Radiology Reimagined: Advancing Clinical Practice Through AI Innovation (2025)-"When to Invest Into AI – AI ROI"

"When to Invest Into AI – AI ROI" – Ernest Montana

Pdf Summary

This presentation by Ernest Montañà Ortiz, AI Manager Clinical Applications at Unilabs, outlines when and how to invest in AI for healthcare, focusing on maximizing return on investment (ROI). It emphasizes first identifying specific needs, such as efficiency gains (e.g., managing X-ray backlogs), quality improvements (e.g., better detection rates), or cost-effective workflows. Success depends on having measurable pain points (turnaround times, reading volumes), suitable AI capabilities, and a positive business case.<br /><br />Current AI strengths include triaging, measurements, segmentations, quantifications, and structured report generation, with emerging tools targeting complex report drafting and multimodal AI. Efficiency gains can reduce outsourcing, cut penalties from delays, increase radiologist productivity, and enable serving more clients. However, these gains must be captured through appropriate value sharing among vendors, radiologists, and companies to avoid burnout and ensure sustainability.<br /><br />AI for enhanced quality offers benefits like reducing adverse events and auditing costs, but European reimbursement models rarely reward quality improvements, creating challenges in monetization. Internal data analysis—examining costs of errors, legal claims, and audits—can help build a quality-driven business case.<br /><br />AI can also enable new or modified services, such as AI-assisted double reading in mammography, demonstrated to increase cancer detection by 29% and reduce workload by 44%, lowering reporting costs by 25%. The AI vendor market currently favors buyers, with discounts and pilot projects common; negotiation tactics include leveraging alternative vendors, partnerships, and volume bundling.<br /><br />The recommendation is to start with low-risk, high-potential projects using established AI solutions, avoiding high-cost, low-volume uncertain ventures. A solid AI business case integrates internal data, project costs, assumptions, value capture plans, and scenario modeling.<br /><br />In conclusion, investing in AI makes sense only when needs are quantifiable, solutions are mature or co-developed, execution capacity exists, and a positive business case can be demonstrated.

Keywords

AI in healthcare

Return on Investment

Efficiency gains

Quality improvements

Radiology AI applications

AI business case

AI vendor negotiation

Healthcare workflow optimization

AI-assisted mammography

Value capture in AI