Document Type: Original Research

Authors

1 Medical Physics and Biomedical Engineering Department, School of Medicine, Tehran University of Medical Sciences, Tehran, Iran

2 Radiology and Nuclear Medicine Department, School of Allied Medicine, Kermanshah University of Medical Sciences, Kermanshah, Iran

Abstract

Background: Quantitative Magnetization Transfer Imaging (QMTI) is often used to quantify the myelin content in multiple sclerosis (MS) lesions and normal appearing brain tissues. Also, automated classifiers such as artificial neural networks (ANNs) can significantly improve the identification and classification processes of MS clinical datasets.
Objective: We classified patients with relapsing-remitting multiple sclerosis (RRMS) from healthy subjects using QMTI and T1 longitudinal relaxation time data of brain white matter, then the performance of three ANN-based classifiers have been investigated.
Materials and Methods: The input features of ANN algorithms, including multilayer perceptron (MLP), radial basis function (RBF) and ensemble neural networks based on Akaike information criterion (ENN-AIC) were extracted in the form of QMTI and T1 mean values from parametric maps. The ANNs quantitative performance is measured by the standard evaluation of confusion matrix criteria.
Results: The results indicate that ENN-AIC-based classification method has achieved 90% accuracy, 92% sensitivity and 86% precision compared to other ANN models. NPV, FPR and FDR values were found to be 0.933, 0.125 and 0.133, respectively, according to the proposed ENN-AIC model. A graphical representation of how to track actual data by the predictive values derived from ANN algorithms, was also presented.
Conclusion: It has been demonstrated that ENN-AIC as an effective neural network improves the quality of classification results compared to MLP and RBF.In addition, this research provides a new direction to classify a large amount of quantitative MRI data that can help the physician in a correct MS diagnosis.

