Skip to main content

Investigation of the support vector machine algorithm to predict lung radiation-induced pneumonitis.

Publication ,  Journal Article
Chen, S; Zhou, S; Yin, F-F; Marks, LB; Das, SK
Published in: Med Phys
October 2007

The purpose of this study is to build and test a support vector machine (SVM) model to predict for the occurrence of lung radiation-induced Grade 2+ pneumonitis. SVM is a sophisticated statistical technique capable of separating the two categories of patients (with/without pneumonitis) using a boundary defined by a complex hypersurface. Despite the complexity, the SVM boundary is only minimally influenced by outliers that are difficult to separate. By contrast, the simple hyperplane boundary computed by the more commonly used and related linear discriminant analysis method is heavily influenced by outliers. Two SVM models were built using data from 219 patients with lung cancer treated using radiotherapy (34 diagnosed with pneumonitis). One model (SVM(all)) selected input features from all dose and non-dose factors. For comparison, the other model (SVM(dose)) selected input features only from lung dose-volume factors. Model predictive ability was evaluated using ten-fold cross-validation and receiver operating characteristics (ROC) analysis. For the model SVM(all), the area under the cross-validated ROC curve was 0.76 (sensitivity/specificity = 74%/75%). Compared to the corresponding SVM(dose) area of 0.71 (sensitivity/specificity = 68%/68%), the predictive ability of SVM(all) was improved, indicating that non-dose features are important contributors to separating patients with and without pneumonitis. Among the input features selected by model SVM(all), the two with highest importance for predicting lung pneumonitis were: (a) generalized equivalent uniform doses close to the mean lung dose, and (b) chemotherapy prior to radiotherapy. The model SVM(all) is publicly available via internet access.

Duke Scholars

Published In

Med Phys

DOI

ISSN

0094-2405

Publication Date

October 2007

Volume

34

Issue

10

Start / End Page

3808 / 3814

Location

United States

Related Subject Headings

  • Software
  • Sensitivity and Specificity
  • Radiotherapy Planning, Computer-Assisted
  • Radiotherapy Dosage
  • Radiation Pneumonitis
  • ROC Curve
  • Nuclear Medicine & Medical Imaging
  • Models, Statistical
  • Lung
  • Humans
 

Citation

APA
Chicago
ICMJE
MLA
NLM
Chen, S., Zhou, S., Yin, F.-F., Marks, L. B., & Das, S. K. (2007). Investigation of the support vector machine algorithm to predict lung radiation-induced pneumonitis. Med Phys, 34(10), 3808–3814. https://doi.org/10.1118/1.2776669
Chen, Shifeng, Sumin Zhou, Fang-Fang Yin, Lawrence B. Marks, and Shiva K. Das. “Investigation of the support vector machine algorithm to predict lung radiation-induced pneumonitis.Med Phys 34, no. 10 (October 2007): 3808–14. https://doi.org/10.1118/1.2776669.
Chen S, Zhou S, Yin F-F, Marks LB, Das SK. Investigation of the support vector machine algorithm to predict lung radiation-induced pneumonitis. Med Phys. 2007 Oct;34(10):3808–14.
Chen, Shifeng, et al. “Investigation of the support vector machine algorithm to predict lung radiation-induced pneumonitis.Med Phys, vol. 34, no. 10, Oct. 2007, pp. 3808–14. Pubmed, doi:10.1118/1.2776669.
Chen S, Zhou S, Yin F-F, Marks LB, Das SK. Investigation of the support vector machine algorithm to predict lung radiation-induced pneumonitis. Med Phys. 2007 Oct;34(10):3808–3814.

Published In

Med Phys

DOI

ISSN

0094-2405

Publication Date

October 2007

Volume

34

Issue

10

Start / End Page

3808 / 3814

Location

United States

Related Subject Headings

  • Software
  • Sensitivity and Specificity
  • Radiotherapy Planning, Computer-Assisted
  • Radiotherapy Dosage
  • Radiation Pneumonitis
  • ROC Curve
  • Nuclear Medicine & Medical Imaging
  • Models, Statistical
  • Lung
  • Humans