Tuberculosis is currently the world's leading cause of death from a single infectious disease. In the case of an epidemic the only option of diagnosis remains is the sputum examination. To improve the diagnostic process we are developing an automated method for the detection of tuberculosis bacilli in clinical specimens, preferably sputum smears. Our proposed method makes use of image processing techniques and neural network classifiers for the automatic identification of TB bacilli using Auramine stained specimens of sputum. The developed system , currently shows 93. 5% sensitivity for identifying individual bacilli. There are numerous TB bacilli with active pulmonary TB in the patient's sputum. The overall diagnostic accuracy of the patients with positive smear is expected to be very high. Some potential benefits of automated screening for TB are accurate and rapid diagnosis, increased population screening and reduced health risk.

1. How Many Clusters? Which Clustering Method? Answers Via Model-Based Cluster AnalysisChrisFraley and Adrian. RafteryDepartment of Statistics, University of Washington, USA
2. Identification of Tuberculosis Bacilli using image ProcessingHarshalTalathi, Jinal Patel, KakoliLaha , Mrs. J. M. Kundargi, Department Of Electronics and Telecommunication ,K. J. Somaiya College of Engineering, Mumbai, India.
3. S. Debanne, R. Bielefeld, G. Cauthen, T. Daniel, and D. Rowland, "Multivariate markovian modeling of tuberculosis; forecast for the united states," Emergent infectious diseases 6(2).
4. Ginsberg, A. , The tuberculosis epidemic: scientific challenges and opportunities. . Public Health Reports. v113 i2. 128-136.
5. Veropoulos, K. , Learmonth, G. , Campbell, C. , Knight, B. and Simpson, J. , Automatic identification of tubercle bacilli in sputum. A preliminary investigation. Analytical and Quantitative Cytology and Histology. v21 i4. 277-281.
6. M. Forero, F. Sroubek, J. Alvarez-Borrego, N. Malpica, G. Crist´obal, A. Santos, L. Alcal´a, M. Desco, and L. Cohen, "Segmentation, autofocusing and signature extraction of tuberculosis sputum images," in Photonic devices and algorithms for computing, 2002.
7. Vishnu Makkapati , Ravindra Agrawal , Raviraja Acharya, Segmentation and classification of tuberculosis bacillifrom ZN-stained sputum smear images, Proceedings of the fifth annual IEEE international conference on Automation science and engineering, p. 217-220, August 22-25, 2009, Bangalore, India.
8. An Introduction to Neural NetworksVincent Cheung&Kevin CannonsSignal & Data Compression Laboratory Electrical & Computer EngineeringUniversity of Manitoba Winnipeg, Manitoba, Canada.

Fluorescence Microscopy Pattern Matching Tuberculosis.