Background: Tuberculosis (TB) is a major cause of illness and death in many countries, especially in Asia and Africa. Repeated tests of microscopic examination are needed to be performed for early detection of the disease. Hence there is a need to automate the diagnostic process for improvement in the sensitivity and accuracy of the test.
Objective: To automate the decision support system for tuberculosis digital images using histogram based statistical features and evolutionary based extreme learning machines.
Materials and methods: The sputum smear positive and negative images recorded under standard image acquisition protocol are subjected to histogram based feature extraction technique. Most significant features are selected using student ‘t’ test. These significant features are further used as input to the differential evolutionary extreme learning machine classifier.
Results: Results demonstrate that the histogram based significant features are able to differentiate TB positive and negative images with a higher specificity and accuracy.
Conclusion: The methodology used in this work seems to be useful for the automated analysis of TB sputum smear images in mass screening disorders such as pulmonary tuberculosis.