What challenges did European Institute of Oncology face before AI implementation?

Unstructured clinical medical records and reports are difficult to analyze quickly for research and clinical decision-making. Manual processing of large volumes of medical records is slow and inefficient, hindering timely insights. Ensuring data privacy and compliance with GDPR while making data accessible for research was essential.

What was the impact of the AI solution?

Medical report classification speed improved 300 times over manual methods, enabling faster research and clinical analysis. Improved data accuracy and consistency through standardized, anonymized datasets. Non-technical users gained self-service access to data insights via dashboards, supporting better decision-making. Automated KPI extraction saved significant time and reduced workload for research and certification processes.

Source

GCPEvidence: Medium50/100

IEO-Monzino using Vertex AI NLP models for large-scale clinical data structuring and research acceleration

Use case typeClinical documentationUpdated Jun 13, 2026

The European Institute of Oncology (IEO) and Monzino Cardiology Center in Milan, Italy, serving 1.7 million patients, developed a Clinical Data Platform (CDP) to convert vast unstructured clinical records into structured, anonymized data to support medical research and clinical analysis. They built proprietary NLP models using Google Cloud's Vertex AI to classify and standardize 76,000+ medical reports, achieving 300x faster processing than manual methods, enabling rapid data-driven research and decision-making. The solution includes dashboards on Looker Studio for non-technical medical staff to access data insights easily and automated KPI extraction for certifications, improving reliability and saving time for clinicians and researchers.

Organization: European Institute of Oncology
Industry: Healthcare
Location: Italy
Published: May 2024

Reported outcomes

300x

timeTime & speed

Strategic outcomes

New product / capabilityConverted clinical records into structured anonymized dataSpeed & agilityEnabled much faster medical report processingBetter decisions & insightGave non-technical staff self-service data accessRisk & complianceImproved privacy-safe data accessibility for research

Why do we believe this?Outcome claims, sources, and evidence checks

Normalized claim

Time: 300 x decrease

Google Cloud Customer StoriesMay 9, 2024Customer storyInferred claimMedium evidence strength

Medical report classification speed improved 300 times over manual methods, enabling faster research and clinical analysis.

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

Customer: European Institute of Oncology, Monzino Cardiology Center
Provider: GCP
Maturity: Unknown
Linked source: Google Cloud Customer Stories

No explicit deployment-stage evidence found.

Customer identity supportedSource describes one deploymentMaturity evidence evaluated

Primary read

Use case focus

Showing 3 of 3

1Natural Language Processing
2Data Structuring
3Clinical Research Acceleration

Unstructured clinical medical records and reports are difficult to analyze quickly for research and clinical decision-making.
Manual processing of large volumes of medical records is slow and inefficient, hindering timely insights.
Ensuring data privacy and compliance with GDPR while making data accessible for research was essential.

Developed and trained proprietary Natural Language Processing (NLP) and large language models on Google Cloud's Vertex AI to standardize and anonymize clinical data.
Built interactive dashboards using Looker Studio to allow easy access and filtering of data for medical staff and researchers without technical expertise.
Automated the extraction and processing of key clinical performance indicators (KPIs) for certifications, reducing manual workload.

Technologies

Vertex AI Looker Studio Google Cloud

Medical report classification speed improved 300 times over manual methods, enabling faster research and clinical analysis.
Improved data accuracy and consistency through standardized, anonymized datasets.
Non-technical users gained self-service access to data insights via dashboards, supporting better decision-making.
Automated KPI extraction saved significant time and reduced workload for research and certification processes.

Architecture

The architecture involves a Clinical Data Platform (CDP) built on Google Cloud, leveraging Vertex AI for NLP and large language model training and inference. Looker Studio dashboards provide user-friendly data access and filtering. The platform ensures data anonymization and GDPR compliance.

Sources & evidence1

Evidence: Medium50/100Evidence strength