What challenges did Erasmus MC face before AI implementation?

Need to guarantee ethical, safe, and responsible use of AI in clinical practice. Fragmented data and isolated AI deployments across healthcare providers hinder broad benefit realization. Lack of standardized processes to register, monitor, and benchmark AI in clinical operations. Concerns around AI algorithm bias, efficacy, and patient privacy. Ensuring low-resource healthcare settings can adopt responsible AI practices. Sustaining public and clinician trust in digital health innovation.

What was the impact of the AI solution?

Improved quality, safety, and accountability for AI tools in patient care. Enables safe AI use even in low-resource healthcare settings. Federated registry and best practice sharing accelerate adoption of proven, unbiased AI solutions. Supports EU-level compliance and collaboration, safeguarding patient data.

Source

MicrosoftExpandedScaled productionEvidence: Medium65/100

European Hospitals Collaborate to Embed Responsible AI in Healthcare Delivery

Use case typeAI development platformUpdated Jun 13, 2026

A consortium of leading European healthcare institutions has joined the Trustworthy & Responsible AI Network (TRAIN) to operationalize responsible AI at scale across clinical and operational healthcare settings. Organizations including Erasmus MC (Netherlands), Sahlgrenska University Hospital and Skåne University Hospital (Sweden), HUS Helsinki University Hospital (Finland), Universita Vita-Salute San Raffaele (Italy), University Medical Center Utrecht (Netherlands), and Foundation 29 (ES) collaborate with Microsoft as technology partner. The initiative aims to share best practices, provide robust technology-based guardrails, and improve the safety, efficacy, and trustworthiness of AI algorithms in clinical operations. Outcomes-focused measurement tools and federated AI outcome registries foster impartial evaluation, and privacy-enhancing technologies protect patient data. This cross-country network aims to enable equitable AI benefits while maintaining data privacy and compliance with EU healthcare regulations.

Organization: Erasmus MC
Industry: Healthcare
Location: Sweden
Published: June 2024

Why do we believe this deployment?Customer identity, provider attribution, maturity, and source checks

Customer: Erasmus MC, Sahlgrenska University Hospital, Skåne University Hospital, University Medical Center Utrecht, Universita Vita-Salute San Raffaele, Foundation 29, HUS Helsinki University Hospital
Provider: Microsoft
Maturity: Scaled Production
Linked source: news.microsoft.com

Customer identity supportedSource describes one deploymentMaturity supported

Primary read

Use case focus

Showing 3 of 3

1AI evaluation and monitoring
2Federated outcomes registry
3Responsible AI compliance

Need to guarantee ethical, safe, and responsible use of AI in clinical practice.
Fragmented data and isolated AI deployments across healthcare providers hinder broad benefit realization.
Lack of standardized processes to register, monitor, and benchmark AI in clinical operations.
Concerns around AI algorithm bias, efficacy, and patient privacy.
Ensuring low-resource healthcare settings can adopt responsible AI practices.
Sustaining public and clinician trust in digital health innovation.

Formation of European TRAIN, a collaborative network of hospitals and non-profits with Microsoft as enabling technology partner.
Deployment of technology-based guardrails for AI applications.
Development of secure, federated AI outcomes registries to evaluate and share real-world outcomes among members.
Sharing of best practices for AI deployment, including privacy-preserving collaboration and bias detection technologies.
Provision of measurement tools to study AI effectiveness in different healthcare subpopulations.
Open membership to healthcare organizations across Europe.

Technologies

Azure AI

Improved quality, safety, and accountability for AI tools in patient care.
Enables safe AI use even in low-resource healthcare settings.
Federated registry and best practice sharing accelerate adoption of proven, unbiased AI solutions.
Supports EU-level compliance and collaboration, safeguarding patient data.

Architecture

Member hospitals implement responsible AI guardrails and outcome measurement tools; federated registries securely track and analyze clinical AI deployments. Microsoft provides cloud technology and privacy-preserving AI capabilities to connect, monitor, and benchmark registered algorithms across institutions, while best practices and outcome data are shared without transferring raw patient data.

Sources & evidence1

Evidence: Medium65/100Evidence strength