Predicting Lung Radiotherapy-Induced Pneumonitis Using a Model Combining Parametric Lyman Probit With Nonparametric Decision Trees

Shiva K. Das, Sumin Zhou, Junan Zhang, Fang Fang Yin, Mark W. Dewhirst, Lawrence B. Marks

Research output: Contribution to journalArticle

32 Citations (Scopus)

Abstract

Purpose: To develop and test a model to predict for lung radiation-induced Grade 2+ pneumonitis. Methods and Materials: The model was built from a database of 234 lung cancer patients treated with radiotherapy (RT), of whom 43 were diagnosed with pneumonitis. The model augmented the predictive capability of the parametric dose-based Lyman normal tissue complication probability (LNTCP) metric by combining it with weighted nonparametric decision trees that use dose and nondose inputs. The decision trees were sequentially added to the model using a "boosting" process that enhances the accuracy of prediction. The model's predictive capability was estimated by 10-fold cross-validation. To facilitate dissemination, the cross-validation result was used to extract a simplified approximation to the complicated model architecture created by boosting. Application of the simplified model is demonstrated in two example cases. Results: The area under the model receiver operating characteristics curve for cross-validation was 0.72, a significant improvement over the LNTCP area of 0.63 (p = 0.005). The simplified model used the following variables to output a measure of injury: LNTCP, gender, histologic type, chemotherapy schedule, and treatment schedule. For a given patient RT plan, injury prediction was highest for the combination of pre-RT chemotherapy, once-daily treatment, female gender and lowest for the combination of no pre-RT chemotherapy and nonsquamous cell histologic type. Application of the simplified model to the example cases revealed that injury prediction for a given treatment plan can range from very low to very high, depending on the settings of the nondose variables. Conclusions: Radiation pneumonitis prediction was significantly enhanced by decision trees that added the influence of nondose factors to the LNTCP formulation.

Original languageEnglish (US)
Pages (from-to)1212-1221
Number of pages10
JournalInternational Journal of Radiation Oncology Biology Physics
Volume68
Issue number4
DOIs
StatePublished - Jul 15 2007

Fingerprint

Decision Trees
lungs
radiation therapy
Pneumonia
Radiotherapy
Lung
Drug Therapy
Appointments and Schedules
Wounds and Injuries
Radiation Pneumonitis
chemotherapy
ROC Curve
Lung Neoplasms
schedules
predictions
Therapeutics
Databases
Radiation
dosage
radiation

Keywords

  • Lung
  • Models
  • Pneumonitis
  • Prediction
  • Radiotherapy

ASJC Scopus subject areas

  • Radiation
  • Oncology
  • Radiology Nuclear Medicine and imaging
  • Cancer Research

Cite this

Predicting Lung Radiotherapy-Induced Pneumonitis Using a Model Combining Parametric Lyman Probit With Nonparametric Decision Trees. / Das, Shiva K.; Zhou, Sumin; Zhang, Junan; Yin, Fang Fang; Dewhirst, Mark W.; Marks, Lawrence B.

In: International Journal of Radiation Oncology Biology Physics, Vol. 68, No. 4, 15.07.2007, p. 1212-1221.

Research output: Contribution to journalArticle

@article{f8f2d4c1068c4e53b962d8ff066a9fea,
title = "Predicting Lung Radiotherapy-Induced Pneumonitis Using a Model Combining Parametric Lyman Probit With Nonparametric Decision Trees",
abstract = "Purpose: To develop and test a model to predict for lung radiation-induced Grade 2+ pneumonitis. Methods and Materials: The model was built from a database of 234 lung cancer patients treated with radiotherapy (RT), of whom 43 were diagnosed with pneumonitis. The model augmented the predictive capability of the parametric dose-based Lyman normal tissue complication probability (LNTCP) metric by combining it with weighted nonparametric decision trees that use dose and nondose inputs. The decision trees were sequentially added to the model using a {"}boosting{"} process that enhances the accuracy of prediction. The model's predictive capability was estimated by 10-fold cross-validation. To facilitate dissemination, the cross-validation result was used to extract a simplified approximation to the complicated model architecture created by boosting. Application of the simplified model is demonstrated in two example cases. Results: The area under the model receiver operating characteristics curve for cross-validation was 0.72, a significant improvement over the LNTCP area of 0.63 (p = 0.005). The simplified model used the following variables to output a measure of injury: LNTCP, gender, histologic type, chemotherapy schedule, and treatment schedule. For a given patient RT plan, injury prediction was highest for the combination of pre-RT chemotherapy, once-daily treatment, female gender and lowest for the combination of no pre-RT chemotherapy and nonsquamous cell histologic type. Application of the simplified model to the example cases revealed that injury prediction for a given treatment plan can range from very low to very high, depending on the settings of the nondose variables. Conclusions: Radiation pneumonitis prediction was significantly enhanced by decision trees that added the influence of nondose factors to the LNTCP formulation.",
keywords = "Lung, Models, Pneumonitis, Prediction, Radiotherapy",
author = "Das, {Shiva K.} and Sumin Zhou and Junan Zhang and Yin, {Fang Fang} and Dewhirst, {Mark W.} and Marks, {Lawrence B.}",
year = "2007",
month = "7",
day = "15",
doi = "10.1016/j.ijrobp.2007.03.064",
language = "English (US)",
volume = "68",
pages = "1212--1221",
journal = "International Journal of Radiation Oncology Biology Physics",
issn = "0360-3016",
publisher = "Elsevier Inc.",
number = "4",

