Mycobacterium Tuberculosis Segmentation through New Pixel Classification Techniques

• Pamela C. Levy UFAM
• Clahildek M. Xavier UFAM
• Marly G. F. Costa UFAM
• Cícero F. F. C. Filho UFAM
• Luciana B. M. Fujimoto INPA
• Julia I. do N. Salem José INPA

Abstract

This paper presents a new method for segmentation of tuberculosis bacillus in conventional sputum smear microscopy. The method comprises two main steps. In the first step, a scalar feature selection is performed. In the second step, two types of pixel classifiers are optimized, using these selected features as inputs: a support vector machine classifier and a feedforward neural network classifier. The inputs for classifiers are selected from features extracted from four color spaces: RGB, HSI, YCbCr and Lab. A sensitivity of 94% is achieved in pixel classification. Nevertheless, as shown, further steps for noise reduction are necessary to minimize the bacilli classification errors.

References

Abe, S. (2010) Support Vector Machines for Pattern Classification: Advances in Pattern Recognition, Springer, London.

Bishop, C. M. (2006) Pattern Recognition and Machine Learning, Springer Science Business Media, Singapore.

Cichoki, A., Unbehauen, R. (1993), Neural Networks for Optimization and Signal Processing, John Wiley & Sons ltd., New York.

Costa, M. G. F., Costa Filho, C. F. F., Sena, J. F., Salem, J., and de Lima, M. O. (2008) “Automatic Identification of Mycobacterium Tuberculosis with Conventional Light Microscopy”, In: 30th Annual International Conference of the IEEE EMBS, pp. 382-385, Vancouver.

Haykin, S. (1999) Neural Networks and Learning Machines, Pearson Prentice Hall, New Jersey.

Khutlang, R., Krishnan, S., Dendere, R., Whitelaw, A., Veropoulos, K., Learmonth, G., Douglas, T. S. (2010) “Classification of Mycobacterium Tuberculosis in Images of ZN-Stained Sputum Smears” In: IEEE transactions on information technology in biomedicine, v. 14, p. 949-957.

Kimura, A., Costa, M.G.F., Filho, C.F.F.C., Fujimoto, L.B.M., Salem, J. (2010) “Evaluation of Autofocus Functions of Conventional Sputum Smear Microscopy for Tuberculosis”, In: Annual International Conference of the IEEE EMBS, p. 3041-3044, Buenos Aires.

Luna, J. A. C. (2004) “A Tuberculosis Guide for Specialist Physicians”, International Union against Tuberculosis and Lung Disease.

Makkapati, V., Agrawal, R., Acharya, R. (2009) “Segmentation and Classification of Tuberculosis Bacilli from ZN-stained Sputum Smear Images”, In: 5th Annual IEEE Conference on Automation Science and Engineering CASE, p. 217-220.

Raof, R.A.A., Salleh, Z., Sahidan, S.I., Mashor, M.Y., Noor, S.S. Md., Mohamad Idris F., Hasan, H. (2008) “Color Thresholding Method for Image Segmentation Algorithm of Ziehl-Neelsen Sputum Slide Images”, In: 5th International Conference on Electrical Engineering, Computing Science and Automatic Control CCE, p. 212-217, Mexico City.

Sadaphal, P., Rao, J., Comstock, G.W., Beg, M.F. (2008)” Image Processing Techniques for Identifying Mycobacterium Tuberculosis in Ziehl-Nelseen Stains”, Int. J. of Tuberculosis and Lung Disease, v. 12, p. 579 – 582.

Sotaquirá, M., Rueda, L., Narvaez, R. (2009) “Detection and Quantification of Bacilli and Clusters Present in Sputum Smear Samples: A Novel Algorithm for Pulmonary Tuberculosis Diagnosis”, In: International Conference on Digital Image Processing, p. 117-121, Bangkok.

Taylor, J. S., Cristianini, N. (2004) Kernel Methods for Pattern Analysis, Cambridge University Press, New York.

Theodoridis , S., Koutroumbas, K. (2006) Pattern Recognition, Elsevier Academic Press, San Diego.

Who (2012) “Tuberculosis facts 2011/2012”, [link], January.