AI in Microbiology and Infectious Diseases
AI is routine in chest X-ray reading for TB and in gene-based tools that find antibiotic-resistance genes in sequenced bacteria. Models that predict resistance from clinical data or genomes are published widely but are still moving into practice. Data gaps and weak interpretability are the main problems.
What AI does in Microbiology and Infectious Diseases
AI reads the X-ray and prioritises people for confirmatory testing. WHO has recommended CAD for people aged 15 and above.
Examples: qXR (Qure.ai), CAD4TB, Lunit INSIGHT CXR
Tools scan sequenced isolates for acquired resistance genes and mutations.
Examples: AMRFinderPlus (NCBI), CARD Resistance Gene Identifier
Systems suggest empirical antibiotics from local data and patient factors.
Examples: Not listed
Models predict which antibiotics will fail before culture results return. Reviews list data quality and bias as limits.
Examples: Random forest, support vector machine and deep learning models in published studies
Models combine epidemiological and genomic data to forecast resistance spread.
Examples: Not listed
routine = in everyday use in many centres · early adoption = in use at some sites, evidence still building · research = studies only
Key ideas for this specialty
New to the basics? Start with the theory lessons.
Tools
Curated resistance gene database with the Resistance Gene Identifier for web or command-line use.
Regulatory status: Not listed
Finds acquired resistance genes and point mutations in protein or nucleotide sequences; documented on GitHub.
Regulatory status: Not listed
Chest X-ray AI with TB screening; the vendor page shows CE and FDA marks and no CDSCO mark.
Regulatory status: CE marked and FDA cleared per vendor page; CDSCO not shown on page
Literature search and data extraction for reviews on resistance or infection topics.
Regulatory status: Not listed
Checks whether later studies support or dispute a paper's finding.
Regulatory status: Not listed
A listing here is not an endorsement. Ask any vendor for its CDSCO licence and class before clinical use.
Where to learn
Open review of the methods and obstacles for resistance prediction (2024).
2025 review that covers data sources and clinical decision support.
Focuses on what stops models reaching the lab bench.
Install guide and usage for the standard resistance gene tool.
Plain introduction to ML and how to judge accuracy claims.
Teaches how models are built and tested on medical data.
Do it yourself
Download a bacterial genome FASTA from NCBI and upload it to the CARD Resistance Gene Identifier. Compare the genes found with the known susceptibility profile of that strain.
Open the dataset or tool →Follow the GitHub wiki to install it and run it on one assembled genome. Write down which resistance genes it reports and what drugs they affect.
Open the dataset or tool →Use public or properly de-identified data only, and get ethics approval before any project on your own patients.
Papers worth reading
- Artificial intelligence in predicting pathogenic microorganisms' antimicrobial resistance: challenges, progress, and prospects Frontiers in Cellular and Infection Microbiology, 2024
- The role of artificial intelligence and machine learning in predicting and combating antimicrobial resistance Computational and Structural Biotechnology Journal, 2025
- AI-CAD enhances pulmonary TB detection and yield in active case finding IJTLD Open, 2025
Societies and guidance
TB reports and technical guidance including CAD recommendations.
Ethical guidelines for AI in biomedical research.
Recent updates
Chhattisgarh study: AI chest X-ray screening for TB raised case notifications by 80%
In Open Forum Infectious Diseases, qXR reading of 2,745 X-rays in a tribal population flagged 363 presumptive cases, 162 were confirmed, and notifications rose 80.21% against baseline. It also found 51 asymptomatic people who symptom screening would have missed.
Source: Open Forum Infectious Diseases (PMC)Links checked 3 October 2026. “Not listed” means we could not confirm the detail, not that it is absent. Educational material, not clinical advice.