Decision fusion of machine learning models to predict radiotherapyinduced lung pneumonitis

Shiva K. Das, Shifeng Chen, Joseph O. Deasy, Sumin Zhou, Fang Fane Yin, Lawrence B. Marks

Research output: Chapter in Book/Report/Conference proceedingConference contribution

1 Citation (Scopus)

Abstract

Combining different machine learning models (decision fusion) has been shown to be an effective method for estimating the underlying physical mechanism by allowing the models to reinforce each other when consensus exists, or, conversely, negate each other when there is no consensus. To be effective, decision fusion requires that the different models provide some degree of complementary information. In this work, we fuse the results of four different machine learning models (Boosted Decision Trees, Neural Networks, Support Vector Machines, Self Organizing Maps) to predict the risk of lung pneumonitis in patients undergoing thoracic radiotherapy. Fusion was achieved by simple averaging of the 10-fold cross validated predictions for each patient from all four models. To reduce prediction dependence on the manner in which the data set was split, 10-fold cross-validation was repeated 100 times for random data splitting. The area under the receiver operating characteristics curve for the fused cross-validated results was 0.79, higher than the individual models and with (generally) lower variance. The fusion extracted three important features as the consensus among all four models in predicting radiation pneumonitis risk: chemotherapy prior to radiotherapy, equivalent Uniform Dose (EUD) for exponent a = 1.2 to 3, and female gender. The results show great promise for machine learning in radiotherapy outcomes modeling.

Original languageEnglish (US)
Title of host publicationProceedings - 7th International Conference on Machine Learning and Applications, ICMLA 2008
Pages545-550
Number of pages6
DOIs
StatePublished - Dec 1 2008
Event7th International Conference on Machine Learning and Applications, ICMLA 2008 - San Diego, CA, United States
Duration: Dec 11 2008Dec 13 2008

Publication series

NameProceedings - 7th International Conference on Machine Learning and Applications, ICMLA 2008

Conference

Conference7th International Conference on Machine Learning and Applications, ICMLA 2008
CountryUnited States
CitySan Diego, CA
Period12/11/0812/13/08

Fingerprint

Learning systems
Fusion reactions
Radiotherapy
Chemotherapy
Self organizing maps
Electric fuses
Decision trees
Dosimetry
Support vector machines
Neural networks
Radiation

ASJC Scopus subject areas

  • Artificial Intelligence
  • Computer Science Applications
  • Software

Cite this

Das, S. K., Chen, S., Deasy, J. O., Zhou, S., Yin, F. F., & Marks, L. B. (2008). Decision fusion of machine learning models to predict radiotherapyinduced lung pneumonitis. In Proceedings - 7th International Conference on Machine Learning and Applications, ICMLA 2008 (pp. 545-550). [4725027] (Proceedings - 7th International Conference on Machine Learning and Applications, ICMLA 2008). https://doi.org/10.1109/ICMLA.2008.122

Decision fusion of machine learning models to predict radiotherapyinduced lung pneumonitis. / Das, Shiva K.; Chen, Shifeng; Deasy, Joseph O.; Zhou, Sumin; Yin, Fang Fane; Marks, Lawrence B.

