lung Cancer

JOURNAL=Frontiers in Oncology

VOLUME=Volume 11 - 2021

YEAR=2021

URL=https://www.frontiersin.org/journals/oncology/articles/10.3389/fonc.2021.726865

DOI=10.3389/fonc.2021.726865

ISSN=2234-943X

ABSTRACT=Background: The aim of this work was to investigate the ability of building prognostic models in non-small cell lung cancer (NSCLC) using radiomic features from 18FDG-PET/CT images based on a “rough” volume of interest (VOI) containing the tumor, instead of its accurate delineation, which is a significant time-consuming bottleneck of radiomic analyses.
Methods: A cohort of 138 NSCLC stage II-III patients treated with radiochemotherapy recruited retrospectively (n=87) and prospectively (n=51) was used. Two approaches were compared: first, the radiomic features were extracted from the delineated primary tumor volumes in both PET (using the automated fuzzy locally adaptive Bayesian, FLAB) and CT (using a semi-automated approach with 3D SlicerTM) components. Both delineations were carried out within previously manually defined “rough” VOIs containing the tumor and the surrounding tissues, which were exploited for the second approach: the same features were extracted from this alternative VOI. Both sets for features were then combined with clinical variables and processed through the same machine learning (ML) pipelines after splitting the cohort into training and testing sets. Logistic regression (LR), random forest (RF) and support vector machines (SVM), as well as their consensus through averaging the output probabilities, were considered for feature selection and modeling for overall survival (OS) prediction as a binary classification (either median OS or 6 months OS). Resulting models were compared in terms of balanced accuracy, sensitivity and specificity.
Results: Overall, better performance was achieved using features from delineated tumor volumes. This was observed consistently across ML algorithms and for the two clinical endpoints. However, the loss of performance was not significant, especially when a consensus of the 3 ML algorithms was considered (0.89 vs. 0.88 and 0.78 vs. 0.77).  
Conclusion: Our findings suggest it is possible to achieve a similar level of prognostic accuracy in radiomic based modeling by relying on a faster and easier VOI definition, skipping a time-consuming tumor delineation step, thus facilitating automation of the whole radiomics workflow. The associated cost is a loss of performance in the resulting models, although this loss can be greatly mitigated is a consensus of several models is relied upon.