Detection of pseudo brain tumors via stacked LSTM neural networks using MR spectroscopy signals

• Autorzy: Dandil Emre , Karaca Semih

Identyfikatory

DOI

10.1016/j.bbe.2020.12.003

• Języki publikacji: EN

Abstrakty

EN

Magnetic resonance spectroscopy (MRS) is one of the non-invasive tools used in the detection of brain tumors. MRS provides a metabolic profile about the brain. In this profile, MRS patterns of the tumors and pseudo tumors can be similar to each other. For this reason, accurate diagnosis and classification of brain tumor is of vital importance for the patient's treatment planning. It has been widely preferred by physicians in recent years because it does not pose the risk of infection and death due to surgery like biopsy. In this study, binary classification of brain tumors and normal brain tissue with pseudo-brain tumors is achieved via deep neural networks using MRS data. For the classification of MRS signals, a stacked model based on Long Short-Term Memory (LSTM) and Bidirectional Long Short-Term Memory (Bi-LSTM) deep neural networks is proposed. In the experimental studies in the study, MRS signals from normal brain tissue, brain tumor and pseudo-brain tumors in the INTERPRET database are used. Since the MRS data belonging to a large number oftumors and pseudo-tumors are required for training and testing of the LSTM neural networks, the number of data for the MRS dataset is increased by data augmentation methods. Training and testing of the LSTM neural networks used are performed with a repeated 5-fold cross validation and 10 repetitions for each model. As a result of this study, proposed a stacked model for computer-aided binary classification of MRS data, classification results of 93.44%, 85.56%, 88.33% and 99.23% are obtained for the classification of pseudo brain tumor with glioblastoma, diffuse astrocytoma, metastatic brain tumors and normal brain tissue, respectively. Therefore, it is confirmed that the proposed LSTM-based stacked method is successful in detecting pseudo brain tumors using MRS signals.

Słowa kluczowe

EN

pseudo brain tumor; magnetic resonance spectroscopy; deep learning; LSTM

PL

guz rzekomy mózgu; rezonans magnetyczny; uczenie głębokie; LSTM

• Wydawca: Nałęcz Institute of Biocybernetics and Biomedical Engineering of the Polish Academy of Sciences ; Elsevier
• Czasopismo: Biocybernetics and Biomedical Engineering
• Rocznik: 2021
• Tom: Vol. 41, no. 1
• Strony: 173--195
• Opis fizyczny: Bibliogr. 76 poz., rys., tab., wykr.

Twórcy

autor

Dandil Emre

• Department of Computer Engineering, Faculty of Engineering, Bilecik Seyh Edebali University, 11210, Bilecik, Turkey

autor

Karaca Semih

• Information Technologies Department, Bilecik Seyh Edebali University, Bilecik, Turkey

