Exploring Statistical Parameters of Machine Learning Techniques for Detection and Classification of Brain Tumor (original) (raw)

Abstract

A computerized system can improve the disease identifying abilities of doctor and also reduce the time needed for the identification and decision-making in healthcare. Gliomas are the brain tumors that can be labeled as Benign (non- cancerous) or Malignant (cancerous) tumor. Hence, the different stages of the tumor are extremely important for identification of appropriate medication. In this paper, a system has been proposed to detect brain tumor of different stages by MR images. The proposed system uses Fuzzy C-Mean (FCM) as a clustering technique for better outcome. The main focus in this paper is to refine the required features in two steps with the help of Discrete Wavelet Transform (DWT) and Independent Component Analysis (ICA) using three machine learning techniques i.e. Random Forest (RF), Artificial Neural Network (ANN) and Support Vector Machine (SVM). The final outcome of our experiment indicated that the proposed computerized system identifies the brain tumor using RF, AN...

Key takeaways

sparkles

AI

  1. The proposed system detects brain tumor stages using MRI with 100% accuracy in Random Forest classification.
  2. Fuzzy C-Mean (FCM) enhances clustering for MR image analysis, improving segmentation accuracy.
  3. Independent Component Analysis (ICA) refines feature reduction for better classification results.
  4. AUC-ROC values indicate Random Forest outperforms ANN and SVM in distinguishing tumor types.
  5. The study emphasizes the importance of accurate ROI extraction for effective tumor identification.

Loading...

Loading Preview

Sorry, preview is currently unavailable. You can download the paper by clicking the button above.

References (28)

