skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Statistical-learning strategies generate only modestly performing predictive models for urinary symptoms following external beam radiotherapy of the prostate: A comparison of conventional and machine-learning methods

Abstract

Purpose: Given the paucity of available data concerning radiotherapy-induced urinary toxicity, it is important to ensure derivation of the most robust models with superior predictive performance. This work explores multiple statistical-learning strategies for prediction of urinary symptoms following external beam radiotherapy of the prostate. Methods: The performance of logistic regression, elastic-net, support-vector machine, random forest, neural network, and multivariate adaptive regression splines (MARS) to predict urinary symptoms was analyzed using data from 754 participants accrued by TROG03.04-RADAR. Predictive features included dose-surface data, comorbidities, and medication-intake. Four symptoms were analyzed: dysuria, haematuria, incontinence, and frequency, each with three definitions (grade ≥ 1, grade ≥ 2 and longitudinal) with event rate between 2.3% and 76.1%. Repeated cross-validations producing matched models were implemented. A synthetic minority oversampling technique was utilized in endpoints with rare events. Parameter optimization was performed on the training data. Area under the receiver operating characteristic curve (AUROC) was used to compare performance using sample size to detect differences of ≥0.05 at the 95% confidence level. Results: Logistic regression, elastic-net, random forest, MARS, and support-vector machine were the highest-performing statistical-learning strategies in 3, 3, 3, 2, and 1 endpoints, respectively. Logistic regression, MARS, elastic-net, random forest, neural network, and support-vectormore » machine were the best, or were not significantly worse than the best, in 7, 7, 5, 5, 3, and 1 endpoints. The best-performing statistical model was for dysuria grade ≥ 1 with AUROC ± standard deviation of 0.649 ± 0.074 using MARS. For longitudinal frequency and dysuria grade ≥ 1, all strategies produced AUROC>0.6 while all haematuria endpoints and longitudinal incontinence models produced AUROC<0.6. Conclusions: Logistic regression and MARS were most likely to be the best-performing strategy for the prediction of urinary symptoms with elastic-net and random forest producing competitive results. The predictive power of the models was modest and endpoint-dependent. New features, including spatial dose maps, may be necessary to achieve better models.« less

Authors:
 [1];  [2];  [3];  [4];  [5];  [6];  [7]
  1. School of Physics, University of Western Australia, Western Australia 6009, Australia and School of Health Sciences, National University of Malaysia, Bangi 43600 (Malaysia)
  2. School of Physics, University of Western Australia, Western Australia 6009, Australia and Department of Radiation Oncology, Sir Charles Gairdner Hospital, Western Australia 6008 (Australia)
  3. Institute for Health Research, University of Notre Dame, Fremantle, Western Australia 6959 (Australia)
  4. School of Physics, University of Western Australia, Western Australia 6009 (Australia)
  5. Department of Radiation Oncology, Sir Charles Gairdner Hospital, Western Australia 6008 (Australia)
  6. Department of Radiation Oncology, Sir Charles Gairdner Hospital, Western Australia 6008, Australia and School of Surgery, University of Western Australia, Western Australia 6009 (Australia)
  7. School of Medicine and Public Health, University of Newcastle, New South Wales 2308 (Australia)
Publication Date:
OSTI Identifier:
22620865
Resource Type:
Journal Article
Journal Name:
Medical Physics
Additional Journal Information:
Journal Volume: 43; Journal Issue: 5; Other Information: (c) 2016 American Association of Physicists in Medicine; Country of input: International Atomic Energy Agency (IAEA); Journal ID: ISSN 0094-2405
Country of Publication:
United States
Language:
English
Subject:
60 APPLIED LIFE SCIENCES; 61 RADIATION PROTECTION AND DOSIMETRY; COMPARATIVE EVALUATIONS; LEARNING; MARS PLANET; MULTIVARIATE ANALYSIS; NEURAL NETWORKS; PERFORMANCE; PROSTATE; RADIATION DOSES; RADIOTHERAPY; SYMPTOMS

