Brain Tumour Classification by Machine Learning Applications with Selected Biological Features: Towards A Newer Diagnostic Regime
DOI:
https://doi.org/10.30683/1927-7229.2020.09.02Keywords:
Brain tumour, Astrocytoma, Meningioma, Ependymoma, Support Vector Machine (SVM), Auto-Encoder (AE).Abstract
Histopathologically classified low-grade brain tumours show overlapping biological characteristics making them difficult to distinguish. In the present study low-grade brain tumour patient samples of three different histopathological types have been trained through machine learning technique using selected features for its classification. We used specifically the fundamental proliferation, invasion, macrophage infiltration triangle of cancer hallmark with propidium iodide (PI) marked cell-cycle, Ki67 marked proliferative indexing, invasion with MMP2 expression and presence of macrophage/microglia by silver-gold staining, CD11b+ and Iba1+ cell presence as biological parameters. These parameters when trained with proper machine learning protocol through extraction of underling features and represented in a 2D perceivable space are found capable of distinguishing the tumour types. Extracted features from such parameters in a six-dimensional featured space were trained through statistical learning theory while support vector machine (SVM) maximizes their predictive precision. The leave one out (LOO) cross validation process was applied to judge the accuracy of training followed by auto-encoder (AE) to reduce feature dimension at two which is visually perceptible. From the biological features quantified with standard methods it was found impossible to demarcate the three types of low grade brain tumours. However, after training through SVM and LOO cross validation when the six-dimensional featured space had been reduced into two-dimension using AE, the combined output of the features showed clear zonation in that 2D space. This indicates that the overlapping biological characteristics of these tumour types, when trained through proper support vector machine and reduced from multiple to two dimensional space provides a clear patho-clinical classification edge using a combination of common biological features. Hence, machine learning applications may potentially be used as a complementary diagnostic protocol with the conventional practice.
References
Hanahan D, Weinberg RA. The hallmarks of cancer. Cell 2000; 100: 57-70. https://doi.org/10.1016/S0092-8674(00)81683-9 DOI: https://doi.org/10.1016/S0092-8674(00)81683-9
Hanahan D, Weinberg RA. Hallmarks of cancer: the next generation. Cell 2011; 144: 646-674. https://doi.org/10.1016/j.cell.2011.02.013 DOI: https://doi.org/10.1016/j.cell.2011.02.013
Eckhouse S, Lewison G, Sullivan R. Trends in the global funding and activity of cancer research. Mol Oncol 2008; 2: 20-32. https://doi.org/10.1016/j.molonc.2008.03.007 DOI: https://doi.org/10.1016/j.molonc.2008.03.007
American Brain Tumor Association (Official Home Page, USA). http: //www.abta.org/about-us/news/brain-tumor statistics
Maldjian JA, Patel RS. Cerebral Neoplasms in adults. Semin Roentgenol 1999; 34(2): 102-22. https://doi.org/10.1016/S0037-198X(99)80025-X DOI: https://doi.org/10.1016/S0037-198X(99)80025-X
Bondy ML, Scheurer ME, Malmer B, et al. Brain Tumor Epidemiology: Consensus From the Brain Tumor Epidemiology Consortium. CANCER Suppl 2008; 113(7). 1953-68. https://doi.org/10.1002/cncr.23741 DOI: https://doi.org/10.1002/cncr.23741
Duong LM, McCarthy BJ, McLendon RE, et al. Descriptive epidemiology of malignant and nonmalignant primary spinal cord, spinal meninges, and caudaequina tumors, United States, 2004‐2007. CANCER 2012; 118: 4220-4227. https://doi.org/10.1002/cncr.27390 DOI: https://doi.org/10.1002/cncr.27390
Zitnik M, Nguyen F, Wang B, Leskovec J, Goldenberg A, Hoffman MM. Machine learning for integrating data in biology and medicine: Principals, practice and opportunities. Information Fusion 2019; 50: 71-91. https://doi.org/10.1016/j.inffus.2018.09.012 DOI: https://doi.org/10.1016/j.inffus.2018.09.012
Xu C, Jackson SA. Machine learning and complex biological data. Genome Biology 2019; 20: 76. https://doi.org/10.1186/s13059-019-1689-0 DOI: https://doi.org/10.1186/s13059-019-1689-0
Shavers C, Li R, Lebby G. An SVM based approach to face detection. Proceedings of the 38th Southeastern Symposium on System Theory, Tennessee Technological University Cookeville, TN, USA, IEEE 2006; pp. 362-366.