  1. Gonzalez, R. C., & Woods, R. E. (2008). Digital image processing (3rd ed.). Pearson Prentice Hall (Prentice Hall, 2002).
  2. E.-S. A. El-Dahshan, H. M. Mohsen, K. Revett, & A.-B. M. Salem, "Computer-aided diagnosis of human brain tumor through MRI: A survey and a new algorithm", Expert Systems with Applications, vol. 41, pp. 5526-5545, 2014.
  3. F. P. Polly, S. K. Shil, M. A. Hossain, A. Ayman, & Y. M. Jang, "Detection and Classification of HGG and LGG Brain Tumor Using Machine Learning", International Conference on Information Networking (ICOIN), pp. 813-817, 2018.
  4. E. Dandil, M. Cakiroglu, Z. Eksi, "Computer-aided diagnosis of malign and benign brain tumors on MR images", International Conference on ICT Innovations, pp. 157-166, 2014.
  5. Midas: healthy human brain database. Available online: http://www.insightjournal.org/midas/collection/view/27
  6. Brainweb (2014). Brainweb: Simulated brain database.http://brainweb.bic.mni.mcgill.ca/cgi/brainweb2
  7. M. A. Balafar, "Fuzzy C-mean based brain MRI segmentation algorithms", Artificial Intelligence Review, vol. 41, pp. 441-449, 2014.
  8. Gasmi, K., Kharrat, A., Messaoud, M. B., & Abid, M., "Automated segmentation of brain tumor using optimal texture features and support vector machine classifier", International Conference on Image Analysis and Recognition LNCS, pp. 230-239, 2012.
  9. Jafari, M., & Kasaei, S., "Automatic brain tissue detection in MRI images using seeded region growing segmentation and neural network classification", Australian Journal of Basic and Applied Sciences, vol. 5(8), pp.1066-1079, 2011.
  10. Masulli, F., & Schenone, A., "A fuzzy clustering based segmentation system as support to diagnosis in medical imaging", Artificial Intelligence in Medicine, vol.16, pp.129-147, 1999.
  11. Tabatabai, G., Stupp, R., Bent, M. J., Hegi, M.E., Tonn, J.C., Wick, W., & Weller, M., "Molecular diagnostics of gliomas: the clinical perspective", Acta Neuropathologica, vol. 120, pp. 585-592, 2010.
  12. Chaplot, S., Patnaik, L.M., & Jagannathan, N.R., "Classification of magnetic resonance brain images using wavelets as input to support vector machine and neural network", Biomedical Signal Processing and Control, vol. 1, pp. 86-92, 2006.
  13. Van Meir, E.G., Hadjipanayis, C.G., Norden, A.D., Shu, H.K., Wen, P.Y., & Olson, J.J., "Exciting new advances in neuro-oncology: theavenue to a cure for malignant glioma", A Cancer Journal for Clinicians, vol. 60, pp. 166-193, 2010.
  14. Phillips, R. Velthuizen,S. Phupanich, L. Hall, L. Clarke, & M. Silbiger, "Application of fuzzy C-means segmentation technique for tissue differentiation in MR images of a hemorrhagic glioblastoma multiform", Magnetic Resonance Imaging, vol. 29, pp. 277-290, 1995.
  15. Khotanlou, H., Colliot, O., Atif, J., & Bloch, I., "3D brain tumor segmentation in MRI using fuzzy classification, symmetry analysis and spatially constrained deformable models", Fuzzy Sets and System, vol. 160, pp. 1457-1473, 2009.
  16. A. E. Lashkari, "A Neural Network based Method for Brain Abnormality Detection in MR Images Using Gabor Wavelets", International Journal of Computer Applications, vol. 4, pp.841-1140, 2010.
  17. Pauline, J., "Brain Tumor Classification Using Wavelet and Texture Based Neural Network", International Journal of Scientific & Engineering Research, vol. 3, pp.1-7, 2012.
  18. Kolen, J.F., Hutcheson, T., "Reducing the time complexity of the fuzzy c-means algorithm", IEEE Transactions on Fuzzy Systems, vol. 10, pp.263-267, 2002.
  19. Murugavalli, S., Rajamani, V., "A high speed parallel fuzzy c-mean algorithm for brain tumor segmentation", Bioinform. Medical Engineering Journal, vol. 6, pp.29-34, 2006.
  20. Zacharaki, E.I., Wang, S., Chawla, S., Yoo, D.D., Wolf, R., Melhem, E.R., & Davatzikos, C., "Classification of Brain Tumor Type and Grade Using MRI Texture and Shape in a Machine Learning Scheme", Magnetic Resonance in Medicine, vol. 62, pp. 1609-1618, 2009.
  21. Gambino, O., Daidone, E., Sciortino, M., Pirrone, R., & Ardizzone, E., "Automatic Skull Stripping in MRI based on Morphological Filters and Fuzzy C-means Segmentation", Annual International Conference of the IEEE Engineering in Medicine and Biology Society, pp.5040-5043, 2011.
  22. Bahadure, N.B., Ray, A.K., & Thethi, H.P., "Image Analysis for MRI Based Brain Tumor Detection and Feature Extraction Using Biologically Inspired BWT and SVM", Hindawi International Journal of Biomedical Imaging, vol. 2017, pp. 1-12, 2017.
  23. AyseDemirhan, Mustafa Toru, InanGuler, "Segmentation of Tumor and Edema Along with healthy tissues of brain using Wavelets and Neural Networks", IEEE Journal of Biomedical and Health Informatics, vol.19, pp.1451-1458, 2015.
  24. Dipali M Joshi, Rana N K & Misra VM, "Classification of Brain Cancer Using Artificial Neural Network", IEEE Conference on Electronic Computer Technology(ICECT), pp.112-116, 2010.
  25. L. Breiman, Random forests, Machine learning, vol. 45, pp. 5-32, 2001.
  26. S. Haykin, Neural Networks: A Comprehensive Foundation, second ed., PHI, 1994.
  27. D. M. W. Powers, Evaluation: From Precision, Recall and F-Measure to ROC, Informedness, Markedness & Correlation, 2011, vol. 2, pp. 37-63.
  28. T. Fawcett. (2006, June). An Introduction to ROC Analysis. Pattern Recognition Letters. vol. 27, pp. 861-874.

FAQs

sparkles

AI

What classification techniques achieve the highest accuracy in brain tumor detection?add

The study finds that the Random Forest (RF) classifier achieves the highest accuracy at 99.16% and AUC-ROC of 1.0 in differentiating tumor types.

How does Independent Component Analysis (ICA) improve feature reduction?add

ICA effectively maximizes independence in image processing, aiding in feature extraction by separating unknown signal sources without prior knowledge.

What is the role of Fuzzy C-Mean (FCM) in MRI image segmentation?add

FCM enhances accuracy in classifying tumor regions by allowing data points to belong to multiple clusters, thereby improving segmentation.

How significant is the AUC-ROC metric in evaluating classifier performance?add

An AUC-ROC of 1.0 for RF suggests perfect discrimination, while SVM and ANN demonstrated AUCs of 0.96 and 0.92, indicating their comparative effectiveness.

What preprocessing techniques are vital for MRI image analysis?add

Crucial preprocessing techniques include binarization and morphological operations, which enhance image quality and reduce noise before segmentation.