Journal article
Evaluating the Effect of Right-Censored End Point Transformation for Radiomic Feature Selection of Data From Patients With Oropharyngeal Cancer
JCO clinical cancer informatics, Vol.2(2), pp.1-19
12/2018
DOI: 10.1200/CCI.18.00052
PMCID: PMC6354589
PMID: 30652615
Abstract
To evaluate the effect of transforming a right-censored outcome into binary, continuous, and censored-aware representations on radiomics feature selection and subsequent prediction of overall survival (OS) and relapse-free survival (RFS) of patients with oropharyngeal cancer.
Different feature selection techniques were applied using a binary outcome indicating event occurrence before median follow-up time, a continuous outcome using the Martingale residuals from a proportional hazards model, and the raw right-censored time-to-event outcome. Radiomic signatures combined with clinical variables were used for risk prediction. Three metrics for accuracy and calibration were used to evaluate eight feature selectors and six predictive models.
Feature selection across 529 patients on more than 3,800 radiomic features resulted in increases ranging from 0.01 to 0.11 in C-index and area under the curve (AUC) scores compared with clinical features alone. The ensemble model yielded the best scores for AUC and C-index (often > 0.7) and calibration (Nam-D'Agostino test statistic often < 15.5 with 8 df). The random forest feature selectors achieved the best performance considering all metrics. Random regression forest performed the best in OS prediction with the ensemble model (AUC, 0.75; C-index, 0.76; calibration, 8.7). Random survival forest performed the best in RFS prediction with the ensemble model (AUC, 0.71; C-index, 0.68; calibration, 19.1).
Including a radiomic signature results in better prediction than using only clinical data. Signatures generated randomly or without considering the outcome result in poor calibration scores. The random forest feature selectors for each of the three transformations typically selected the greatest number of features and produced the best predictions at acceptable calibration levels. In particular, random regression forest and random survival forest performed best for OS and RFS, respectively.
Details
- Title: Subtitle
- Evaluating the Effect of Right-Censored End Point Transformation for Radiomic Feature Selection of Data From Patients With Oropharyngeal Cancer
- Creators
- Luka Zdilar - University of IowaDavid M Vock - University of MinnesotaG Elisabeta Marai - University of Illinois ChicagoClifton D Fuller - University of Illinois ChicagoAbdallah S R Mohamed - University of Illinois ChicagoHesham Elhalawani - University of Illinois ChicagoBaher Ahmed Elgohari - University of Illinois ChicagoCarly Tiras - University of Illinois ChicagoAustin Miller - University of Illinois ChicagoGuadalupe Canahuate - University of Iowa
- Resource Type
- Journal article
- Publication Details
- JCO clinical cancer informatics, Vol.2(2), pp.1-19
- DOI
- 10.1200/CCI.18.00052
- PMID
- 30652615
- PMCID
- PMC6354589
- NLM abbreviation
- JCO Clin Cancer Inform
- ISSN
- 2473-4276
- eISSN
- 2473-4276
- Grant note
- R01 CA218148 / NCI NIH HHS R01 CA225190 / NCI NIH HHS P30 CA016672 / NCI NIH HHS R01 DE025248 / NIDCR NIH HHS R01 LM012527 / NLM NIH HHS R56 DE025248 / NIDCR NIH HHS R01 CA214825 / NCI NIH HHS P50 CA097007 / NCI NIH HHS
- Language
- English
- Date published
- 12/2018
- Academic Unit
- Electrical and Computer Engineering
- Record Identifier
- 9984197197702771
Metrics
34 Record Views