MRI based genomic analysis of glioma using three pathway deep convolutional neural network for IDH classification

As per 2016 updates by World Health Organization (WHO) on cancer disease, gliomas are categorized and further treated based on genomic mutations. The imaging modalities support a complimentary but immediate noninvasive diagnosis of cancer based on genetic mutations. Our aim is to train a deep convolutional neural network for isocitrate dehydrogenase (IDH) genotyping of glioma by auto-extracting the most discriminative features from magnetic resonance imaging (MRI) volumes. MR imaging data of total 217 patients were obtained from The Cancer Imaging Archives (TCIA) of high and low-grade gliomas. A 3-pathway convolutional neural network was trained for IDH classification. The multipath neural network, consisting of one shallow and two deep neural network paths, is used to auto-extract the significant imaging features for successful IDH discrimination into IDH mutant and wild type. An accuracy of 93.67% and cross-entropy loss of 0.052 is achieved for IDH classification. The results of 3-pathway convolutional neural network (CNN) are better than the results achieved from individual paths of 3-pathway model. The results have demonstrated the multipath convolutional neural networks as state-of-the-art method with simple design to predict IDH genotypes in glioma with auto-extraction of radiogenomic features.

PDF

___

[1] Cancer Genome Atlas Research Network. Comprehensive genomic characterization defines human glioblastoma genes and core pathways. Nature 2008; 455 (7216): 1061-1068.
[2] Nalawade S, Murugesan G, Vejdani-Jahromi M, Fisicaro R, Yogananda C et al. Classification of brain tumor isocitrate dehydrogenase status using MRI and deep learning. Journal of Medical Imaging 2019; 6 (4): 1-13.
[3] Yogananda CGB, Shah B, Vejdani-Jahromi M, Nalawade S, Murugesan G et al. A novel fully automated MRIbased deep-learning method for classification of IDH mutation status in brain gliomas. Neuro-oncology 2020; 22 (3): 402-411.
[4] Zhang X, Tian Q, Wang L, Liu Y, Li B et al. Radiomics strategy for the molecular subtype stratification of lowergrade glioma: detecting IDH and TP53 mutations based on multimodal MRI. Journal of Magnetic Resonance Imaging 2018; 48 (4): 916-926.
[5] Meraj T, Hassan A, Zahoor S, Rauf HT, Lali MI et al. Lungs nodule detection using semantic segmentation and classification with optimal features. Preprints 2019; 2019090139. doi: 10.20944/preprints 201909.0139.v1
[6] Korfiatis P, Kline TL, Coufalova L, Lachance DH, Parney IF et al. MRI texture features as biomarkers to predict MGMT methylation status in glioblastomas. Medical Physics 2016; 43 (6): 2835–2844.
[7] Villanueva-Meyer JE, Wood MD, Choi BS, Mabray MC, Butowski NA et al. MRI features and IDH mutational status of grade II diffuse gliomas: impact on diagnosis and prognosis. AJR Americal Journal of Roentgenology 2018; 210 (3): 621-628.
[8] Mossa A, Çevik U. Ensemble learning of multiview CNN models for survival time prediction of braintumor patients using multimodal MRI scans. Turkish Journal of Electrical Engineering & Computer Science 2021; 29: 616-631. doi: 10.3906/elk-2002-175
[9] Zadeh Shirazi A, Fornaciari E, McDonnell MD, Yaghoobi M, Cevallos Y et al. The application of deep convolutional neural networks to brain cancer images: a survey. Journal of Personalized Medicine 2020; 10 (4): 224. doi: 10.3390/jpm10040224
[10] Stefan B, Roland W. A Survey of MRI-based medical image analysis for brain tumor studies. Physics in Medicine and Biology 2013; 58 (13): R97-129.
[11] Bastanlar Y, Ozuysal M. Introduction to machine learning. Methods in Molecular Biology 2014; 1107: 105-128.
[12] Inza I, Calvo B, Armananzas R, Bengoetxea E, Larranaga P et al. Machine learning: an indispensable tool in bioinformatics. Methods in Molecular Biology 2010; 593: 25-48.
[13] Zhang B, Chang K, Ramkissoon S, Tanguturi S, Bi WL et al. Multimodal MRI features predict isocitrate dehydrogenase genotype in high-grade gliomas. Neuro-oncology 2016; 19 (1): 109-117.
[14] Dalu Y, Ganesh R. Evaluation of tumor-derived MRI-texture features for discrimination of molecular subtypes and prediction of 12-month survival status in glioblastoma. Medical Physics 2015; 42 (11): 6725-6735.
[15] Songtao Q, Lei Y. Isocitrate dehydrogenase mutation is associated with tumor location and magnetic resonance imaging characteristics in astrocytic neoplasms. Oncology Letters 2014; 7: 1895-1902.
[16] Bini S. Artificial intelligence, machine learning, deep learning, and cognitive computing: what do these terms mean and how will they impact health care? The Jornal of Arthoplasty 2018; 33 (8): 2358-2361.
[17] Ueda D, Shimazaki A, Miki Y. Technical and clinical overview of deep learning in radiology. Japnese Journal of Radiology 2019; 37: 15-33.
