Deep Learning versus Classical Regression for Brain Tumor Patient Survival Prediction

Deep learning for regression tasks on medical imaging data has shown promising results. However, compared to other approaches, their power is strongly linked to the dataset size. In this study, we evaluate 3D-convolutional neural networks (CNNs) and classical regression methods with hand-crafted features for survival time regression of patients with high-grade brain tumors. The tested CNNs for regression showed promising but unstable results. The best performing deep learning approach reached an accuracy of \(51.5\%\) on held-out samples of the training set. All tested deep learning experiments were outperformed by a Support Vector Classifier (SVC) using 30 radiomic features. The investigated features included intensity, shape, location and deep features.

[1]  D. Cox.  The Regression Analysis of Binary Sequences , 1958 .

[2]  R. Tibshirani.  Regression Shrinkage and Selection via the Lasso , 1996 .

[3]  Trevor Hastie,et al.  The Elements of Statistical Learning , 2001 .

[4]  Robert Tibshirani,et al.  The Elements of Statistical Learning: Data Mining, Inference, and Prediction , 2001, Springer Series in Statistics.

[5]  B. Scheithauer,et al.  The 2007 WHO classification of tumours of the central nervous system , 2007, Acta Neuropathologica.

[6]  Christoph H. Lampert.  Kernel Methods in Computer Vision , 2009, Found. Trends Comput. Graph. Vis..

[7]  Wei-Yin Loh,et al.  Classification and regression trees , 2011, WIREs Data Mining Knowl. Discov..

[8]  Mitchel S Berger,et al.  An extent of resection threshold for newly diagnosed glioblastomas. , 2011, Journal of neurosurgery.

[9]  Bruce Fischl,et al.  FreeSurfer , 2012, NeuroImage.

[10]  Brian B. Avants,et al.  The Multimodal Brain Tumor Image Segmentation Benchmark (BRATS) , 2015, IEEE Transactions on Medical Imaging.

[11]  Jimmy Ba,et al.  Adam: A Method for Stochastic Optimization , 2014, ICLR.

[12]  G. Reifenberger,et al.  The 2016 World Health Organization Classification of Tumors of the Central Nervous System: a summary , 2016, Acta Neuropathologica.

[13]  Anders M. Dale,et al.  Differential localization of glioblastoma subtype: implications on glioblastoma pathogenesis , 2016, Oncotarget.

[14]  Paul Kinahan,et al.  Radiomics: Images Are More than Pictures, They Are Data , 2015, Radiology.

[15]  Estanislao Arana,et al.  Glioblastoma: does the pre-treatment geometry matter? A postcontrast T1 MRI-based study , 2017, European Radiology.

[16]  Víctor M. Pérez-García,et al.  Towards Uncertainty-Assisted Brain Tumor Segmentation and Survival Prediction , 2017, BrainLes@MICCAI.

[17]  Sébastien Ourselin,et al.  Automatic Brain Tumor Segmentation Using Cascaded Anisotropic Convolutional Neural Networks , 2017, BrainLes@MICCAI.

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

[19]  Christos Davatzikos,et al.  Advancing The Cancer Genome Atlas glioma MRI collections with expert segmentation labels and radiomic features , 2017, Scientific Data.

[20]  Mauricio Reyes,et al.  Automatic estimation of extent of resection and residual tumor volume of patients with glioblastoma. , 2017, Journal of neurosurgery.

[21]  Andriy Fedorov,et al.  Computational Radiomics System to Decode the Radiographic Phenotype. , 2017, Cancer research.

[22]  R. Spetzler,et al.  Impact of removed tumor volume and location on patient outcome in glioblastoma , 2017, Journal of Neuro-Oncology.

[23]  LinLin Shen,et al.  Deep Learning Based Multimodal Brain Tumor Diagnosis , 2017, BrainLes@MICCAI.

[24]  Victor Alves,et al.  Enhancing interpretability of automatically extracted machine learning features: application to a RBM‐Random Forest system on brain lesion segmentation , 2018, Medical Image Anal..

[25]  et al.,et al.  Identifying the Best Machine Learning Algorithms for Brain Tumor Segmentation, Progression Assessment, and Overall Survival Prediction in the BRATS Challenge , 2018, ArXiv.

[26]  Zev A. Binder,et al.  Radiomic MRI signature reveals three distinct subtypes of glioblastoma with different clinical and molecular characteristics, offering prognostic value beyond IDH1 , 2018, Scientific Reports.