Kirby Bauer Reader

Abstract

The objective of this work is to improve a laboratory experiment called "KirbyBauer". Its a test that allows the identification of the best antibiotic that leads to the treatment of a known bacteria. Several tests have been developed to replace the "KirbyBauer" test, but none of them could have the same cost efficiency. Thus, the objective of this project is to decrease the manual labor and the time needed for this test. The proposed approach consists of five main stages. It is implemented using the software "Matlab". The first stage is the acquisition of the specimens images. It can be acquired using a camera or selected from a set of collected images. Then, the pills are extracted from the image (second stage) and the inhibition zones are detected (Third stage). Each stage is accomplished using a set of morphological operations. The fourth stage involves the classification of the various antibiotics. Having computed the diameter of each inhibition zone associated with a given antibiotic, a comparison is performed with two respective critical diameters to classify the bacteria as resistant, intermediate or sensitive to the antibiotic under study. The identification of a particular antibiotic is achieved based on the list (known a priori) and the order of the antibiotics placement on the agar plate. The final stage (fifth stage) is concerned with the execution of the algorithm, the collection and the presentation of the results on a display (using a graphical user interface) or in a printing form. The implemented approach can save about 20 min of delicate manual labor and a cost of up to {dollar}20 000 in comparisons with a professional device.