Optimization of treatment strategy by using a machine learning model to predict survival time of patients with malignant glioma after radiotherapy

ABSTRACT The purpose of this study was to predict the survival time of patients with malignant glioma after radiotherapy with high accuracy by considering additional clinical factors and optimize the prescription dose and treatment duration for individual patient by using a machine learning model. A total of 35 patients with malignant glioma were included in this study. The candidate features included 12 clinical features and 192 dose–volume histogram (DVH) features. The appropriate input features and parameters of the support vector machine (SVM) were selected using the genetic algorithm based on Akaike’s information criterion, i.e. clinical, DVH, and both clinical and DVH features. The prediction accuracy of the SVM models was evaluated through a leave-one-out cross-validation test with residual error, which was defined as the absolute difference between the actual and predicted survival times after radiotherapy. Moreover, the influences of various values of prescription dose and treatment duration on the predicted survival time were evaluated. The prediction accuracy was significantly improved with the combined use of clinical and DVH features compared with the separate use of both features (P < 0.01, Wilcoxon signed rank test). Mean ± standard deviation of the leave-one-out cross-validation using the combined clinical and DVH features, only clinical features and only DVH features were 104.7 ± 96.5, 144.2 ± 126.1 and 204.5 ± 186.0 days, respectively. The prediction accuracy could be improved with the combination of clinical and DVH features, and our results show the potential to optimize the treatment strategy for individual patients based on a machine learning model.

[1]  Luke Macyszyn,et al.  Imaging patterns predict patient survival and molecular subtype in glioblastoma via machine learning techniques. , 2016, Neuro-oncology.

[2]  Christos Davatzikos,et al.  Classification of brain tumor type and grade using MRI texture and shape in a machine learning scheme , 2009, Magnetic resonance in medicine.

[3]  W D Plummer,et al.  Power and sample size calculations for studies involving linear regression. , 1998, Controlled clinical trials.

[4]  Timothy Solberg,et al.  Machine learning algorithms for outcome prediction in (chemo)radiotherapy: An empirical comparison of classifiers , 2018, Medical physics.

[5]  Hsin-Hsiung Huang,et al.  Comparing logistic regression, support vector machines, and permanental classification methods in predicting hypertension , 2014, BMC Proceedings.

[6]  Carsten Brink,et al.  Survival prediction of non-small cell lung cancer patients using radiomics analyses of cone-beam CT images. , 2017, Radiotherapy and oncology : journal of the European Society for Therapeutic Radiology and Oncology.

[7]  Alan Effraim Nahum,et al.  (Radio)Biological Optimization of External-Beam Radiotherapy , 2012, Comput. Math. Methods Medicine.

[8]  Issam El Naqa,et al.  Outcome modeling techniques for prostate cancer radiotherapy: Data, models, and validation. , 2016, Physica medica : PM : an international journal devoted to the applications of physics to medicine and biology : official journal of the Italian Association of Biomedical Physics.

[9]  Umberto Castellani,et al.  Classification of first-episode psychosis in a large cohort of patients using support vector machine and multiple kernel learning techniques , 2017, NeuroImage.

[10]  Dahai Li,et al.  An ankle rehabilitation robot based on 3-RRS spherical parallel mechanism , 2017 .

[11]  Chih-Jen Lin,et al.  A Practical Guide to Support Vector Classication , 2008 .

[12]  Hyunjin Park,et al.  Classification of the glioma grading using radiomics analysis , 2018, PeerJ.

[13]  De-Chen Lin,et al.  Non-malignant epithelial cells preferentially proliferate from nasopharyngeal carcinoma biopsy cultured under conditionally reprogrammed conditions , 2017, Scientific Reports.

[14]  T. Beyer,et al.  Glioma Survival Prediction with Combined Analysis of In Vivo 11C-MET PET Features, Ex Vivo Features, and Patient Features by Supervised Machine Learning , 2017, The Journal of Nuclear Medicine.

