Oropharyngeal cancer patient stratification using random forest based-learning over high-dimensional radiomic features

Harsh Patel; David M. Vock; G. Elisabeta Marai; Clifton D. Fuller; Abdallah S. R. Mohamed; Guadalupe Canahuate

doi:10.1038/s41598-021-92072-8

Back

Oropharyngeal cancer patient stratification using random forest based-learning over high-dimensional radiomic features

Journal article

Open access

Peer reviewed

Oropharyngeal cancer patient stratification using random forest based-learning over high-dimensional radiomic features

Harsh Patel, David M. Vock, G. Elisabeta Marai, Clifton D. Fuller, Abdallah S. R. Mohamed and Guadalupe Canahuate

Scientific reports, Vol.11(1), 14057

07/01/2021

DOI: 10.1038/s41598-021-92072-8

Files and links (1)

url

https://doi.org/10.1038/s41598-021-92072-8View

Published (Version of record) Open Access

Abstract

Abstract To improve risk prediction for oropharyngeal cancer (OPC) patients using cluster analysis on the radiomic features extracted from pre-treatment Computed Tomography (CT) scans. 553 OPC Patients randomly split into training (80%) and validation (20%), were classified into 2 or 3 risk groups by applying hierarchical clustering over the co-occurrence matrix obtained from a random survival forest (RSF) trained over 301 radiomic features. The cluster label was included together with other clinical data to train an ensemble model using five predictive models (Cox, random forest, RSF, logistic regression, and logistic-elastic net). Ensemble performance was evaluated over the independent test set for both recurrence free survival (RFS) and overall survival (OS). The Kaplan–Meier curves for OS stratified by cluster label show significant differences for both training and testing (p val < 0.0001). When compared to the models trained using clinical data only, the inclusion of the cluster label improves AUC test performance from .62 to .79 and from .66 to .80 for OS and RFS, respectively. The extraction of a single feature, namely a cluster label, to represent the high-dimensional radiomic feature space reduces the dimensionality and sparsity of the data. Moreover, inclusion of the cluster label improves model performance compared to clinical data only and offers comparable performance to the models including raw radiomic features.

Details

Title: Subtitle: Oropharyngeal cancer patient stratification using random forest based-learning over high-dimensional radiomic features
Creators: Harsh Patel - University of Iowa
David M. Vock - University of Minnesota
G. Elisabeta Marai - University of Illinois Chicago
Clifton D. Fuller - University of Texas MD Anderson Cancer Center
Abdallah S. R. Mohamed - University of Texas MD Anderson Cancer Center
Guadalupe Canahuate - University of Iowa
Resource Type: Journal article
Publication Details: Scientific reports, Vol.11(1), 14057
DOI: 10.1038/s41598-021-92072-8
ISSN: 2045-2322
eISSN: 2045-2322
Publisher: Nature Publishing Group
Grant note: DOI: 10.13039/100000002, name: National Institutes of Health, award: R01CA214825, R01CA214825, R01CA214825, R01CA214825, R01CA214825, R01CA214825
Language: English
Date published: 07/01/2021
Academic Unit: Electrical and Computer Engineering
Record Identifier: 9984197195902771

Metrics

6 Record Views

9 Times Cited - Web of Science