Development & Deployment
Researchers from IIIT-Delhi have developed AI-driven data integration and predictive analytics tools for real-time monitoring of antibiotic resistance.
AMRSense, the AI-powered tool, is deployed in collaboration with CHRI-PATH, Tata 1mg, and ICMR.
The tool utilizes routine hospital data to generate early insights on antimicrobial resistance at global, national, and hospital levels.
Key Findings from Research
Published in The Lancet Regional Health - Southeast Asia, the study analyzed six years of data from 21 tertiary care centers under ICMR’s AMR surveillance network.
The study established relationships between antibiotic pairs and the directional influence of resistance in both community and hospital-acquired infections.
Cost-Effective Approach:
Instead of expensive genomics-based methods, the tool uses hospital routine data to track resistance trends.
If resistance to one antibiotic increases, resistance to related antibiotics may rise in the following months.
AI’s Role in AMR Surveillance
AI can enhance antimicrobial stewardship and surveillance beyond clinical decision-making.
Hospital data (blood, urine, sputum, pus cultures) can be used to create AI-based predictive models for better treatment decisions.
AI can facilitate timely interventions by tracking resistance trends at hospital, regional, and global levels.
AMROrbit Scorecard
Award-winning tool developed by IIIT-Delhi researchers for AMR Surveillance.
Visualizes AMR resistance trends at hospital or department levels against global medians.
Goal: Hospitals should aim for low baseline resistance and low rate of change in AMR trends.
The AI tool allows real-time comparisons between different hospitals, departments, and cities.
Limitations & Future Scope
Reliability:
Models accurately captured historical AMR trends, but future predictions require more data.
Unexpected factors (e.g., COVID-19) can impact predictions.
Data Availability:
AI models may be ineffective in settings with inconsistent or non-digitized surveillance data.
Future Goals:
Integrating hospital data with antibiotic sales and environmental factors (e.g., poultry industry antibiotic use, soil contamination).
COMMENTS