Ab.ai - A Novel Automated AI Tool for Reporting Antibiograms.

Area of Science:

• Microbiology
• Medical Technology
• Artificial Intelligence

Background:

• Antimicrobial resistance (AMR) poses a significant global health threat, exacerbated by antibiotic misuse.
• Accurate antibiotic sensitivity testing (AST) is crucial for effective treatment, guiding appropriate antibiotic selection.
• Current manual interpretation of AST, like the Kirby-Bauer disk diffusion test, suffers from inter-operator variability and requires specialized personnel, particularly challenging in low-resource settings.

Purpose of the Study:

• To develop and validate an automated tool, Ab.ai, for interpreting antibiotic sensitivity testing (AST) results.
• To address the limitations of manual AST interpretation, including subjectivity and resource dependency.
• To enhance the efficiency and reliability of AST in combating antimicrobial resistance.

Main Methods:

• The Ab.ai tool was developed using OpenCV and EasyOCR for automated data collection, processing, and report generation in three phases.
• A total of 50 antibiograms (Gram-positive cocci - GPC and Gram-negative bacilli - GNB) were analyzed.
• The performance of the Ab.ai tool was evaluated against the manual interpretation, considered the gold standard.

Main Results:

• The Ab.ai tool demonstrated high agreement with the manual gold standard: 98.4% for GPC antibiotic susceptibility and 97.6% for GNB antibiotic susceptibility.
• The automation significantly reduces inter-operator variability and potential biases inherent in manual interpretation.
• The tool provides accurate and consistent antibiotic sensitivity reports.

Conclusions:

• The Ab.ai tool is a highly accurate and reliable solution for automating antibiotic sensitivity testing (AST).
• This automated approach can overcome the challenges of manual interpretation, especially in resource-limited environments.
• Ab.ai has the potential to be a game-changer in the global fight against antimicrobial resistance (AMR).