Artificial Neural Network-Based Variable Importance Analysis of Prognostic Factors Related to Radiation Pneumonitis in Patients with Lung Cancer: Preliminary Study

Eunbin Ju, Kwang Hyeon Kim, Kyung Hwan Chang, Jang Bo Shim, Chul Yong Kim, Nam Kwon Lee, Suk Lee, Chun Gun Park

Research output: Contribution to journalArticle

Abstract

Radiation pneumonitis (RP) is a major radiation-induced lung injury in patients with lung cancer. When an extremely high risk of severe RP is predicted, it is a critical issue that severely affects radiation dose and treatment planning. It is essential to assess which prognostic factor affects the occurrence of RP before initiating radiation treatment. In this study, we aimed to identify the variable importance (VI) of prognostic factors related to RP by using an artificial neural network (ANN) containing complex association between several prognostic factors. We reviewed 110 cases in patients with lung cancer who received radiation therapy (RT) from August 2000 to December 2018. The multi-layer perceptron algorithm, which is a back-propagation ANN algorithm, was implemented by SPSS Modeler (verl3.1, IBM SPSS Inc., Chicago, IL). The fifteen input variables were set, and the target variable was the occurrence of RP. The VI, which indicates the effect of each prognostic factor on the occurrence of RP, was analyzed by the variance-based method. To evaluate the VI of the ANN, we qualitatively compared the VI of the ANN with the odds ratios (OR) obtained from previously published literature. Patients who had an RP grade >2 comprised 13.6%. The accuracy of the ANN model was 76.92% and the and the area under the curve (AUC) was 0.774. From the results of the VI from the ANN, mean lung dose (MLD) had the highest VI among the prognostic factors at 15.96%. The OR of interstitial lung disease (ILD) was the highest at 25.70. While the VI of the ANN includes complex associations of the prognostic factors, ORs include independent associations between exposure and outcome. This preliminary study is meaningful as it proposes a method to quantify the VI of prognostic factors related to patient outcome using even simple ML algorithm. By further acquiring the more refined big data, optimizing the ML model, and performing clinical validation, the accurate VI can be adopted for personalized RT planning.

Original languageEnglish
Pages (from-to)277-282
Number of pages6
JournalJournal of the Korean Physical Society
Volume75
Issue number4
DOIs
Publication statusPublished - 2019 Aug 1

Fingerprint

lungs
cancer
radiation
occurrences
planning
radiation therapy
dosage
self organizing systems
grade

Keywords

  • Artificial neural network
  • Lung cancer
  • Radiation pneumonitis
  • Variable importance

ASJC Scopus subject areas

  • Physics and Astronomy(all)

Cite this

Artificial Neural Network-Based Variable Importance Analysis of Prognostic Factors Related to Radiation Pneumonitis in Patients with Lung Cancer : Preliminary Study. / Ju, Eunbin; Kim, Kwang Hyeon; Chang, Kyung Hwan; Shim, Jang Bo; Kim, Chul Yong; Lee, Nam Kwon; Lee, Suk; Park, Chun Gun.

In: Journal of the Korean Physical Society, Vol. 75, No. 4, 01.08.2019, p. 277-282.

Research output: Contribution to journalArticle

@article{e803085b5d944e7cb968cc71ee45a23c,
title = "Artificial Neural Network-Based Variable Importance Analysis of Prognostic Factors Related to Radiation Pneumonitis in Patients with Lung Cancer: Preliminary Study",
abstract = "Radiation pneumonitis (RP) is a major radiation-induced lung injury in patients with lung cancer. When an extremely high risk of severe RP is predicted, it is a critical issue that severely affects radiation dose and treatment planning. It is essential to assess which prognostic factor affects the occurrence of RP before initiating radiation treatment. In this study, we aimed to identify the variable importance (VI) of prognostic factors related to RP by using an artificial neural network (ANN) containing complex association between several prognostic factors. We reviewed 110 cases in patients with lung cancer who received radiation therapy (RT) from August 2000 to December 2018. The multi-layer perceptron algorithm, which is a back-propagation ANN algorithm, was implemented by SPSS Modeler (verl3.1, IBM SPSS Inc., Chicago, IL). The fifteen input variables were set, and the target variable was the occurrence of RP. The VI, which indicates the effect of each prognostic factor on the occurrence of RP, was analyzed by the variance-based method. To evaluate the VI of the ANN, we qualitatively compared the VI of the ANN with the odds ratios (OR) obtained from previously published literature. Patients who had an RP grade >2 comprised 13.6{\%}. The accuracy of the ANN model was 76.92{\%} and the and the area under the curve (AUC) was 0.774. From the results of the VI from the ANN, mean lung dose (MLD) had the highest VI among the prognostic factors at 15.96{\%}. The OR of interstitial lung disease (ILD) was the highest at 25.70. While the VI of the ANN includes complex associations of the prognostic factors, ORs include independent associations between exposure and outcome. This preliminary study is meaningful as it proposes a method to quantify the VI of prognostic factors related to patient outcome using even simple ML algorithm. By further acquiring the more refined big data, optimizing the ML model, and performing clinical validation, the accurate VI can be adopted for personalized RT planning.",
keywords = "Artificial neural network, Lung cancer, Radiation pneumonitis, Variable importance",
author = "Eunbin Ju and Kim, {Kwang Hyeon} and Chang, {Kyung Hwan} and Shim, {Jang Bo} and Kim, {Chul Yong} and Lee, {Nam Kwon} and Suk Lee and Park, {Chun Gun}",
year = "2019",
month = "8",
day = "1",
doi = "10.3938/jkps.75.277",
language = "English",
volume = "75",
pages = "277--282",
journal = "Journal of the Korean Physical Society",
issn = "0374-4884",
publisher = "Korean Physical Society",
number = "4",