Bibliografia

• [1] Fitzmaurice C, Allen C, Barber RM, Barregard L, Bhutta ZA, Brenner H, et al. Global, regional, and national cancer incidence, mortality, years of life lost, years lived with disability, and disability-adjusted life-years for 32 cancer groups, 1990 to 2015: a systematic analysis for the global burden of disease study. JAMA Oncol 2017;3:524–48.
• [2] WHO. WHO - Cancer;Accessed from: https://www.who.int/news-room/fact-sheets/detail/ cancer2018.
• [3] Neugut AI, Sackstein P, Hillyer GC, Jacobson JS, Bruce J, Lassman AB, et al. Magnetic resonance imaging-based screening for asymptomatic brain tumors: a review. Oncologist 2019;24:375–84.
• [4] Wild C, Weiderpass E, Stewart B. World cancer report: cancer research for cancer prevention. Lyon: International Agency for Research on Cancer; 2020.
• [5] Devos A, Lukas L, Suykens J, Vanhamme L, Tate A, Howe F, et al. Classification of brain tumours using short echo time 1H MR spectra. J Magn Reson 2004;170:164–75.
• [6] Yang G, Raschke F, Barrick TR, Howe FA. Nonlinear laplacian eigenmaps dimension reduction of in-vivo magnetic resonance spectroscopic imaging analysis. International Society for Magnetic Resonance in Medicine (ISMRM). 2013. p. 1967.
• [7] Ladd ME, Bachert P, Meyerspeer M, Moser E, Nagel AM, Norris DG, et al. Pros and cons of ultra-high-field MRI/MRS for human application. Prog Nucl Magn Reson Spectrosc 2018;109:1–50.
• [8] Soares D, Law M. Magnetic resonance spectroscopy of the brain: review of metabolites and clinical applications. Clin Radiol 2009;64:12–21.
• [9] Hekmatnia A, Sabouri M, Ghazavi AH, Far PS, Hekmatnia F, Sofi GJ, et al. Diagnostic value of Magnetic Resonance Spectroscopy (MRS) for detection of Brain Tumors in patients. Med Sci 2019;23:939–45.
• [10] Majós C, Alonso J, Aguilera C, Serrallonga M, Pérez-Martín J, Acebes JJ, et al. Proton magnetic resonance spectroscopy (1 H MRS) of human brain tumours: assessment of differences between tumour types and its applicability in brain tumour categorization. Eur Radiol 2003;13:582–91.
• [11] Louis DN, Perry A, Reifenberger G, Von Deimling A, Figarella-Branger D, Cavenee WK, et al. The 2016 World Health Organization classification of tumors of the central nervous system: a summary. Acta Neuropathol 2016;131:803–20.
• [12] Horská A, Barker PB. Imaging of brain tumors: MR spectroscopy and metabolic imaging. Neuroimaging Clin N Am 2010;20:293–310.
• [13] Burkett JG, Ailani J. An up to date review of pseudotumor cerebri syndrome. Curr Neurol Neurosci Rep 2018;18:1–7.
• [14] Katz VL, Peterson R, Cefalo RC. Pseudotumor cerebri and pregnancy. Am J Perinatol 1989;6:442–5.
• [15] Celebisoy N, Secil Y, Akyürekli Ö. Pseudotumor cerebri: etiological factors, presenting features and prognosis in the western part of Turkey. Acta Neurol Scand 2002;106:367–70.
• [16] Tasdemir HA, Dilber C, Totan M, Onder A. Pseudotumor cerebri complicating measles: a case report and literature review. Brain Dev 2006;28:395–7.
• [17] Milhorat TH. Classification of the cerebral edemas with reference to hydrocephalus and pseudotumor cerebri. Childs Nerv Syst 1992;8:301–6.
• [18] Silva HCA, Callegaro D, Marchiori PE, Scaff M, Tsanaclis A, Maria C. Magnetic resonance imaging in five patients with a tumefactive demyelinating lesion in the central nervous system. Arq Neuropsiquiatr 1999;57:921–6.
• [19] Cianfoni A, Niku S, Imbesi S. Metabolite findings in tumefactive demyelinating lesions utilizing short echo time proton magnetic resonance spectroscopy. Am J Neuroradiol 2007;28:272–7.
• [20] Majos C, Aguilera C, Alonso J, Julia-Sape M, Castaner S, Sanchez J, et al. Proton MR spectroscopy improves discrimination between tumor and pseudotumoral lesion in solid brain masses. Am J Neuroradiol 2009;30:544–51.
