Machine learning methods for MRI biomarkers analysis of pediatric posterior fossa tumors

Abstract Medical imaging technologies provide an increasing number of opportunities for disease prediction and prognosis. Specifically, imaging biomarkers can quantify the entire tumor phenotypes to enhance the prediction. Machine learning technology can be explored to mine and analyze these biomarkers and to establish predictive models for the clinical applications. Several studies have applied various machine learning methods to imaging biomarkers based clinical predictions of different diseases. Here we seek to evaluate different machine learning methods in pediatric posterior fossa tumor prediction. We present a machine learning based magnetic resonance imaging biomarkers analysis framework for two kinds of pediatric posterior fossa tumors. In details, three feature extraction methods are used to obtain 300 imaging biomarkers. 10 feature selection methods and 11 classifiers are evaluated by the quantified predictive performance and stability, and importance consistency of features and the influence of the experimental factors are also analyzed. Our results demonstrate that the CFS feature selection method (accuracy: 83.85 ± 5.51%, stability: [0.84, 0.06]) and SVM classifier (accuracy: 85.38 ± 3.47%, RSD: 4.77%) show relatively better performance than others and should be preferred. Among all the biomarkers, 17 texture features seem to be more important. Multifactor analysis results indicate the choice of classifier accounts for the most contribution to the variability in performance (37.25%). The machine learning based framework is efficient for pediatric posterior fossa tumors biomarkers analysis, and could provide valuable references and decision support for assisted clinical diagnosis.

[1]  Sung-Bae Cho,et al.  Machine Learning in DNA Microarray Analysis for Cancer Classification , 2003, APBC.

[2]  David H. Kim,et al.  CT texture features of liver parenchyma for predicting development of metastatic disease and overall survival in patients with colorectal cancer , 2018, European Radiology.

[3]  P. Lambin,et al.  Machine Learning methods for Quantitative Radiomic Biomarkers , 2015, Scientific Reports.

[4]  Senén Barro,et al.  Do we need hundreds of classifiers to solve real world classification problems? , 2014, J. Mach. Learn. Res..

[5]  M. Teshnehlab,et al.  Feature Selection and Classification of Breast Cancer on Dynamic Magnetic Resonance Imaging Using ANN and SVM , 2012 .

[6]  Dinggang Shen,et al.  Deep ensemble learning of sparse regression models for brain disease diagnosis , 2017, Medical Image Anal..

[7]  Raymond Y Huang,et al.  Radiographic prediction of meningioma grade by semantic and radiomic features , 2017, PloS one.

[8]  Colin C Pritchard,et al.  Targeted Sequencing of Malignant Supratentorial Pediatric Brain Tumors Demonstrates a High Frequency of Clinically Relevant Mutations , 2018, Pediatric and developmental pathology : the official journal of the Society for Pediatric Pathology and the Paediatric Pathology Society.

[9]  T. Jaspan,et al.  Metrics and Textural Features of MRI Diffusion to Improve Classification of Pediatric Posterior Fossa Tumors , 2014, American Journal of Neuroradiology.

[10]  Mehrbakhsh Nilashi,et al.  Diseases diagnosis using fuzzy logic methods: A systematic and meta-analysis review , 2018, Comput. Methods Programs Biomed..

[11]  Jean-Philippe Vert,et al.  The Influence of Feature Selection Methods on Accuracy, Stability and Interpretability of Molecular Signatures , 2011, PloS one.

[12]  Jinzhong Yang,et al.  Measuring Computed Tomography Scanner Variability of Radiomics Features , 2015, Investigative radiology.

[13]  B. Zheng,et al.  Computer-aided diagnosis of lung cancer: the effect of training data sets on classification accuracy of lung nodules , 2018, Physics in medicine and biology.

[14]  U ManChon,et al.  Prediction and Prioritization of Rare Oncogenic Mutations in the Cancer Kinome Using Novel Features and Multiple Classifiers , 2014, PLoS Comput. Biol..

[15]  Balaji Ganeshan,et al.  Texture analysis of non-small cell lung cancer on unenhanced computed tomography: initial evidence for a relationship with tumour glucose metabolism and stage , 2010, Cancer imaging : the official publication of the International Cancer Imaging Society.

[16]  Theodoros N. Arvanitis,et al.  Three‐dimensional textural features of conventional MRI improve diagnostic classification of childhood brain tumours , 2015, NMR in biomedicine.

[17]  Axel Wismüller,et al.  Classification of micro-CT images using 3D characterization of bone canal patterns in human osteogenesis imperfecta , 2017, Medical Imaging.

[18]  Hussein Hijazi,et al.  A classification framework applied to cancer gene expression profiles. , 2013, Journal of healthcare engineering.

[19]  S. Resnick,et al.  Detection of prodromal Alzheimer's disease via pattern classification of magnetic resonance imaging , 2008, Neurobiology of Aging.

[20]  Theodoros N. Arvanitis,et al.  Texture analysis of T1- and T2-weighted MR images and use of probabilistic neural network to discriminate posterior fossa tumours in children , 2014, NMR in biomedicine.

[21]  Michel Verleysen,et al.  The Curse of Dimensionality in Data Mining and Time Series Prediction , 2005, IWANN.

[22]  S. Plevritis,et al.  Glioblastoma multiforme: exploratory radiogenomic analysis by using quantitative image features. , 2014, Radiology.

[23]  Mehrbakhsh Nilashi,et al.  A predictive method for hepatitis disease diagnosis using ensembles of neuro-fuzzy technique. , 2019, Journal of infection and public health.

[24]  P. Lambin,et al.  Radiomic Machine-Learning Classifiers for Prognostic Biomarkers of Head and Neck Cancer , 2015, Front. Oncol..

[25]  Mehrbakhsh Nilashi,et al.  An analytical method for diseases prediction using machine learning techniques , 2017, Comput. Chem. Eng..

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

[27]  Klaas Nicolay,et al.  1H MR spectroscopy of the brain: absolute quantification of metabolites. , 2006, Radiology.

[28]  Zhiqiang Zhang,et al.  Brain medical image diagnosis based on corners with importance-values , 2017, BMC Bioinformatics.

[29]  Filippo Molinari,et al.  Comparative assessment of texture features for the identification of cancer in ultrasound images: a review , 2018 .

[30]  Sotiris B. Kotsiantis,et al.  Supervised Machine Learning: A Review of Classification Techniques , 2007, Informatica.

[31]  Robert J. Gillies,et al.  Predicting Outcomes of Nonsmall Cell Lung Cancer Using CT Image Features , 2014, IEEE Access.

[32]  T. Merchant,et al.  Quantitative imaging analysis of posterior fossa ependymoma location in children , 2016, Child's Nervous System.

[33]  Jan Luts,et al.  Effect of feature extraction for brain tumor classification based on short echo time 1H MR spectra , 2008, Magnetic resonance in medicine.

[34]  J. Connor,et al.  Biomarker-based predictive models for prognosis in amyotrophic lateral sclerosis. , 2013, JAMA neurology.

[35]  Billie Anderson,et al.  Comparison of the predictive qualities of three prognostic models of colorectal cancer. , 2010, Frontiers in bioscience.

[36]  Avner Meoded,et al.  Neuroimaging of pediatric posterior fossa tumors including review of the literature , 2012, Journal of magnetic resonance imaging : JMRI.