Proceedings - 7th International Conference on Machine Learning and Applications, ICMLA 2008. 2008. p. 545-550 4725027 (Proceedings - 7th International Conference on Machine Learning and Applications, ICMLA 2008).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Das, SK, Chen, S, Deasy, JO, Zhou, S, Yin, FF & Marks, LB 2008, Decision fusion of machine learning models to predict radiotherapyinduced lung pneumonitis. in Proceedings - 7th International Conference on Machine Learning and Applications, ICMLA 2008., 4725027, Proceedings - 7th International Conference on Machine Learning and Applications, ICMLA 2008, pp. 545-550, 7th International Conference on Machine Learning and Applications, ICMLA 2008, San Diego, CA, United States, 12/11/08. https://doi.org/10.1109/ICMLA.2008.122
Das SK, Chen S, Deasy JO, Zhou S, Yin FF, Marks LB. Decision fusion of machine learning models to predict radiotherapyinduced lung pneumonitis. In Proceedings - 7th International Conference on Machine Learning and Applications, ICMLA 2008. 2008. p. 545-550. 4725027. (Proceedings - 7th International Conference on Machine Learning and Applications, ICMLA 2008). https://doi.org/10.1109/ICMLA.2008.122
Das, Shiva K. ; Chen, Shifeng ; Deasy, Joseph O. ; Zhou, Sumin ; Yin, Fang Fane ; Marks, Lawrence B. / Decision fusion of machine learning models to predict radiotherapyinduced lung pneumonitis. Proceedings - 7th International Conference on Machine Learning and Applications, ICMLA 2008. 2008. pp. 545-550 (Proceedings - 7th International Conference on Machine Learning and Applications, ICMLA 2008).
@inproceedings{2f2d4562fb104e7393a0d101cb409c1d,
title = "Decision fusion of machine learning models to predict radiotherapyinduced lung pneumonitis",
abstract = "Combining different machine learning models (decision fusion) has been shown to be an effective method for estimating the underlying physical mechanism by allowing the models to reinforce each other when consensus exists, or, conversely, negate each other when there is no consensus. To be effective, decision fusion requires that the different models provide some degree of complementary information. In this work, we fuse the results of four different machine learning models (Boosted Decision Trees, Neural Networks, Support Vector Machines, Self Organizing Maps) to predict the risk of lung pneumonitis in patients undergoing thoracic radiotherapy. Fusion was achieved by simple averaging of the 10-fold cross validated predictions for each patient from all four models. To reduce prediction dependence on the manner in which the data set was split, 10-fold cross-validation was repeated 100 times for random data splitting. The area under the receiver operating characteristics curve for the fused cross-validated results was 0.79, higher than the individual models and with (generally) lower variance. The fusion extracted three important features as the consensus among all four models in predicting radiation pneumonitis risk: chemotherapy prior to radiotherapy, equivalent Uniform Dose (EUD) for exponent a = 1.2 to 3, and female gender. The results show great promise for machine learning in radiotherapy outcomes modeling.",
author = "Das, {Shiva K.} and Shifeng Chen and Deasy, {Joseph O.} and Sumin Zhou and Yin, {Fang Fane} and Marks, {Lawrence B.}",
year = "2008",
month = "12",
day = "1",
doi = "10.1109/ICMLA.2008.122",
language = "English (US)",
isbn = "9780769534954",
series = "Proceedings - 7th International Conference on Machine Learning and Applications, ICMLA 2008",
pages = "545--550",
booktitle = "Proceedings - 7th International Conference on Machine Learning and Applications, ICMLA 2008",

}

TY - GEN

T1 - Decision fusion of machine learning models to predict radiotherapyinduced lung pneumonitis

AU - Das, Shiva K.

AU - Chen, Shifeng

AU - Deasy, Joseph O.

AU - Zhou, Sumin

AU - Yin, Fang Fane

AU - Marks, Lawrence B.

PY - 2008/12/1

Y1 - 2008/12/1

N2 - Combining different machine learning models (decision fusion) has been shown to be an effective method for estimating the underlying physical mechanism by allowing the models to reinforce each other when consensus exists, or, conversely, negate each other when there is no consensus. To be effective, decision fusion requires that the different models provide some degree of complementary information. In this work, we fuse the results of four different machine learning models (Boosted Decision Trees, Neural Networks, Support Vector Machines, Self Organizing Maps) to predict the risk of lung pneumonitis in patients undergoing thoracic radiotherapy. Fusion was achieved by simple averaging of the 10-fold cross validated predictions for each patient from all four models. To reduce prediction dependence on the manner in which the data set was split, 10-fold cross-validation was repeated 100 times for random data splitting. The area under the receiver operating characteristics curve for the fused cross-validated results was 0.79, higher than the individual models and with (generally) lower variance. The fusion extracted three important features as the consensus among all four models in predicting radiation pneumonitis risk: chemotherapy prior to radiotherapy, equivalent Uniform Dose (EUD) for exponent a = 1.2 to 3, and female gender. The results show great promise for machine learning in radiotherapy outcomes modeling.

AB - Combining different machine learning models (decision fusion) has been shown to be an effective method for estimating the underlying physical mechanism by allowing the models to reinforce each other when consensus exists, or, conversely, negate each other when there is no consensus. To be effective, decision fusion requires that the different models provide some degree of complementary information. In this work, we fuse the results of four different machine learning models (Boosted Decision Trees, Neural Networks, Support Vector Machines, Self Organizing Maps) to predict the risk of lung pneumonitis in patients undergoing thoracic radiotherapy. Fusion was achieved by simple averaging of the 10-fold cross validated predictions for each patient from all four models. To reduce prediction dependence on the manner in which the data set was split, 10-fold cross-validation was repeated 100 times for random data splitting. The area under the receiver operating characteristics curve for the fused cross-validated results was 0.79, higher than the individual models and with (generally) lower variance. The fusion extracted three important features as the consensus among all four models in predicting radiation pneumonitis risk: chemotherapy prior to radiotherapy, equivalent Uniform Dose (EUD) for exponent a = 1.2 to 3, and female gender. The results show great promise for machine learning in radiotherapy outcomes modeling.

UR - http://www.scopus.com/inward/record.url?scp=60649104824&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=60649104824&partnerID=8YFLogxK

U2 - 10.1109/ICMLA.2008.122

DO - 10.1109/ICMLA.2008.122

M3 - Conference contribution

SN - 9780769534954

T3 - Proceedings - 7th International Conference on Machine Learning and Applications, ICMLA 2008

SP - 545

EP - 550

BT - Proceedings - 7th International Conference on Machine Learning and Applications, ICMLA 2008

ER -