• [21] Vieira BH, Santos ACD, Salmon CEG. Pattern recognition of abscesses and brain tumors through MR spectroscopy: comparison of experimental conditions and radiological findings. Res Biomed Eng 2017;33:185–94.
• [22] McBride DQ, Miller BL, Nikas DL, Buchthal S, Chang L, Chiang F, et al. Analysis of brain tumors using 1H magnetic resonance spectroscopy. Surg Neurol 1995;44:137–44.
• [23] Hourani R, Brant L, Rizk T, Weingart JD, Barker PB, Horská A. Can proton MR spectroscopic and perfusion imaging differentiate between neoplastic and nonneoplastic brain lesions in adults? Am J Neuroradiol 2008;29:366–72.
• [24] Weis J, Ring P, Olofsson T, Ortiz-Nieto F, Wikström J. Short echo time MR spectroscopy of brain tumors: grading of cerebral gliomas by correlation analysis of normalized spectral amplitudes. J Magn Reson Imaging 2010;31:39–45.
• [25] Kaur T, Saini BS, Gupta S. An optimal spectroscopic feature fusion strategy for MR brain tumor classification using Fisher Criteria and Parameter-Free BAT optimization algorithm. Biocybern Biomed Eng 2018;38:409–24.
• [26] Lisboa P, Kirby S, Vellido A, Lee Y, El-Deredy W. Assessment of statistical and neural networks methods in NMR spectral classification and metabolite selection. NMR Biomed 1998;11:225–34.
• [27] Butzen J, Prost R, Chetty V, Donahue K, Neppl R, Bowen W, et al. Discrimination between neoplastic and nonneoplastic brain lesions by use of proton MR spectroscopy: the limits of accuracy with a logistic regression model. Am J Neuroradiol 2000;21:1213–9.
• [28] Tate AR, Underwood J, Acosta DM, Julià-Sapé M, Majós C, Moreno-Torres À, et al. Development of a decision support system for diagnosis and grading of brain tumours using in vivo magnetic resonance single voxel spectra. NMR Biomed 2006;19:411–34.
• [29] Arús C, Celda B, Dasmahaptra S, Dupplaw D, GonzalezVelez H, Van Huffel S, et al. On the design of a web-based decision support system for brain tumour diagnosis using distributed agents. 2006 IEEE/WIC/ACM International Conference on Web Intelligence and Intelligent Agent Technology Workshops; 2006. pp. 208–11.
• [30] Georgiadis P, Kostopoulos S, Cavouras D, Glotsos D, Kalatzis I, Sifaki K, et al. Quantitative combination of volumetric MR imaging and MR spectroscopy data for the discrimination of meningiomas from metastatic brain tumors by means of pattern recognition. Magn Reson Imaging 2011;29:525–35.
• [31] Faria AV, Macedo Jr F, Marsaioli A, Ferreira M, Cendes F. Classification of brain tumor extracts by high resolution 1 H MRS using partial least squares discriminant analysis. Braz J Med Biol Res 2011;44:149–64.
• [32] Tsolaki E, Svolos P, Kousi E, Kapsalaki E, Fountas K, Theodorou K, et al. Automated differentiation of glioblastomas from intracranial metastases using 3T MR spectroscopic and perfusion data. Int J Comput Assist Radiol Surg 2013;8:751–61.
• [33] Vicente J, Fuster-Garcia E, Tortajada S, García-Gómez JM, Davies N, Natarajan K, et al. Accurate classification of childhood brain tumours by in vivo1H MRS–a multi-centre study. Eur J Cancer 2013;49:658–67.
• [34] Nachimuthu DS, Baladhandapani A. Multidimensional texture characterization: on analysis for brain tumor tissues using MRS and MRI. J Digit Imaging 2014;27:496–506.
• [35] Yang G, Raschke F, Barrick TR, Howe FA. Manifold Learning in MR spectroscopy using nonlinear dimensionality reduction and unsupervised clustering. Magn Reson Med 2015;74:868–78.
• [36] Lukas L, Devos A, Suykens JA, Vanhamme L, Howe FA, Majós C, et al. Brain tumor classification based on long echo proton MRS signals. Artif Intell Med 2004;31:73–89.
