1. Project GuideBenazir N(810011104303) Mr. Nandhi Kesavan RBhuvaneshwari R(810011104304) Batch no: 32 Department of Computer Science Engineering

2.  Objective  Abstract  Existing System  Proposed system  Performance metrics  Results  Result Evaluation  Conclusion  References

3.  To provide a better understanding of normal brain function and the characteristics of disorders of abnormal brain function.

4. This project proposes an automatic support system to detect brain disorder using artificial neural network and clustering methods for medical application. The performance of the ANN classifier was evaluated in terms of training performance and classification accuracies. Artificial Neural Network gives fast and accurate classification than other neural networks and it is a promising tool for classification.

5.  Existing strategies for K-means cannot be straightforwardly transferred to IKM because of the special cluster notion.  Loss of details due to shift invariant property.  Initial centroid value is assumed randomly.

6.  It is not applicable for multiple signals for abnormal detection in a short time.  Initialization of the centroid value is time consuming and requires number of random initializations.  Difficult to get accurate results.

7.  Brain abnormal Detection is based on,  K-Means Clustering for Segmentation.  Artificial Neural Network for classification.

8.  The use of Artificial Neural Network reduces or eliminates the need for a neurologist to analyze the disorder.  Flexible in features detection.  Time consumption is less.

9.  Preprocessing  Centroid Calculation  Segmentation  ANN Classification

10.  OBJECTIVE  To provide a better understanding of normal and disordered Brain function.  GOAL  Find that a Brain image is normal or abnormal (ie.,disordered ) using Classification.  PERFORMANCE METRICS  Centroid calculation for initialization, Wavelet Transformation for reducing noise in the test image.

13.  The proposed system is suitable for brain abnormality detection. In addition, we propose the Artificial Neural Network for classification and K-means, an efficient algorithm for clustering.  Our experimental evaluation demonstrates that the interaction based cluster notion is a valuable complement to existing methods for clustering multivariate time series. This achieves good results on synthetic data and on real world data. Our algorithm is scalable and robust against noise. Moreover, the interaction patterns detected by the classification and K- means are easy to interpret.

14.  In future work the feature selection for interaction-based clustering motivated by text clustering. Graph-based methods are an exciting area of research within the pattern recognition community.  By extending the technique to perform hierarchical clustering, the concepts present in natural language documents usually display some type of hierarchical structure, whereas the algorithm we have presented here identifies only flat clusters. Our main future objective is to extend these ideas to the development of a hierarchical fuzzy relational clustering algorithm.

15. [1] Fox, M. D. and Raichle, M. E.(2007) “Spontaneous fluctuations in brain activity observed with functional magnetic resonance imaging,” Nat. Rev. Neurosci., vol. 8, no. 9, pp. 700–711. [2] Sorg, C. (2007) “Selective changes of resting-state networks in individuals at risk for alzheimer’s disease,” PNAS, vol. 104, no. 47, pp. 18760–18765. [3] Penny, W. D.,Friston, K. J.,Ashburner, J. T.,Kiebel, S. J. and Nichols, T. E. (2007) “Statistical Parametric Mapping: The Analysis of Functional Brain Images” Boston, MA, USA: Elsevier. [4] Smith, S. et al.,(2009) “Correspondence of the brain’s functional architecture during activation and rest,” PNAS, vol. 106, no. 31, p. 13040. [5] Li, C.,Khan, L. and Prabhakaran, B.(2007) “Feature selection for classification of variable length multiattribute motions,” in MultimediaData Mining and Knowledge Discovery, V. A. Petrushin and L. Khan, Eds. London, U.K.: Springer, 2007.

16. [6] Pamminger, C.(2008) “ Bayesian Clustering of Categorical Time Series: An Approach Using Finite Mixtures of Markov Chain Models” Saarbrucken, Germany: VDM Publishing Group. [7] Liao, T. W. (2005) “Clustering of time series data – A survey,” Pattern Recognit., vol. 38, no. 11, pp. 1857–1874. [8] Lauren J. O’donnell and Carl-Fredrik Westin(2007) “Spectral Clustering Of Tractography” IEEE Paper. [9] Claudia Plant, Andrew Zherdin, Christian Sorg, Anke Meyer-Baese, and Afra M. Wohlschläger (2014) “ Mining Interaction Patterns among Brain Regions by Clustering” IEEE Transactions On Knowledge And Data Engineering, Vol. 26, No. 9. [10] Khalid Abualsaudmassudi Mahmuddin mohammad Salehamr Mohamed(2014) “Classification Algorithms Including Bayesnet, Decisiontable, Ibk, J48/C4.5, And Vfi” IEEE Conference.