Application of Machine Learning Techniques for Prediction of Radiation Pneumonitis in Lung Cancer Patients

Lung cancer patients who receive radiotherapy as part of their treatment are at risk radiation-induced lung injury known as radiation pneumonitis (RP). RP is a potentially fatal side effect to treatment. Hence, new methods are needed to guide physicians to prescribe targeted therapy dosage to patients at high risk of RP. Several predictive models based on traditional statistical methods and machine learning techniques have been reported, however, no guidance to variation in performance has not been provided to date. Therefore, in this study, we compare several widely used classification algorithms in the machine learning field are used to distinguish between different risk groups of RP. The performance of these classification algorithms is evaluated in conjunction with several feature selection strategy and the impact of the feature selection on performance is further evaluated.

[1]  Ian H. Witten,et al.  Data mining: practical machine learning tools and techniques, 3rd Edition , 1999 .

[2]  Jung Hun Oh,et al.  Proteomic Biomarker Identification for Diagnosis of Early Relapse in Ovarian Cancer , 2006, J. Bioinform. Comput. Biol..

[3]  Vladimir N. Vapnik,et al.  The Nature of Statistical Learning Theory , 2000, Statistics for Engineering and Information Science.

[4]  J. Deasy,et al.  Modeling radiation pneumonitis risk with clinical, dosimetric, and spatial parameters. , 2006, International journal of radiation oncology, biology, physics.

[5]  Chong-Ho Choi,et al.  Input feature selection for classification problems , 2002, IEEE Trans. Neural Networks.

[6]  J. Deasy,et al.  Bioinformatics Methods for Learning Radiation-Induced Lung Inflammation from Heterogeneous Retrospective and Prospective Data , 2009, Journal of biomedicine & biotechnology.

[7]  Nikolas P. Galatsanos,et al.  A support vector machine approach for detection of microcalcifications , 2002, IEEE Transactions on Medical Imaging.

[8]  Huan Liu,et al.  Consistency-based search in feature selection , 2003, Artif. Intell..

[9]  J. Deasy,et al.  Multivariable modeling of radiotherapy outcomes, including dose-volume and clinical factors. , 2006, International journal of radiation oncology, biology, physics.

[10]  Huan Liu,et al.  Efficient Feature Selection via Analysis of Relevance and Redundancy , 2004, J. Mach. Learn. Res..

[11]  A. Niemierko,et al.  Methodological issues in radiation dose-volume outcome analyses: summary of a joint AAPM/NIH workshop. , 2002, Medical physics.

[12]  Laurence Anthony,et al.  Relevant, irredundant feature selection and noisy example elimination , 2004, IEEE Transactions on Systems, Man, and Cybernetics, Part B (Cybernetics).

[13]  Richard Nock,et al.  A hybrid filter/wrapper approach of feature selection using information theory , 2002, Pattern Recognit..

[14]  Joseph O. Deasy,et al.  Nonlinear Kernel-Based Approaches for Predicting Normal Tissue Toxicities , 2008, 2008 Seventh International Conference on Machine Learning and Applications.

[15]  Jung Hun Oh,et al.  Biomarker selection and sample prediction for multi-category disease on MALDI-TOF data , 2008, Bioinform..

[16]  I El Naqa,et al.  Dose response explorer: an integrated open-source tool for exploring and modelling radiotherapy dose–volume outcome relationships , 2006, Physics in medicine and biology.

[17]  Jin-Tsong Jeng,et al.  Hybrid approach of selecting hyperparameters of support vector machine for regression , 2005, IEEE Trans. Syst. Man Cybern. Part B.

[18]  Jason Weston,et al.  Gene Selection for Cancer Classification using Support Vector Machines , 2002, Machine Learning.

[19]  Roberto Battiti,et al.  Using mutual information for selecting features in supervised neural net learning , 1994, IEEE Trans. Neural Networks.

[20]  Bruce A. Draper,et al.  Feature selection from huge feature sets , 2001, Proceedings Eighth IEEE International Conference on Computer Vision. ICCV 2001.