AI-based mobile application to fight antibiotic resistance

Marco Pascucci,Ellen Sebastian,Dena Soukieh,Nada Malou,Amine Bezzarga,Clara Nordon,Guilhem Royer,Franck Samson,Christian Curel,Alex Brunner,Laetitia Blanche,Mai Al Asmar,Jean-Philippe Vert,Jakub Adámek ,David Aristizabal,Christophe Ambroise,Gabriel Dulac-Arnold,Vincent Runge,Guillaume Boniface-Chang,Rasheed M Fakhri,Mohammed‐Amin Madoui

doi:10.1038/s41467-021-21187-3

Abstract

Antimicrobial resistance is a major global health threat and its development is promoted by antibiotic misuse. While disk diffusion antibiotic susceptibility testing (AST, also called antibiogram) is broadly used to test for antibiotic resistance in bacterial infections, it faces strong criticism because of inter-operator variability and the complexity of interpretative reading. Automatic reading systems address these issues, but are not always adapted or available to resource-limited settings. We present an artificial intelligence (AI)-based, offline smartphone application for antibiogram analysis. The application captures images with the phone’s camera, and the user is guided throughout the analysis on the same device by a user-friendly graphical interface. An embedded expert system validates the coherence of the antibiogram data and provides interpreted results. The fully automatic measurement procedure of our application’s reading system achieves an overall agreement of 90% on susceptibility categorization against a hospital-standard automatic system and 98% against manual measurement (gold standard), with reduced inter-operator variability. The application’s performance showed that the automatic reading of antibiotic resistance testing is entirely feasible on a smartphone. Moreover our application is suited for resource-limited settings, and therefore has the potential to significantly increase patients’ access to AST worldwide.

Highlights

Antimicrobial resistance is a major global health threat and its development is promoted by antibiotic misuse
The App presented in this paper is an automatic AST reading system capable of running the whole analysis of a disk-diffusion antibiogram offline on mobile devices, from image acquisition to interpreted results
The App can be summarized in three major components: first, a dedicated image processing module (IP) that reads and analyzes the AST image; second, an expert system (ES) responsible for the interpretation of the data extracted by IP; third, a Graphical User Interface (GUI)

Summary

Introduction

Antimicrobial resistance is a major global health threat and its development is promoted by antibiotic misuse. While disk diffusion antibiotic susceptibility testing (AST, called antibiogram) is broadly used to test for antibiotic resistance in bacterial infections, it faces strong criticism because of inter-operator variability and the complexity of interpretative reading. The growth of the bacterial colony stops at a distance from the pellet which corresponds to a critical antibiotic concentration, forming a visible bacteria-free area around the cellulose disk This is called a zone of inhibition. The inhibition zone might not be a perfect disk (e.g., if the inhibition zones overlap) or if the pellet is too close to the border of the dish In this case, the problem of measuring a diameter is ill-posed and, together with intrinsic measurement error, introduces subjectivity and inter-operator variability in the measurement. Interpretation is based on expert rules published and updated by scientific societies, such as EUCAST in Europe[11]

Methods

Results

Conclusion