EARS Oncology AI Recommendation and Decision Support System

GLINTT's EARS Oncology AI Recommendation and Decision Support System is a secure and explainable AI-based solution designed to support personalised cancer care. The system combines AI-based medical imaging analysis for tumour detection and evolution monitoring with patient data processing, sentiment analysis and GeoAI/Location Intelligence. These multimodal data sources are integrated to generate personalised and explainable recommendations on treatments, therapies and patient follow-up. The solution is designed for integration within the EARS federated framework, enabling privacy-preserving use of sensitive healthcare data while supporting healthcare professionals with more accurate, contextualised and trustworthy clinical decision support.

The solution combines AI-based processing of tumour scans with heterogeneous patient information to support personalised oncology recommendations. Computer Vision models analyse medical images to identify relevant regions and monitor changes in tumour evolution. NLP and sentiment-analysis components process patient-generated and unstructured information, while GeoAI enriches the patient context with location-dependent information. A multimodal recommendation layer combines these sources to support treatment and therapy recommendations. Explainability mechanisms provide interpretable information to healthcare professionals, while integration with the EARS federated framework enables privacy-preserving model development and secure exploitation of distributed healthcare data.

Integration requires secure access to authorised clinical information systems and appropriate interfaces for medical images, patient data and contextual information. Interoperability with hospital systems must rely on controlled APIs and standardised data models where applicable. Sensitive health information must remain protected according to applicable data-protection requirements, including GDPR. Federated-learning integration should avoid unnecessary transfer of raw patient data, while authentication, authorisation, traceability and audit mechanisms must be implemented. Clinical recommendations are intended as decision support and should remain subject to validation and interpretation by qualified healthcare professionals

Hospitals, oncology departments, healthcare networks, clinics, diagnostic centres and healthcare organisations seeking AI-assisted oncology decision support. Secondary users include oncologists, radiologists, other healthcare professionals, home carers and organisations providing personalised cancer follow-up services. GLINTT's existing healthcare network provides a potential route for integration and commercial exploitation.

euse and deployment are subject to GLINTT's commercial licensing conditions, applicable intellectual-property agreements and EARS consortium exploitation rules. Deployment with real patient data requires compliance with GDPR, healthcare-sector security requirements, institutional governance procedures and, where applicable, validation and regulatory requirements for clinical software and AI-based medical solutions. Models and components should be adapted and validated for the target clinical environment before operational use.