• [37] Arizmendi C, Sierra DA, Vellido A, Romero E. Automated classification of brain tumours from short echo time in vivo MRS data using Gaussian Decomposition and Bayesian Neural Networks. Expert Syst Appl 2014;41:5296–307.
• [38] Yang G, Nawaz T, Barrick TR, Howe FA, Slabaugh G. Discrete wavelet transform-based whole-spectral and subspectral analysis for improved brain tumor clustering using single Voxel MR Spectroscopy. IEEE Trans Biomed Eng 2015;62:2860-6.
• 39] Crain ID, Elias PS, Chapple K, Scheck AC, Karis JP, Preul MC. Improving the utility of 1 H-MRS for the differentiation of glioma recurrence from radiation necrosis. J Neurooncol 2017;133:97–105.
• [40] Depciuch J, Tolpa B, Witek P, Szmuc K, Kaznowska E, Osuchowski M, et al. Raman and FTIR spectroscopy in determining the chemical changes in healthy brain tissues and glioblastoma tumor tissues. Spectrochim Acta A Mol Biomol Spectrosc 2020;225:1–8.
• [41] Lu D, Polomac N, Gacheva I, Hattingen E, Triesch J. Humanexpert-Level brain tumor detection using deep learning with data distillation and augmentation. arXiv preprint arXiv:200612285 2020.
• [42] Dandil E. aiMRS: a feature extraction method from MRS signals based on artificial immune algorithms for classification of brain tumours. IET Signal Process 2020;14:361–73.
• [43] Dandil E, Biçer A. Automatic grading of brain tumours using LSTM neural networks on magnetic resonance spectroscopy signals. IET Image Process 2020;14:167–1979.
• [44] Callot V, Galanaud D, Le Fur Y, Confort-Gouny S, Ranjeva JP, Cozzone PJ. 1H MR spectroscopy of human brain tumours: a practical approach. Eur J Radiol 2008;67:268–74.
• [45] Kimura T, Sako K, Gotoh T, Tanaka K, Tanaka T. In vivo single-voxel proton MR spectroscopy in brain lesions with ring-like enhancement. NMR Biomed 2001;14:339–49.
• [46] Ramin SL, Tognola WA, Spotti AR. Proton magnetic resonance spectroscopy: clinical applications in patients with brain lesions. Sao Paulo Med J 2003;121:254–9.
• [47] Howe FA, Opstad KS. 1H MR spectroscopy of brain tumours and masses. NMR Biomed 2003;16:123–31.
• [48] Nagori M, Joshi M. Methods and algorithms for extracting values from MRS graph for brain tumour detection. Ieri Procedia 2013;4:331–6.
• [49] Blüml S. Magnetic resonance spectroscopy: basics. MR spectroscopy of pediatric brain disorders. Springer; 2013. p. 11–23.
• [50] Yang G, Raschke F, Barrick TR, Howe FA. Classification of brain tumour 1 h mr spectra: extracting features by metabolite quantification or nonlinear manifold learning? 2014 IEEE 11th International Symposium on Biomedical Imaging (ISBI); 2014. pp. 1039–42.
• [51] Moonen CT, Kienlin MV, Van Zijl PC, Cohen J, Gillen J, Daly P, et al. Comparison of single-shot localization methods (STEAM and PRESS) for in vivo proton NMR spectroscopy. NMR Biomed 1989;2:201–8.
• [52] van der Graaf M. In vivo magnetic resonance spectroscopy: basic methodology and clinical applications. Eur Biophys J 2010;39:527–40.
• [53] INTERPRET Consortium. International network for pattern recognition of tumours using magnetic resonance; 2002.
• [54] Julià-Sapé M, Acosta D, Mier M, Arùs C, Watson D. A multicentre, web-accessible and quality control-checked database of in vivo MR spectra of brain tumour patients. Magn Reson Mater Phys Biol Med 2006;19:22–33.
• [55] García-Gómez JM, Luts J, Julià-Sapé M, Krooshof P, Tortajada S, Robledo JV, et al. Multiproject–multicenter evaluation of automatic brain tumor classification by magnetic resonance spectroscopy. Magn Reson Mater Phys Biol Med 2009;22:5–18.
• [56] Wu Y, Schuster M, Chen Z, Le QV, Norouzi M, Macherey W, et al. Google's neural machine translation system: bridging the gap between human and machine translation. arXiv preprint arXiv:160908144 2016.
