Reflections

Notes on AI, evidence, and systematic reviews

Expert insights on healthcare AI, HEOR, HTA, and the science of getting therapies to patients faster.

Latest · May 25, 2026Calibrated Confidence over Active Learning: Methodological Foundations for Reliable Abstract ScreeningTwo paradigms compete to accelerate abstract screening: active-learning rankers and autonomous, confidence-calibrated agentic AI. The published evidence reframes the choice as one of scientific reliability, reproducibility, and risk transparency.Mara Rada · 5 min read

From the article"Active-learning stopping rules should be applied cautiously — the first systematic empirical test shows each one fails. A platform that produces a deterministic, calibrated decision for every record can meet the reproducibility standard systematic review methodology requires."

Compliance · November 24, 2025Methodological Compliance for The Use of Loon AI® in Evidence Synthesis: How Loon’s Confidence Calibration Meets NICE, CDA-AMC, and ISPOR Auditing RequirementsMara Rada · 5 min read Systematic reviews · May 10, 2025Making Systematic Reviews Feasible: Evidence-Based Automation Through Validated AIDr. Ghayath Janoudi · 6 min read Loon Lens™ · December 10, 2024Loon Lens™: Autonomous AI Agents for Literature Screening in Systematic ReviewsDr. Ghayath Janoudi · 8 min read Compliance · December 10, 2024Loon's Compliance with the NICE Position, CDA-AMC Guidelines, and ISPOR ELEVATE-AI LLMs Framework on The Use of AI in Evidence SynthesisLoon Team · 10 min read Validation · November 20, 2024AI Literature Screening: Evidence-Based Validation for Systematic Review AutomationDr. Ghayath Janoudi · 7 min read