[18] Xue Y, Chen S, Qin J, Liu Y, Huang B et al. Application of deep learning in automated analysis of molecular images in cancer: a survey. Contrast Media Molecular Imaging 2017; 9512370.
[19] Chang P, Grinband XJ, Weinberg XBD. Deep-learning convolutional neural networks accurately classify genetic mutations in gliomas. AJNR American Journal of Neuroradiology 2018; 39: 1201-1207.
[20] Yang Y, Yan LF, Zhang X. Glioma grading on conventional MR images: a deep learning study with transfer learning. Frontiers in Neuroscience 2018; 12 (804): 1-10.
[21] Chang K, Bai HX, Zhou H, Chang S. Residual convolutional neural network for the determination of IDH status in low- and high-grade gliomas from MR imaging. Clinical Cancer Research 2017; OF1-OF9.
[22] Akkus Z, Ali I, Sedlar J. Predicting 1p19q chromosomal deletion of low-grade gliomas from MR images using deep learning. Journal of Digit Imaging 2017; 30 (4): 469-476.
[23] Li Z, Wang Y, Yu J, Guo Y, Cao W. Deep learning based radiomics (DLR) and its usage in non-invasive IDH1 prediction for low grade glioma. Scientific Reports 2017; 7: 5467.
[24] Liang S, Zhang R, Liang D et al. Multimodal 3D DenseNet for IDH genotype prediction in gliomas. Genes 2018; 9 (8): 382.
[25] He K, Xiangyu Z, Shaoqing R, Sun J. Deep residual learning for image recognition. In: Proceedings of the IEEE conference on computer vision and pattern recognition; Las Vegas, US; 2016. pp. 770-778.
[26] Ahmad A, Sarkar S, Shah A, Gore S, Santosh V et al. Predictive and discriminative localization of IDH genotype in high grade gliomas using deep convolutional neural nets. In: IEEE 16th International Symposium on Biomedical Imaging (ISBI); Venice, Italy; 2019. pp. 8-11.
[27] Chougule T, Shinde S, Santosh V, Saini J, Ingalhalikar M. On Validating multimodal MRI based stratification of IDH genotype in high grade gliomas using CNNs and its comparison to radiomics. In: International Workshop on Radiomics and Radiogenomics in Neuro-oncology; Shenzhen, China; 2019. pp. 53-60.
[28] Nalawade S, Yu FF, Bangalore YCG, Murugesan GK, Shah BR et al. Brain Tumor IDH, 1p/19q, and MGMT molecular classification using MRI-based deep learning: effect of motion and motion correction. bioRxiv 2020. doi: 10.1101/2020.06.01.126375
[29] Pasquini L, Napolitano A, Tagliente E, Dellepiane F, Lucignani M et al. Deep learning can differentiate IDH-mutant from IDH-wild type GBM. Journal of Personalized Medicine 2021; 11 (4): 290. doi: 10.3390/jpm11040290
[30] Fukuma R, Yanagisawa T, Kinoshita M et al. Prediction of IDH and TERT promoter mutations in low-grade glioma from magnetic resonance images using a convolutional neural network. Scientific Reports 2019; 9: 20311. doi:10.1038/s41598-019-56767-3
[31] Decuyper M, Bonte S, Deblaere K, Van HR. Automated MRI based pipeline for segmentation and prediction of grade, IDH mutation and 1p19q co-deletion in glioma. Computerized Medical Imaging and Graphics 2021; 88: 101831. doi: 10.1016/j.compmedimag.2020.101831
[32] Gore S, Chougule T, Jagtap J, Saini J, Ingalhalikar M. A review of radiomics and deep predictive modeling in glioma characterization. Academic Radiology 2020. doi: 10.1016/j.acra.2020.06.016
[33] Kessler J, Hohmann T, Güttler A, Petrenko M, Ostheimer C et al. Radiosensitization and a less aggressive phenotype of human malignant glioma cells expressing isocitrate dehydrogenase 1 (IDH1) mutant protein: dissecting the mechanisms. Cancers (Basel) 2019; 11(6): 889. doi: 10.3390/cancers11060889
[34] Houillier C, Wang X, Kaloshi G, Mokhtari K, Guillevin R et al. IDH1 or IDH2 mutations predict longer survival and response to temozolomide in low-grade gliomas. Journal of Neurology 2010; 75: 1560-1566.
[35] Han W, Qin L, Bay C, Chen X, Yu KH et al. Deep transfer learning and radiomics feature prediction of survival of patients with high-grade gliomas. AJNR American Journal of Neuroradiology 2020; 41 (1): 40-48.
[36] Calabrese E, Villanueva-Meyer JE, Cha S. A fully automated artificial intelligence method for non-invasive, imaging-based identification of genetic alterations in glioblastomas. Scientific Reports 2020; 10 (1): 11852. doi: 10.1038/s41598-020-68857-8
[37] Clark K, Vendt B, Smith K, Freymann J, Kirby J et al. The Cancer Imaging Archive (TCIA): maintaining and operating a public information repository. Journal of Digit Imaging 2013; 26 (6): 1045–1057. doi: 10.1007/s10278- 013-9622-7
[38] Ceccarelli M, Barthel FP, Malta TM, Sabedot TS, Salama SR et al. Molecular profiling reveals biologically discrete subsets and pathways of progression in diffuse glioma. Cell 2016; 164 (3): 550-563.
[39] Tustison NJ, Avants BB, Cook PA, Zheng Y, Egan A et al. N4ITK: improved N3 bias correction. IEEE Transaction on Medical Imaging 2010; 29 (6): 1310-1320.
[40] Tustison NJ, Cook PA, Klein A, Song G, Das SR et al. Large-scale evaluation of ANTs and FreeSurfer cortical thickness measurements. Neuroimage 2014; 99: 166-179.