Citation Formats

Yahya, Noorazrul, Ebert, Martin A., Bulsara, Max, House, Michael J., Kennedy, Angel, Joseph, David J., and Denham, James W. Statistical-learning strategies generate only modestly performing predictive models for urinary symptoms following external beam radiotherapy of the prostate: A comparison of conventional and machine-learning methods. United States: N. p., 2016. Web. doi:10.1118/1.4944738.
Yahya, Noorazrul, Ebert, Martin A., Bulsara, Max, House, Michael J., Kennedy, Angel, Joseph, David J., & Denham, James W. Statistical-learning strategies generate only modestly performing predictive models for urinary symptoms following external beam radiotherapy of the prostate: A comparison of conventional and machine-learning methods. United States. https://doi.org/10.1118/1.4944738
Yahya, Noorazrul, Ebert, Martin A., Bulsara, Max, House, Michael J., Kennedy, Angel, Joseph, David J., and Denham, James W. 2016. "Statistical-learning strategies generate only modestly performing predictive models for urinary symptoms following external beam radiotherapy of the prostate: A comparison of conventional and machine-learning methods". United States. https://doi.org/10.1118/1.4944738.
@article{osti_22620865,
title = {Statistical-learning strategies generate only modestly performing predictive models for urinary symptoms following external beam radiotherapy of the prostate: A comparison of conventional and machine-learning methods},
author = {Yahya, Noorazrul and Ebert, Martin A. and Bulsara, Max and House, Michael J. and Kennedy, Angel and Joseph, David J. and Denham, James W.},
abstractNote = {Purpose: Given the paucity of available data concerning radiotherapy-induced urinary toxicity, it is important to ensure derivation of the most robust models with superior predictive performance. This work explores multiple statistical-learning strategies for prediction of urinary symptoms following external beam radiotherapy of the prostate. Methods: The performance of logistic regression, elastic-net, support-vector machine, random forest, neural network, and multivariate adaptive regression splines (MARS) to predict urinary symptoms was analyzed using data from 754 participants accrued by TROG03.04-RADAR. Predictive features included dose-surface data, comorbidities, and medication-intake. Four symptoms were analyzed: dysuria, haematuria, incontinence, and frequency, each with three definitions (grade ≥ 1, grade ≥ 2 and longitudinal) with event rate between 2.3% and 76.1%. Repeated cross-validations producing matched models were implemented. A synthetic minority oversampling technique was utilized in endpoints with rare events. Parameter optimization was performed on the training data. Area under the receiver operating characteristic curve (AUROC) was used to compare performance using sample size to detect differences of ≥0.05 at the 95% confidence level. Results: Logistic regression, elastic-net, random forest, MARS, and support-vector machine were the highest-performing statistical-learning strategies in 3, 3, 3, 2, and 1 endpoints, respectively. Logistic regression, MARS, elastic-net, random forest, neural network, and support-vector machine were the best, or were not significantly worse than the best, in 7, 7, 5, 5, 3, and 1 endpoints. The best-performing statistical model was for dysuria grade ≥ 1 with AUROC ± standard deviation of 0.649 ± 0.074 using MARS. For longitudinal frequency and dysuria grade ≥ 1, all strategies produced AUROC>0.6 while all haematuria endpoints and longitudinal incontinence models produced AUROC<0.6. Conclusions: Logistic regression and MARS were most likely to be the best-performing strategy for the prediction of urinary symptoms with elastic-net and random forest producing competitive results. The predictive power of the models was modest and endpoint-dependent. New features, including spatial dose maps, may be necessary to achieve better models.},
doi = {10.1118/1.4944738},
url = {https://www.osti.gov/biblio/22620865}, journal = {Medical Physics},
issn = {0094-2405},
number = 5,
volume = 43,
place = {United States},
year = {Sun May 15 00:00:00 EDT 2016},
month = {Sun May 15 00:00:00 EDT 2016}
}