Antibiogramj: A tool for analysing images from disk diffusion tests. (May 2017)

• Record Type: Journal Article
• Authors: Alonso, C.A.; Domínguez, C.; Heras, J.; Mata, E.; Pascual, V.; Torres, C.; Zarazaga, M.
• Is Part Of: Computer methods and programs in biomedicine. Volume 143(2017)
• Page Start: 159
• Page End: 169
• Publication Date: 2017-05
• Subjects: Antibiograms -- Inhibition zone reader -- Disk diffusion test -- ImageJ -- OpenCV -- Learning system; Medicine -- Computer programs -- Periodicals; Biology -- Computer programs -- Periodicals; Computers -- Periodicals; Medicine -- Periodicals; Médecine -- Logiciels -- Périodiques; Biologie -- Logiciels -- Périodiques; Biology -- Computer programs; Medicine -- Computer programs; Periodicals; Electronic journals; 610.28
• DOI: 10.1016/j.cmpb.2017.03.010
• Languages: English
• ISSNs: 0169-2607

Abstract

Highlights: AntibiogramJ determines, measures and categorises inhibition zones of antibiograms. AntibiogramJ deals with images captured with any device that incorporates a camera. AntibiogramJ is user-friendly, open-source, guides users, learns, and stores results. AntibiogramJ provides the benefits of automated readers but without any investment. Abstract: Background and objectives: Disk diffusion testing, known as antibiogram, is widely applied in microbiology to determine the antimicrobial susceptibility of microorganisms. The measurement of the diameter of the zone of growth inhibition of microorganisms around the antimicrobial disks in the antibiogram is frequently performed manually by specialists using a ruler. This is a time-consuming and error-prone task that might be simplified using automated or semi-automated inhibition zone readers. However, most readers are usually expensive instruments with embedded software that require significant changes in laboratory design and workflow. Methods: Based on the workflow employed by specialists to determine the antimicrobial susceptibility of microorganisms, we have designed a software tool that, from images of disk diffusion tests, semi-automatises the process. Standard computer vision techniques are employed to achieve such an automatisation. Results: We present AntibiogramJ, a user-friendly and open-source software tool to semi-automatically determine, measure and categorise inhibition zones of images from disk diffusion tests. AntibiogramJ is implemented in Java and deals with images captured with any device that incorporates a camera, including digital cameras and mobile phones. The fully automatic procedure of AntibiogramJ for measuring inhibition zones achieves an overall agreement of 87% with an expert microbiologist; moreover, AntibiogramJ includes features to easily detect when the automatic reading is not correct and fix it manually to obtain the correct result. Conclusions: AntibiogramJ is a user-friendly, platform-independent, open-source, and free tool that, up to the best of our knowledge, is the most complete software tool for antibiogram analysis without requiring any investment in new equipment or changes in the laboratory.