}

TY - JOUR

T1 - Predicting Lung Radiotherapy-Induced Pneumonitis Using a Model Combining Parametric Lyman Probit With Nonparametric Decision Trees

AU - Das, Shiva K.

AU - Zhou, Sumin

AU - Zhang, Junan

AU - Yin, Fang Fang

AU - Dewhirst, Mark W.

AU - Marks, Lawrence B.

PY - 2007/7/15

Y1 - 2007/7/15

N2 - Purpose: To develop and test a model to predict for lung radiation-induced Grade 2+ pneumonitis. Methods and Materials: The model was built from a database of 234 lung cancer patients treated with radiotherapy (RT), of whom 43 were diagnosed with pneumonitis. The model augmented the predictive capability of the parametric dose-based Lyman normal tissue complication probability (LNTCP) metric by combining it with weighted nonparametric decision trees that use dose and nondose inputs. The decision trees were sequentially added to the model using a "boosting" process that enhances the accuracy of prediction. The model's predictive capability was estimated by 10-fold cross-validation. To facilitate dissemination, the cross-validation result was used to extract a simplified approximation to the complicated model architecture created by boosting. Application of the simplified model is demonstrated in two example cases. Results: The area under the model receiver operating characteristics curve for cross-validation was 0.72, a significant improvement over the LNTCP area of 0.63 (p = 0.005). The simplified model used the following variables to output a measure of injury: LNTCP, gender, histologic type, chemotherapy schedule, and treatment schedule. For a given patient RT plan, injury prediction was highest for the combination of pre-RT chemotherapy, once-daily treatment, female gender and lowest for the combination of no pre-RT chemotherapy and nonsquamous cell histologic type. Application of the simplified model to the example cases revealed that injury prediction for a given treatment plan can range from very low to very high, depending on the settings of the nondose variables. Conclusions: Radiation pneumonitis prediction was significantly enhanced by decision trees that added the influence of nondose factors to the LNTCP formulation.

AB - Purpose: To develop and test a model to predict for lung radiation-induced Grade 2+ pneumonitis. Methods and Materials: The model was built from a database of 234 lung cancer patients treated with radiotherapy (RT), of whom 43 were diagnosed with pneumonitis. The model augmented the predictive capability of the parametric dose-based Lyman normal tissue complication probability (LNTCP) metric by combining it with weighted nonparametric decision trees that use dose and nondose inputs. The decision trees were sequentially added to the model using a "boosting" process that enhances the accuracy of prediction. The model's predictive capability was estimated by 10-fold cross-validation. To facilitate dissemination, the cross-validation result was used to extract a simplified approximation to the complicated model architecture created by boosting. Application of the simplified model is demonstrated in two example cases. Results: The area under the model receiver operating characteristics curve for cross-validation was 0.72, a significant improvement over the LNTCP area of 0.63 (p = 0.005). The simplified model used the following variables to output a measure of injury: LNTCP, gender, histologic type, chemotherapy schedule, and treatment schedule. For a given patient RT plan, injury prediction was highest for the combination of pre-RT chemotherapy, once-daily treatment, female gender and lowest for the combination of no pre-RT chemotherapy and nonsquamous cell histologic type. Application of the simplified model to the example cases revealed that injury prediction for a given treatment plan can range from very low to very high, depending on the settings of the nondose variables. Conclusions: Radiation pneumonitis prediction was significantly enhanced by decision trees that added the influence of nondose factors to the LNTCP formulation.

KW - Lung

KW - Models

KW - Pneumonitis

KW - Prediction

KW - Radiotherapy

UR - http://www.scopus.com/inward/record.url?scp=34447266924&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=34447266924&partnerID=8YFLogxK

U2 - 10.1016/j.ijrobp.2007.03.064

DO - 10.1016/j.ijrobp.2007.03.064

M3 - Article

VL - 68

SP - 1212

EP - 1221

JO - International Journal of Radiation Oncology Biology Physics

JF - International Journal of Radiation Oncology Biology Physics

SN - 0360-3016

IS - 4

ER -