Fully Automatic Calibration of Tumor-Growth Models Using a Single mpMRI Scan

Our objective is the calibration of mathematical tumor growth models from a single multiparametric scan. The target problem is the analysis of preoperative Glioblastoma (GBM) scans. To this end, we present a fully automatic tumor-growth calibration methodology that integrates a single-species reaction-diffusion partial differential equation (PDE) model for tumor progression with multiparametric Magnetic Resonance Imaging (mpMRI) scans to robustly extract patient specific biomarkers i.e., estimates for (i) the tumor cell proliferation rate, (ii) the tumor cell migration rate, and (iii) the original, localized site(s) of tumor initiation. Our method is based on a sparse reconstruction algorithm for the tumor initial location (TIL). This problem is particularly challenging due to nonlinearity, ill-posedeness, and ill conditioning. We propose a coarse-to-fine multi-resolution continuation scheme with parameter decomposition to stabilize the inversion. We demonstrate robustness and practicality of our method by applying the proposed method to clinical data of 206 GBM patients. We analyze the extracted biomarkers and relate tumor origin with patient overall survival by mapping the former into a common atlas space. We present preliminary results that suggest improved accuracy for prediction of patient overall survival when a set of imaging features is augmented with estimated biophysical parameters. All extracted features, tumor initial positions, and biophysical growth parameters are made publicly available for further analysis. To our knowledge, this is the first fully automatic scheme that can handle multifocal tumors and can localize the TIL to a few millimeters.

[1]  Tianqi Chen,et al.  XGBoost: A Scalable Tree Boosting System , 2016, KDD.

[2]  Alfredo Quinones-Hinojosa,et al.  Relationship of glioblastoma multiforme to the lateral ventricles predicts survival following tumor resection , 2008, Journal of Neuro-Oncology.

[3]  Martin Bendszus,et al.  Location-Dependent Patient Outcome and Recurrence Patterns in IDH1-Wildtype Glioblastoma , 2019, Cancers.

[4]  Christos Davatzikos,et al.  Coupling brain-tumor biophysical models and diffeomorphic image registration. , 2017, Computer methods in applied mechanics and engineering.

[5]  Joo-Ho Lee,et al.  Abstract 2455: Human glioblastoma arises from the distant subventricular zone normal appearing but harboring tumor-initiating mutations , 2017 .

[6]  Johan Pallud,et al.  MRI Atlas of IDH Wild-Type Supratentorial Glioblastoma: Probabilistic Maps of Phenotype, Management, and Outcomes. , 2019, Radiology.

[7]  Thomas E Yankeelov,et al.  Clinically Relevant Modeling of Tumor Growth and Treatment Response , 2013, Science Translational Medicine.

[8]  Panagiotis Angelikopoulos,et al.  Personalized Radiotherapy Design for Glioblastoma: Integrating Mathematical Tumor Models, Multimodal Scans, and Bayesian Inference , 2018, IEEE Transactions on Medical Imaging.

[9]  Kurt Keutzer,et al.  A Novel Domain Adaptation Framework for Medical Image Segmentation , 2018, BrainLes@MICCAI.

[10]  Thomas Welzel,et al.  A comparison of long-term survivors and short-term survivors with glioblastoma, subventricular zone involvement: a predictive factor for survival? , 2014, Radiation oncology.

[11]  Christos Davatzikos,et al.  Epidermal Growth Factor Receptor Extracellular Domain Mutations in Glioblastoma Present Opportunities for Clinical Imaging and Therapeutic Development. , 2018, Cancer cell.

[12]  G. Biros,et al.  Simulation of glioblastoma growth using a 3D multispecies tumor model with mass effect , 2018, Journal of Mathematical Biology.

[13]  Christos Davatzikos,et al.  CLAIRE: A distributed-memory solver for constrained large deformation diffeomorphic image registration , 2018, SIAM J. Sci. Comput..

[14]  George Biros,et al.  Where did the tumor start? An inverse solver with sparse localization for tumor growth models , 2019, Inverse problems.

[15]  Hervé Delingette,et al.  Realistic simulation of the 3-D growth of brain tumors in MR images coupling diffusion with biomechanical deformation , 2005, IEEE Transactions on Medical Imaging.

[16]  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.

[17]  Hervé Delingette,et al.  Image Guided Personalization of Reaction-Diffusion Type Tumor Growth Models Using Modified Anisotropic Eikonal Equations , 2010, IEEE Transactions on Medical Imaging.

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

[19]  Jeong Ho Lee,et al.  The origin-of-cell harboring cancer-driving mutations in human glioblastoma , 2018, BMB reports.

[20]  D. Ruppert.  The Elements of Statistical Learning: Data Mining, Inference, and Prediction , 2004 .

[21]  Hang-gen Du,et al.  Correlation Between Tumor Location and Clinical Properties of Glioblastomas in Frontal and Temporal Lobes. , 2018, World neurosurgery.

[22]  Antti Honkela,et al.  A Generative Approach for Image-Based Modeling of Tumor Growth , 2011, IPMI.

[23]  Christos Davatzikos,et al.  Population-based MRI atlases of spatial distribution are specific to patient and tumor characteristics in glioblastoma , 2016, NeuroImage: Clinical.

[24]  Deanna Needell,et al.  CoSaMP: Iterative signal recovery from incomplete and inaccurate samples , 2008, ArXiv.

[25]  Kristin R. Swanson,et al.  Patient-Specific Mathematical Neuro-Oncology: Using a Simple Proliferation and Invasion Tumor Model to Inform Clinical Practice , 2015, Bulletin of mathematical biology.

[26]  G. Biros,et al.  An inverse problem formulation for parameter estimation of a reaction–diffusion model of low grade gliomas , 2014, Journal of mathematical biology.

[27]  Marie Blonski,et al.  A Probabilistic Atlas of Diffuse WHO Grade II Glioma Locations in the Brain , 2016, PloS one.

[28]  K. Swanson,et al.  A mathematical modelling tool for predicting survival of individual patients following resection of glioblastoma: a proof of principle , 2007, British Journal of Cancer.

[29]  Christos Davatzikos,et al.  An image-driven parameter estimation problem for a reaction–diffusion glioma growth model with mass effects , 2008, Journal of mathematical biology.

[30]  K. Swanson,et al.  Modeling Tumor-Associated Edema in Gliomas during Anti-Angiogenic Therapy and Its Impact on Imageable Tumor , 2013, Front. Oncol..

[31]  Michael Hinczewski,et al.  The 2019 mathematical oncology roadmap , 2019, Physical biology.

[32]  Zhongli Jiang,et al.  Co-Deletion of Chromosome 1p/19q and IDH1/2 Mutation in Glioma Subsets of Brain Tumors in Chinese Patients , 2012, PloS one.

[33]  Christos Davatzikos,et al.  Radiomic signature of infiltration in peritumoral edema predicts subsequent recurrence in glioblastoma: implications for personalized radiotherapy planning , 2018, Journal of medical imaging.

[34]  Hervé Delingette,et al.  Tumor growth parameters estimation and source localization from a unique time point: Application to low-grade gliomas , 2013, Comput. Vis. Image Underst..

[35]  Christos Davatzikos,et al.  GLISTR: Glioma Image Segmentation and Registration , 2012, IEEE Transactions on Medical Imaging.