[15]  Issam El-Naqa,et al.  Radiomics strategies for risk assessment of tumour failure in head-and-neck cancer , 2017, Scientific Reports.

[16]  Bernhard Schölkopf,et al.  A tutorial on support vector regression , 2004, Stat. Comput..

[17]  Bongile Mzenda,et al.  A radiobiological optimization approach in VMAT prostate planning using RayStation , 2014 .

[18]  Raymond Y Huang,et al.  Multimodal MRI features predict isocitrate dehydrogenase genotype in high-grade gliomas , 2017, Neuro-oncology.

[19]  Cheng-Lung Huang,et al.  A GA-based feature selection and parameters optimizationfor support vector machines , 2006, Expert Syst. Appl..

[20]  D. Townsend,et al.  Impact of Image Reconstruction Settings on Texture Features in 18F-FDG PET , 2015, The Journal of Nuclear Medicine.

[21]  Lawrence H. Schwartz,et al.  Quantitative imaging biomarkers for risk stratification of patients with recurrent glioblastoma treated with bevacizumab , 2017, Neuro-oncology.

[22]  David R. Anderson,et al.  Understanding AIC and BIC in Model Selection , 2004 .

[23]  G. Weidlich,et al.  Artificial Intelligence in Medicine and Radiation Oncology , 2018, Cureus.

[24]  Guangtao Zhai,et al.  A Deep Learning-Based Radiomics Model for Prediction of Survival in Glioblastoma Multiforme , 2017, Scientific Reports.

[25]  Xiaomin Luo,et al.  A Genetic Algorithm Based Support Vector Machine Model for Blood-Brain Barrier Penetration Prediction , 2015, BioMed research international.

[26]  I Syndikus,et al.  Prostate Dose-painting Radiotherapy and Radiobiological Guided Optimisation Enhances the Therapeutic Ratio. , 2016, Clinical oncology (Royal College of Radiologists (Great Britain)).

[27]  S Webb,et al.  A model for calculating tumour control probability in radiotherapy including the effects of inhomogeneous distributions of dose and clonogenic cell density. , 1993, Physics in medicine and biology.

[28]  Ron Kohavi,et al.  A Study of Cross-Validation and Bootstrap for Accuracy Estimation and Model Selection , 1995, IJCAI.

[29]  Samuel H. Hawkins,et al.  Reproducibility and Prognosis of Quantitative Features Extracted from CT Images. , 2014, Translational oncology.

[30]  P. Lambin,et al.  Decoding tumour phenotype by noninvasive imaging using a quantitative radiomics approach , 2014, Nature Communications.

[31]  Changqing Shen,et al.  A hybrid technique based on convolutional neural network and support vector regression for intelligent diagnosis of rotating machinery , 2017 .

[32]  Daniel L. Rubin,et al.  Deep Learning for Brain MRI Segmentation: State of the Art and Future Directions , 2017, Journal of Digital Imaging.

[33]  Ahmad Chaddad,et al.  Prediction of survival with multi-scale radiomic analysis in glioblastoma patients , 2018, Medical & Biological Engineering & Computing.

[34]  Joseph O Deasy,et al.  Predicting radiotherapy outcomes using statistical learning techniques , 2009, Physics in medicine and biology.

[35]  D. Nelson,et al.  Recursive partitioning analysis of prognostic factors in three Radiation Therapy Oncology Group malignant glioma trials. , 1993, Journal of the National Cancer Institute.

[36]  Gustavo Carneiro,et al.  Automated 5-year mortality prediction using deep learning and radiomics features from chest computed tomography , 2016, 2017 IEEE 14th International Symposium on Biomedical Imaging (ISBI 2017).

[37]  Alex Rubinsteyn,et al.  Using a Machine Learning Approach to Predict Outcomes after Radiosurgery for Cerebral Arteriovenous Malformations , 2016, Scientific Reports.

[38]  David R. Anderson,et al.  Multimodel Inference , 2004 .

[39]  Susan M. Chang,et al.  Temozolomide in the treatment of recurrent malignant glioma , 2004, Cancer.