• [57] Graves A, Mohamed A-R, Hinton G. Speech recognition with deep recurrent neural networks. 2013 IEEE International Conference on Acoustics, Speech and Signal Processing;2013. pp. 6645-9.
• [58] Zhang X-Y, Xie G-S, Liu C-L, Bengio Y. End-to-end online writer identification with recurrent neural network. IEEE Trans Hum Mach Syst 2016;47:285–92.
• [59] Hopfield JJ. Neural networks and physical systems with emergent collective computational abilities. Proc Natl Acad Sci 1982;79:2554–8.
• [60] Hochreiter S, Schmidhuber J. Long short-term memory. Neural Comput 1997;9:1735–80.
• [61] Schuster M, Paliwal KK. Bidirectional recurrent neural networks. Ieee Trans Signal Process 1997;45:2673–81.
• [62] Pascanu R, Gulcehre C, Cho K, Bengio Y. How to construct deep recurrent neural networks. arXiv preprint arXiv:13126026 2013.
• [63] Peebles Jr PZ. Probability, random variables, and random signal principles. McGraw Hill Book Company; 1987.
• [64] Sengur A. Multiclass least-squares support vector machines for analog modulation classification. Expert Syst Appl 2009;36:6681–5.
• [65] Bottou L. Large-scale machine learning with stochastic gradient descent. Proceedings of COMPSTAT'2010; 2010. p. 177–86.
• [66] Bulik M, Jancalek R, Vanicek J, Skoch A, Mechl M. Potential of MR spectroscopy for assessment of glioma grading. Clin Neurol Neurosurg 2013;115:146–53.
• [67] Rand S, Prost R, Haughton V, Mark L, Strainer J, Johansen J, et al. Accuracy of single-voxel proton MR spectroscopy in distinguishing neoplastic from nonneoplastic brain lesions. Am J Neuroradiol 1997;18:1695–704.
• [68] Ding H, Guo L, Zhao C, Wang F, Wang G, Jiang Z, et al. RFnet: automatic gesture recognition and human identification using time series RFID signals. Mob Netw Appl 2020;1–14.
• [69] Selvaraju RR, Cogswell M, Das A, Vedantam R, Parikh D, Batra D. Grad-cam: visual explanations from deep networks via gradient-based localization. Proceedings of the IEEE International Conference on Computer Vision; 2017. p. 618-26.
• [70] Li Y, Yang H, Li J, Chen D, Du M. EEG-based intention recognition with deep recurrent-convolution neural network: performance and channel selection by Grad-CAM. Neurocomputing 2020;415:225–33.
• [71] Majós C, Julià-Sapé M, Alonso J, Serrallonga M, Aguilera C, Acebes JJ, et al. Brain tumor classification by proton MR spectroscopy: comparison of diagnostic accuracy at short and long TE. Am J Neuroradiol 2004;25:1696–704.
• [72] Luts J, Heerschap A, Suykens JA, Van Huffel S. A combined MRI and MRSI based multiclass system for brain tumour recognition using LS-SVMs with class probabilities and feature selection. Artif Intell Med 2007;40:87–102.
• [73] González-Navarro FF, Belanche-Muñoz LA. Using machine learning techniques to explore 1H-MRS data of brain tumors. 2009 Eighth Mexican International Conference on Artificial Intelligence; 2009. pp. 134–9.
• [74] Dimou I, Tsougos I, Tsolaki E, Theodorou K. Classification of pathological human brain lesions using magnetic resonance spectroscopy at 3T. World congress on medical physics and biomedical engineering. Munich, Germany: Springer; 2009. p. 1368–70. September 7-12, 2009.
• [75] Lu D, Sun Y, Wan S. Brain tumor classification using nonnegative and local non-negative matrix factorization. 2013 IEEE International Conference on Signal Processing, Communication and Computing (ICSPCC 2013); 2013. pp. 1–4.
• [76] Wang L, Wan S, Sun Y, Zhang B, Zhang X. Automatic classification of brain tumor by in vivo MRS data based on LDA and SVM. 2015 Seventh International Conference on Measuring Technology and Mechatronics Automation; 2015. pp. 213–6.

Uwagi

Opracowanie rekordu ze środków MNiSW, umowa Nr 461252 w ramach programu "Społeczna odpowiedzialność nauki" - moduł: Popularyzacja nauki i promocja sportu (2021).