Automated Machine Learning Based on Radiomics Features Predicts H3 K27M Mutation in Midline Gliomas of the Brain.

BACKGROUND Conventional MRI cannot be used to identify H3 K27M mutation status. This study aimed to investigate the feasibility of predicting H3 K27M mutation status by applying an automated machine learning (autoML) approach to the MR radiomics features of patients with midline gliomas. METHODS This single-institution retrospective study included 100 patients with midline gliomas, including 40 patients with H3 K27M mutations and 60 wild-type patients. Radiomics features were extracted from FLAIR images. Prior to autoML analysis, the dataset was randomly stratified into separate 75% training and 25% testing cohorts. The Tree-based Optimization Tool (TPOT) was applied to optimize the machine learning pipeline and select important radiomics features. We compared the performance of ten independent TPOT-generated models based on training and testing cohorts using the area under the curve (AUC) and average precision to obtain the final model. An independent cohort of 22 patients was used to validate the best model. RESULTS Ten prediction models were generated by TPOT, and the accuracy obtained with the best pipeline ranged from 0.788 to 0.867 for the training cohort and from 0.60 to 0.84 for the testing cohort. After comparison, the AUC value and average precision of the final model were 0.903 and 0.911 in the testing cohort, respectively. In the validation set, the AUC was 0.85, and the average precision was 0.855 for the best model. CONCLUSIONS The autoML classifier using radiomics features of conventional MR images provides high discriminatory accuracy in predicting the H3 K27M mutation status of midline glioma.

[1]  Michael Platten,et al.  K27M-mutant histone-3 as a novel target for glioma immunotherapy , 2017, Oncoimmunology.

[2]  Guido Gerig,et al.  User-guided 3D active contour segmentation of anatomical structures: Significantly improved efficiency and reliability , 2006, NeuroImage.

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

[4]  Marc Parizeau,et al.  DEAP: evolutionary algorithms made easy , 2012, J. Mach. Learn. Res..

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

[6]  Arie Perry,et al.  Diffuse Midline Gliomas with Histone H3‐K27M Mutation: A Series of 47 Cases Assessing the Spectrum of Morphologic Variation and Associated Genetic Alterations , 2016, Brain pathology.

[7]  Baojuan Li,et al.  Radiomics Strategy for Molecular Subtype Stratification of Lower‐Grade Glioma: Detecting IDH and TP53 Mutations Based on Multimodal MRI , 2018, Journal of magnetic resonance imaging : JMRI.

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

[9]  Randal S. Olson,et al.  Considerations for automated machine learning in clinical metabolic profiling: Altered homocysteine plasma concentration associated with metformin exposure , 2017, PSB.

[10]  Sung Soo Ahn,et al.  Differentiation between spinal cord diffuse midline glioma with histone H3 K27M mutation and wild type: comparative magnetic resonance imaging , 2019, Neuroradiology.

[11]  B. Kleinschmidt-DeMasters,et al.  H3 K27M-mutant gliomas in adults vs. children share similar histological features and adverse prognosis , 2018, Clinical neuropathology.

[12]  Dominik Sturm,et al.  Diffuse high-grade gliomas with H3 K27M mutations carry a dismal prognosis independent of tumor location , 2018, Neuro-oncology.

[13]  Yuanyuan Wang,et al.  Noninvasive IDH1 mutation estimation based on a quantitative radiomics approach for grade II glioma , 2017, European Radiology.

[14]  Li Ding,et al.  Somatic Histone H3 Alterations in Paediatric Diffuse Intrinsic Pontine Gliomas and Non-Brainstem Glioblastomas , 2012, Nature Genetics.

[15]  M. Aboian,et al.  Imaging Characteristics of Pediatric Diffuse Midline Gliomas with Histone H3 K27M Mutation , 2017, American Journal of Neuroradiology.

[16]  Stephen S F Yip,et al.  Radiomics‐based Assessment of Radiation‐induced Lung Injury After Stereotactic Body Radiotherapy , 2017, Clinical lung cancer.

[17]  Tobias Gauer,et al.  Radiomics of Brain MRI: Utility in Prediction of Metastatic Tumor Type. , 2019, Radiology.

[18]  Geoffrey G. Zhang,et al.  Reproducibility of F18‐FDG PET radiomic features for different cervical tumor segmentation methods, gray‐level discretization, and reconstruction algorithms , 2017, Journal of applied clinical medical physics.

[19]  Pieter Wesseling,et al.  cIMPACT-NOW update 2: diagnostic clarifications for diffuse midline glioma, H3 K27M-mutant and diffuse astrocytoma/anaplastic astrocytoma, IDH-mutant , 2018, Acta Neuropathologica.

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

[21]  Arie Perry,et al.  Genetics of Glioblastomas in Rare Anatomical Locations: Spinal Cord and Optic Nerve , 2016, Brain pathology.

[22]  Alex Vitkin,et al.  Pixel classification method in optical coherence tomography for tumor segmentation and its complementary usage with OCT microangiography , 2018, Journal of biophotonics.

[23]  Kai Wang,et al.  MRI features predict p53 status in lower-grade gliomas via a machine-learning approach , 2017, NeuroImage: Clinical.

[24]  Edward Pan,et al.  Adult Brainstem Gliomas With H3K27M Mutation: Radiology, Pathology, and Prognosis , 2018, Journal of neuropathology and experimental neurology.

[25]  Uri Tabori,et al.  Targeted detection of genetic alterations reveal the prognostic impact of H3K27M and MAPK pathway aberrations in paediatric thalamic glioma , 2016, Acta Neuropathologica Communications.

[26]  Ninon Burgos,et al.  New advances in the Clinica software platform for clinical neuroimaging studies , 2019 .

[27]  Hans Kristian Bø,et al.  Quantitative texture analysis in the prediction of IDH status in low-grade gliomas , 2018, Clinical Neurology and Neurosurgery.

[28]  Dafna Ben-Bashat,et al.  MRI radiomics analysis of molecular alterations in low-grade gliomas , 2018, International Journal of Computer Assisted Radiology and Surgery.

[29]  Patrick Granton,et al.  Radiomics: extracting more information from medical images using advanced feature analysis. , 2012, European journal of cancer.

[30]  Randal S. Olson,et al.  Automating Biomedical Data Science Through Tree-Based Pipeline Optimization , 2016, EvoApplications.

[31]  Randal S. Olson,et al.  TPOT: A Tree-based Pipeline Optimization Tool for Automating Machine Learning , 2016, AutoML@ICML.

[32]  Kuniaki Saito,et al.  H3F3A K27M mutations in thalamic gliomas from young adult patients. , 2014, Neuro-oncology.

[33]  Lars Schmidt-Thieme,et al.  Beyond Manual Tuning of Hyperparameters , 2015, KI - Künstliche Intelligenz.