FLARE Framework Evaluates AI Adoption in Healthcare

The integration of artificial intelligence into healthcare workflows is accelerating, yet most evaluations focus solely on model accuracy rather than the economic and operational viability of adoption in real clinical settings. To address this gap, researchers have developed FLARE, a systematic framework designed to assess the financial and operational implications of AI adoption while explicitly accounting for uncertainty. FLARE integrates fuzzy logic, time-driven activity-based costing, and return on investment (ROI) analysis. This allows it to estimate the costs associated with clinical service delivery, AI development and operation, and the economic consequences of integrating AI into existing workflows. The framework's utility was demonstrated through a case study involving AI-assisted detection of large vessel occlusion in the CT stroke pathway. The case study revealed that FLARE can quantify conventional pathway costs, AI-related expenses, and potential savings within a unified activity-based model. The analysis identified a break-even point and positive first-year ROI under typical patient volumes, emphasizing that economic benefit depends not only on algorithmic performance but also on factors like patient volume, verification time, infrastructure choices, and workflow design. FLARE offers a transparent decision-support tool for early-stage AI adoption evaluation in healthcare, making trade-offs and uncertainties explicit for stakeholders.