Finalist
Team 13353 from Indraprastha Institute of Information Technology Delhi: Tavpritesh Sethi (Team lead) and Jasmine Kaur
Antimicrobial resistance (AMR) is a systems-level global health challenge requiring coordinated surveillance, stewardship, implementation readiness, and sustained research investment. Whereas AMR surveillance models are becoming increasingly common, these lack integration of systemic factors shaping resistance trajectories, including implementation capacity, health-system preparedness, and allocation of R&D resources. To address this gap bridging analytical insights to learning effective policies, we propose PolarisAMR, a surveillance intelligence and decision-support platform.
Inspired by Polaris, the North Star, PolarisAMR builds upon the trajectory modelling foundations established by our prior Vivli Innovation Award-recognized framework, AMROrbit (preprint: https://doi.org/10.1101/2025.06.05.25328985), which has progressed from retrospective resistance trajectory modelling to prospective real-world validation across two large hospital systems. PolarisAMR is not incremental, instead it is a major enhancement that will integrate indicators of implementation readiness, health-system preparedness, and R&D investment patterns to decrypt the AMROrbit trajectories, guide surveillance-informed interventions and investments which may achieve the greatest public health impact.
Methods: PolarisAMR will estimate how implementation readiness, health-system preparedness, governance capacity, and R&D investment patterns collectively shape the direction, acceleration, and stability of longitudinal AMR trajectories across policy units. Indicators from WHO TrACSS indicators, World Bank health-system metrics, and Global AMR R&D Hub investment datasets will be used. Using mixed-effects dynamical modelling, PolarisAMR will identify hotspots, response archetypes, and investment-alignment patterns associated with resilient or deteriorating resistance trajectories. Specifically, we will define and integrate the vector Li(t) = [Hi, Pi, Gi, Ri, Si] representing the principal components of H = health-system indicators, P = policy implementation readiness, G = governance indicators, R = R&D investment intensity, S = socioeconomic resilience vector, into AMROrbit trajectories. Using mixed effects dynamical models, we will then estimate, and visualize trajectory curvature, κi ∼ f(Li).
Outputs will include integrated AMR intelligence scorecards, alignment indicators, and an open-source interactive dashboard supporting transparent policy translation and evidence-based AMR prioritization.