}

TY - JOUR

T1 - Artificial Neural Network-Based Variable Importance Analysis of Prognostic Factors Related to Radiation Pneumonitis in Patients with Lung Cancer

T2 - Preliminary Study

AU - Ju, Eunbin

AU - Kim, Kwang Hyeon

AU - Chang, Kyung Hwan

AU - Shim, Jang Bo

AU - Kim, Chul Yong

AU - Lee, Nam Kwon

AU - Lee, Suk

AU - Park, Chun Gun

PY - 2019/8/1

Y1 - 2019/8/1

N2 - Radiation pneumonitis (RP) is a major radiation-induced lung injury in patients with lung cancer. When an extremely high risk of severe RP is predicted, it is a critical issue that severely affects radiation dose and treatment planning. It is essential to assess which prognostic factor affects the occurrence of RP before initiating radiation treatment. In this study, we aimed to identify the variable importance (VI) of prognostic factors related to RP by using an artificial neural network (ANN) containing complex association between several prognostic factors. We reviewed 110 cases in patients with lung cancer who received radiation therapy (RT) from August 2000 to December 2018. The multi-layer perceptron algorithm, which is a back-propagation ANN algorithm, was implemented by SPSS Modeler (verl3.1, IBM SPSS Inc., Chicago, IL). The fifteen input variables were set, and the target variable was the occurrence of RP. The VI, which indicates the effect of each prognostic factor on the occurrence of RP, was analyzed by the variance-based method. To evaluate the VI of the ANN, we qualitatively compared the VI of the ANN with the odds ratios (OR) obtained from previously published literature. Patients who had an RP grade >2 comprised 13.6%. The accuracy of the ANN model was 76.92% and the and the area under the curve (AUC) was 0.774. From the results of the VI from the ANN, mean lung dose (MLD) had the highest VI among the prognostic factors at 15.96%. The OR of interstitial lung disease (ILD) was the highest at 25.70. While the VI of the ANN includes complex associations of the prognostic factors, ORs include independent associations between exposure and outcome. This preliminary study is meaningful as it proposes a method to quantify the VI of prognostic factors related to patient outcome using even simple ML algorithm. By further acquiring the more refined big data, optimizing the ML model, and performing clinical validation, the accurate VI can be adopted for personalized RT planning.

AB - Radiation pneumonitis (RP) is a major radiation-induced lung injury in patients with lung cancer. When an extremely high risk of severe RP is predicted, it is a critical issue that severely affects radiation dose and treatment planning. It is essential to assess which prognostic factor affects the occurrence of RP before initiating radiation treatment. In this study, we aimed to identify the variable importance (VI) of prognostic factors related to RP by using an artificial neural network (ANN) containing complex association between several prognostic factors. We reviewed 110 cases in patients with lung cancer who received radiation therapy (RT) from August 2000 to December 2018. The multi-layer perceptron algorithm, which is a back-propagation ANN algorithm, was implemented by SPSS Modeler (verl3.1, IBM SPSS Inc., Chicago, IL). The fifteen input variables were set, and the target variable was the occurrence of RP. The VI, which indicates the effect of each prognostic factor on the occurrence of RP, was analyzed by the variance-based method. To evaluate the VI of the ANN, we qualitatively compared the VI of the ANN with the odds ratios (OR) obtained from previously published literature. Patients who had an RP grade >2 comprised 13.6%. The accuracy of the ANN model was 76.92% and the and the area under the curve (AUC) was 0.774. From the results of the VI from the ANN, mean lung dose (MLD) had the highest VI among the prognostic factors at 15.96%. The OR of interstitial lung disease (ILD) was the highest at 25.70. While the VI of the ANN includes complex associations of the prognostic factors, ORs include independent associations between exposure and outcome. This preliminary study is meaningful as it proposes a method to quantify the VI of prognostic factors related to patient outcome using even simple ML algorithm. By further acquiring the more refined big data, optimizing the ML model, and performing clinical validation, the accurate VI can be adopted for personalized RT planning.

KW - Artificial neural network

KW - Lung cancer

KW - Radiation pneumonitis

KW - Variable importance

UR - http://www.scopus.com/inward/record.url?scp=85071675974&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85071675974&partnerID=8YFLogxK

U2 - 10.3938/jkps.75.277

DO - 10.3938/jkps.75.277

M3 - Article

AN - SCOPUS:85071675974

VL - 75

SP - 277

EP - 282

JO - Journal of the Korean Physical Society

JF - Journal of the Korean Physical Society

SN - 0374-4884

IS - 4

ER -