Abstract
Background: The Kirby-Bauer method is a widely used technique to evaluate bacterial antibiotic susceptibility based on the diameter of the inhibition zone. However, manual measurement is time-consuming. The application of artificial intelligence (AI) allows for the automation of the image analysis process, minimizing subjective factors and enhancing efficiency.
Material and Methods: The study was conducted on 608 antibiotic susceptibility images of five common bacterial species: Escherichia coli, Staphylococcus aureus, Pseudomonas aeruginosa, Klebsiella pneumoniae, and Acinetobacter baumannii. Each plate contained 4–6 antibiotics, and the images were captured using a smartphone under standard lighting conditions in a DDr box. The collected data were analyzed by an AI model that identifies antibiotic labels and measures the inhibition zone diameter, with the results compared to manual measurements using a technical ruler.
Results: The AI achieved an initial accuracy of approximately 97% in recognizing antibiotic labels, which increased to approximately 99.6% after fine-tuning. The accuracy in measuring the inhibition zone diameter was generally high, although some discrepancies were noted in cases with overlapping zones or unclear boundaries. The average processing time of the AI was significantly lower than that of manual measurement. The integration of a “Human-In-The-Loop” (HITL) mechanism helped in early detection and minimization of errors.
Conclusion: The application of AI in Kirby-Bauer antibiotic susceptibility analysis demonstrates the potential to improve diagnostic efficiency, save time, and reduce errors compared to manual methods. Integrating HITL and fine-tuning the model are key solutions to maintain accuracy while expanding deployment in testing laboratories.
| Published | 2025-04-25 | |
| Fulltext |
|
|
| Language |
|
|
| Issue | Vol. 15 No. 1 (2025) | |
| Section | Original Articles | |
| DOI | 10.34071/jmp.2025.1.24 | |
| Keywords | trí tuệ nhân tạo, Human in the loop, tinh chỉnh, Kirby-Bauer Artificial intelligent, Human-in-the-loop, fine-tune, Kirby-Bauer |
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
Copyright (c) 2025 Hue Journal of Medicine and Pharmacy
Bauer AW, Kirby WM, Sherris JC, Turck M. Antibiotic susceptibility testing by a standardized single disk method. Am J Clin Pathol 1966;45:493–6.
Höring S, Massarani AS, Löffler B, Rödel J. Rapid antibiotic susceptibility testing in blood culture diagnostics performed by direct inoculation using the VITEK®-2 and BD PhoenixTM platforms. European Journal of Clinical Microbiology & Infectious Diseases 2019;38:471–8. https://doi.org/10.1007/s10096-018-03445-3.
Khan ZA, Siddiqui MF, Park S. Current and Emerging Methods of Antibiotic Susceptibility Testing. Diagnostics 2019;9:49. https://doi.org/10.3390/diagnostics9020049.
Balouiri M, Sadiki M, Ibnsouda SK. Methods for in vitro evaluating antimicrobial activity: A review. J Pharm Anal 2016;6:71–9. https://doi.org/10.1016/j. jpha.2015.11.005.
Alonso CA, Domínguez C, Heras J, Mata E, Pascual V, Torres C, et al. Antibiogramj: A tool for analysing images from disk diffusion tests. Comput Methods Programs Biomed 2017;143:159–69. https://doi.org/10.1016/j. cmpb.2017.03.010.
Pascucci M, Royer G, Adamek J, Asmar M Al, Aristizabal D, Blanche L, et al. AI-based mobile application to fight antibiotic resistance. Nat Commun 2021;12:1173. https://doi.org/10.1038/s41467-021-21187-3.
Topol EJ. High-performance medicine: the convergence of human and artificial intelligence. Nat Med 2019;25:44–56. https://doi.org/10.1038/s41591-018- 0300-7.
Litjens G, Kooi T, Bejnordi BE, Setio AAA, Ciompi F, Ghafoorian M, et al. A survey on deep learning in medical image analysis. Med Image Anal 2017;42:60–88. https:// doi.org/10.1016/j.media.2017.07.005.
Mosqueira-Rey E, Hernández-Pereira E, Alonso-Ríos D, Bobes-Bascarán J, Fernández-Leal Á. Human-in-the-loop machine learning: a state of the art. Artif Intell Rev 2023;56:3005–54. https://doi.org/10.1007/s10462-022- 10246-w.
Holzinger A. Interactive machine learning for health informatics: when do we need the human-in-the-loop? Brain Inform 2016;3:119–31. https://doi.org/10.1007/ s40708-016-0042-6.
Sezgin E. Artificial intelligence in healthcare: Complementing, not replacing, doctors and healthcare providers. Digit Health 2023;9. https://doi. org/10.1177/20552076231186520.
Sahiner B, Chen W, Samala RK, Petrick N. Data drift in medical machine learning: implications and potential remedies. Br J Radiol 2023;96. https://doi.org/10.1259/ bjr.20220878.
Kim MJ, Kim SH, Kim SM, Nam JH, Hwang YB, Lim YJ. The Advent of Domain Adaptation into Artificial Intelligence for Gastrointestinal Endoscopy and Medical Imaging. Diagnostics 2023;13:3023. https://doi. org/10.3390/diagnostics13193023.
Feng L, Feng F, Yang Y, Tan M, Wang L. Model adaptive parameter fine-tuning Based on contribution measure for image classification. Neurocomputing 2025;632:129634. https://doi.org/10.1016/j.neucom.2025.129634.