Keywords

  1. Koriem KMM. Multiple sclerosis: New insights and trends. Asian Pacific Journal of Tropical Biomedicine. 2016;6:429-40.
  2. MacKay A, Laule C, Li DK, Meyers SM, Russell-Schulz B, Vavasour IM, editors Magnetic resonance techniques for investigation of multiple sclerosis. AIP Conference Proceedings. 2014;1626:22–35.
  3. Giorgio A, De Stefano N. Advanced Structural and Functional Brain MRI in Multiple Sclerosis. Semin Neurol. 2016;36:163-76. doi: 10.1055/s-0036-1579737. PubMed PMID: 27116723.
  4. Poloni G, Minagar A, Haacke EM, Zivadinov R. Recent developments in imaging of multiple sclerosis. Neurologist. 2011;17:185-204. doi: 10.1097/NRL.0b013e31821a2643. PubMed PMID: 21712664.
  5. Bodini B, Louapre C, Stankoff B. Advanced imaging tools to investigate multiple sclerosis pathology. Presse Med. 2015;44:e159-67. doi: 10.1016/j.lpm.2015.02.011. PubMed PMID: 25818312.
  6. Moore GR, Laule C. Neuropathologic correlates of magnetic resonance imaging in multiple sclerosis. J Neuropathol Exp Neurol. 2012;71:762-78. doi: 10.1097/NEN.0b013e3182676388. PubMed PMID: 22892523.
  7. Wolff SD, Balaban RS. Magnetization transfer contrast (MTC) and tissue water proton relaxation in vivo. Magn Reson Med. 1989;10:135-44. PubMed PMID: 2547135.
  8. Vavasour IM, Laule C, Li DK, Traboulsee AL, MacKay AL. Is the magnetization transfer ratio a marker for myelin in multiple sclerosis? J Magn Reson Imaging. 2011;33:713-8. doi: 10.1002/jmri.22441. PubMed PMID: 21563257.
  9. Neema M, Stankiewicz J, Arora A, Guss ZD, Bakshi R. MRI in multiple sclerosis: what’s inside the toolbox? Neurotherapeutics. 2007;4:602-17. doi: 10.1016/j.nurt.2007.08.001. PubMed PMID: 17920541.
  10. Ropele S, Strasser-Fuchs S, Augustin M, Stollberger R, Enzinger C, Hartung HP, et al. A comparison of magnetization transfer ratio, magnetization transfer rate, and the native relaxation time of water protons related to relapsing-remitting multiple sclerosis. AJNR Am J Neuroradiol. 2000;21:1885-91. PubMed PMID: 11110542.
  11. Karampekios S, Papanikolaou N, Papadaki E, Maris T, Uffman K, Spilioti M, et al. Quantification of magnetization transfer rate and native T1 relaxation time of the brain: correlation with magnetization transfer ratio measurements in patients with multiple sclerosis. Neuroradiology. 2005;47:189-96. doi: 10.1007/s00234-005-1344-1. PubMed PMID: 15711987.
  12. Cercignani M, Symms MR, Schmierer K, Boulby PA, Tozer DJ, Ron M, et al. Three-dimensional quantitative magnetisation transfer imaging of the human brain. Neuroimage. 2005;27:436-41. doi: 10.1016/j.neuroimage.2005.04.031. PubMed PMID: 15978842.
  13. Levesque IR, Giacomini PS, Narayanan S, Ribeiro LT, Sled JG, Arnold DL, et al. Quantitative magnetization transfer and myelin water imaging of the evolution of acute multiple sclerosis lesions. Magn Reson Med. 2010;63:633-40. doi: 10.1002/mrm.22244. PubMed PMID: 20146232.
  14. Silwattananusarn T, Tuamsuk K. Data mining and its applications for knowledge management: a literature review from 2007 to 2012. arXiv preprint arXiv:12102872. 2012.
  15. Gaur P. Neural networks in data mining. International Journal of Electronics and Computer Science Engineering. 2013;1.
  16. Rajini NH, Bhavani R, editors . Classification of MRI brain images using k-nearest neighbor and artificial neural network. 3-5 June 2011 . Chennai: International Conference on Recent Trends in Information Technology (ICRTIT); 2011.
  17. Chaplot S, Patnaik L, Jagannathan N. Classification of magnetic resonance brain images using wavelets as input to support vector machine and neural network. Biomedical signal processing and control. 2006;1:86-92.
  18. Ramani RG, Sivaselvi K, editors . Classification of Pathological Magnetic Resonance Images of Brain Using Data Mining Techniques. 3-4 Feb. 2017. Tindivanam: Second International Conference on Recent Trends and Challenges in Computational Models (ICRTCCM); 2017.
  19. Amato F, López A, Peña-Méndez EM, Vanhara P, Hampl A, Havel J. Artificial neural networks in medical diagnosis. Journal of Applied Biomedicine. 2013;11:47–58. doi: 10.2478/v10136-012-0031-x.
  20. D Deepa AR, Emmanuel S, Mercy WR. Identification and classification of brain tumor through mixture model based on magnetic resonance imaging segmentation and artificial neural network. Concepts in Magnetic Resonance Part A. 2016;45.
  21. Ahmad F, Ahmad I, Dar WM. Identification and classification of voxels of human brain for rewardless-related decision making using ANN technique. Neural Computing and Applications. 2017;28:1035-41.
  22. Garcia-Lorenzo D, Francis S, Narayanan S, Arnold DL, Collins DL. Review of automatic segmentation methods of multiple sclerosis white matter lesions on conventional magnetic resonance imaging. Med Image Anal. 2013;17:1-18. doi: 10.1016/j.media.2012.09.004. PubMed PMID: 23084503.
  23. Mahbod A, Wang C, Smedby O. Automatic Multiple Sclerosis Lesion Segmentation Using Hybrid Artificial Neural Networks. MSSEG Challenge Proceedings: Multiple Sclerosis Lesions Segmentation Challenge Using a Data Management and Processing Infrastructure. 2016:29.
  24. Valverde S, Cabezas M, Roura E, Gonzalez-Villa S, Pareto D, Vilanova JC, et al. Improving automated multiple sclerosis lesion segmentation with a cascaded 3D convolutional neural network approach. Neuroimage. 2017;155:159-68. doi: 10.1016/j.neuroimage.2017.04.034. PubMed PMID: 28435096.
  25. Soyer P, Rondeau Y, Dufresne A, Spelle L, Somveille E, Scherrer A, et al. T1-weighted spoiled gradient-echo MR imaging of focal hepatic lesion: comparison of in-phase vs opposed-phase pulse sequence. Eur Radiol. 1997;7:1048-53. PubMed PMID: 9265672.
  26. Gan W, editor . Application of neural networks to the processing of medical images. Neural Networks, 1991 1991 IEEE International Joint Conference on; 1991: IEEE.
  27. Dunstone ES, editor . Image processing using an image approximation neural network. Image Processing, 1994 Proceedings ICIP-94, IEEE International Conference; 1994: IEEE.
  28. Hecht-Nielsen R. Theory of the backpropagation neural network. Neural networks for perception: Elsevier; 1992. p. 65-93.
  29. Park J, Sandberg IW. Approximation and radial-basis-function networks. Neural Comput. 1993;5:305-16.
  30. O Orr MJ. Introduction to radial basis function networks. Technical Report, Center for Cognitive Science, University of Edinburgh; 1996.
  31. Krogh A, Vedelsby J, editors. Neural network ensembles, cross validation, and active learning. Adv Neural Inf Process Syst; 1995.
  32. Hansen LK, Salamon P. Neural network ensembles. IEEE transactions on pattern analysis and machine intelligence. 1990;12:993-1001.
  33. Zhao Z, Zhang Y, Liao H. Design of ensemble neural network using the Akaike information criterion. Eng Appl Artif Intell. 2008;21:1182-8.
  34. Posada D, Buckley TR. Model selection and model averaging in phylogenetics: advantages of akaike information criterion and bayesian approaches over likelihood ratio tests. Syst Biol. 2004;53:793-808. doi: 10.1080/10635150490522304. PubMed PMID: 15545256.
  35. Tomas-Fernandez X, Warfield SK, editors. A new classifier feature space for an improved multiple sclerosis lesion segmentation. Biomedical Imaging: From Nano to Macro, 2011 IEEE International Symposium on; 2011: IEEE.
  36. Valverde S, Oliver A, Diez Y, Cabezas M, Vilanova JC, Ramio-Torrenta L, et al. Evaluating the effects of white matter multiple sclerosis lesions on the volume estimation of 6 brain tissue segmentation methods. AJNR Am J Neuroradiol. 2015;36:1109-15. doi: 10.3174/ajnr.A4262. PubMed PMID: 25678478.
  37. Roy S, Butman JA, Reich DS, Calabresi PA, Pham DL. Multiple Sclerosis Lesion Segmentation from Brain MRI via Fully Convolutional Neural Networks. arXiv preprint arXiv:180309172. 2018.
  38. Farahani ES, Choudhury SH, Cortese F, Costello F, Goodyear B, Smith MR. Three-way ROC validation of rs-fMRI visual information propagation transfer functions used to differentiate between RRMS and CIS optic neuritis patients. Conf Proc IEEE Eng Med Biol Soc. 2017;2017:525-8. doi: 10.1109/EMBC.2017.8036877. PubMed PMID: 29059925.
  39. Dong J-r, Zheng C-y, Kan G-y, Zhao M, Wen J, Yu J. Applying the ensemble artificial neural network-based hybrid data-driven model to daily total load forecasting. Neural Computing and Applications. 2015;26:603-11.