IHI Call 9 · Grant Agreement No. 101253015

AI-powered guidance for chronic disease management

GUIDE-AI develops intelligent Guideline Navigators — LLM-based navigational AI systems that help clinicians navigate complex guidelines for HFrEF (Heart Failure with reduced Ejection Fraction), CKD (Chronic Kidney Disease), COPD (Chronic Obstructive Pulmonary Disease), asthma, and IBD (Inflammatory Bowel Disease). Together, these chronic conditions are responsible for a large share of morbidity and mortality in Europe.

19
Consortium Partners
5
Target Disease Areas
8
European Countries
€9.5M
Total Project Budget

Bridging the gap between guidelines and bedside care

Clinical practice guidelines summarise the best available evidence, yet applying them to individual patients with different chronic conditions remains a complex, time-consuming challenge. GUIDE-AI makes guideline-based navigation fast, reliable, and explainable.

Large Language Models

We align state-of-the-art LLMs with curated clinical guideline corpora to enable natural-language guideline query and reasoning.

Clinical Validation

Guideline Navigators are validated in real clinical workflows across two EU health systems, ensuring safety, usability, and clinical relevance.

Explainable & Safe

Every recommendation is traceable to its source guideline. We rigorously assess AI safety, hallucination risk, and regulatory compliance.

From complex guidelines to clinical action

⚠ The Clinical Challenge

  • Clinical guidelines run to hundreds of pages and are updated frequently, making it impossible for busy clinicians to stay current
  • Existing guideline navigation tools are rigid, keyword-based, and unable to handle natural clinical questions
  • Guideline adherence in HFrEF, CKD, COPD, and asthma remains suboptimal, leading to avoidable hospitalisations and undertreatment
  • Static guidelines struggle to account for the complexity of multimorbid patients, and forcing automated AI decisions in ambiguous scenarios risks overriding essential human clinical judgment

✦ The GUIDE-AI Approach

  • LLM-based Guideline Navigators that understand free-text clinical questions and retrieve evidence-grounded answers
  • Transparent source attribution — every answer links back to the specific guideline recommendation
  • Prospective clinical validation in two EU healthcare settings with measurable adherence outcomes
  • Open science approach: validated datasets and model benchmarks shared with the research community
  • A cooperative AI partnership that prevents errors of omission in clear-cut cases, while mapping valid alternative pathways in ambivalent scenarios to empower, not replace, human responsibility and patient preference

Work Packages

GUIDE-AI is organised into seven work packages covering the full lifecycle from clinical requirements to real-world validation and dissemination.

WP01

Project Management & Coordination

Overall project governance, risk management, financial reporting, and quality assurance across all consortium partners.

WP02

People Centred Innovation & Interestholder Engagement

Ensuring the integration of GUIDE-AI into clinical workflows by understanding interestholder needs.

WP03

GUIDE-AI Navigator Development

Delivering the technical development of the GUIDE-AI system.

WP04

Benchmarking Performance & Robustness

Benchmarking, red-teaming and exploring generalizability and scalability.

WP05

Real-World Validation & Multicentric Insights

A rigorous, inclusive and transparent prospective study.

WP06

Navigating Risks, Ethics & Compliance

Ensuring the project meets the highest standards of ethical integrity, legal and regulatory compliance.

Latest News

All news →
Project Update

GUIDE-AI Consortium Convenes for Kick-Off Meeting

Representatives from all partner institutions gathered to officially launch GUIDE-AI, align on project objectives, and establish working group structures.

Announcement

GUIDE-AI Awarded €4.8 Million by the Innovative Health Initiative

The IHI JU has awarded GUIDE-AI funding under Call 9, recognising the project's potential to transform navigational AI for guideline adherence across Europe.

Overview

Introducing GUIDE-AI: Intelligent Guideline Navigators for Chronic Disease

An introduction to the GUIDE-AI project — its scientific vision, the clinical problems it addresses, and how the consortium will tackle them over four years.

19 Partners · 8 Countries

Academic institutions, clinical centres, industry, and patient organisations across Europe and beyond.

🇩🇪 Germany
🇮🇹 Italy
🇳🇱 Netherlands
🇪🇪 Estonia
🇸🇮 Slovenia
🇬🇧 United Kingdom
🇨🇭 Switzerland
🇮🇱 Israel
View all partners →