Tutorial on Support Vector Machine (SVM) by Vikramaditya Jakkula, School of EECS, Washington State University.
Comparative Study of Dimension Reduction Approaches With Respect to Visualization in 3-Dimensional Space, Pooja Chenna, Master's Thesis, Department of Computer Science Kennesaw State University, USA, May, 2016.
Penfield W. A method of staining oligodendroglia and microglia (combined method). Am J Pathol 1928; 4: 153-157.
Penfield W, Cone W. Neuroglia and microglia (the metallic methods). In: McClung CE, Paul B(ed) Handbook of Microscopical Techniques, Hoeber, New York, 1937; pp. 489-521.
Jordanova ES, Corver WE, Vonk MJ, et al. Flow cytometric sorting of paraffin-embedded tumour tissues considerably improves molecular genetic analysis. Am J Clin Pathol 2003; 120: 327-334. https://doi.org/10.1309/HPR11R7LQ9NNCCG8 DOI: https://doi.org/10.1309/HPR11R7LQ9NNCCG8
Burgess CJC. A tutorial on support vector machines for pattern recognition. Data Min Knowl Disc 1998; 2: 955-974. DOI: https://doi.org/10.1162/089976698300017575
Wu Y, Liu Y. Robust Truncated Hinge Loss Support Vector Machines. J Am Stat Assoc 2007; 102(479): 974-983. https://doi.org/10.1198/016214507000000617 DOI: https://doi.org/10.1198/016214507000000617
Elisseeff A, Pontil M. Leave-one-out error and stability of learning algorithms with applications Stability of Randomized Learning Algorithms Source, J Mach Learn Res; January, 2002.
Zhu J, Wu L, Hao H, Song X, Lu Y. Auto-encoder based for high spectral dimensional data classification and visualization, in Proc. IEEE 2nd Int. Conf. Data Sci. Cyberspace, Shenzhen, China, Jun. 2017; pp. 350-354. https://doi.org/10.1109/DSC.2017.32 DOI: https://doi.org/10.1109/DSC.2017.32
Improving Photoplethysmographic Measurements Under Motion Artifacts Using Artificial neural Network for Personal Healthcare, M. Singha Roy, R Gupta, J. K. Chandra, K. Das Sharma, A. Talukdar, IEEE transaction on Instrumentation and Measurement 2018; 67(12). https://doi.org/10.1109/TIM.2018.2829488 DOI: https://doi.org/10.1109/TIM.2018.2829488
Snuderl M, Chi SN, De Santis SM, et al. Prognostic value of tumour microinvasion and metalloproteinases expression in intracranial pediatric ependymomas. J Neuropath Exp Neurol 2008; 67: 911-920. https://doi.org/10.1097/NEN.0b013e318184f413 DOI: https://doi.org/10.1097/NEN.0b013e318184f413
Rooprai HK, Rucklidge GJ, Panou C, et al. The effects of exogenous growth factors on matrix metalloproteinase secretion by human brain tumour cells. Br J Cancer 2000; 82: 52-55. https://doi.org/10.1054/bjoc.1999.0876 DOI: https://doi.org/10.1054/bjoc.1999.0876
Ghosh K, Ghosh S, Chatterjee U, et al. Microglial contribution to glioma progression: An immunohistochemical study in Eastern India. Asian Pac J Cancer Prev 2016; 17: 2767-2773.
Roggendorf W, Strupp S, Paulus W. Distribution and characterization of microglia/macrophages in human brain tumours. Acta Neuropathol 1996; 92: 288-293. https://doi.org/10.1007/s004010050520 DOI: https://doi.org/10.1007/s004010050520
Ghosh A, Sarkar S, Begum Z, et al. The first cross sectional survey on intracranial malignancy in Kolkata, India: Reflection of the state of the art in southern West Bengal. Asian Pac J Cancer Prev 2004; 5: 259-267.
Jaiswal J, Shastry AH, Ramesh A, et al. Spectrum of primary intracranial tumours at a tertiary care neurological institute: A hospital-based brain tumour registry. Neurol India 2016; 64: 494-501. https://doi.org/10.4103/0028-3886.181535 DOI: https://doi.org/10.4103/0028-3886.181535