12.12.2024

AICCELERATE project brought important lessons on scaling AI into hospital care processes

HUS Helsinki University Hospital formed a consortium around the idea they developed, consisting of five European university hospitals, one university and ten technology companies. The consortium received EUR 10 million in funding from the European Union's Horizon2020 programme. Out of over 80 applications, only four received funding, of which the AICCELERATE project (Smart Hospital Care Pathway Engine) coordinated by HUS started at the beginning of 2021 and concluded in April 2024. The AICCELERATE joint project developed digital solutions using artificial intelligence to scale hospital care processes.

The existing digital solutions of the partners involved in the project were utilized to build the Smart Hospital Care Pathway Engine (SHCP) tool. This tool implements AI models and robotics to improve the efficiency of healthcare and the quality of patient care. The technology developed in the AICCELERATE project was tested in three pilots, from which feedback and learning experiences were gathered to develop scalable AI. The key goal of the project was to enhance healthcare efficiency and improve the quality of patient care.

 

In the pilot led by HUS, remote monitoring of the patient's condition and symptoms, analytics and robot-assisted drug dispensing were integrated into the digital care pathway for Parkinson's disease. As the monitoring of a chronic disease moves more into the patient's own living environment, smart remote monitoring enables healthcare resources to be allocated to those phases of the disease where hospital stays or doctor visits are needed. The project's outcomes included open-source components for data collection from various sources and data harmonization into a common FHIR data model, which can be used in building AI-assisted patient pathways. Additionally, tools and interfaces for training and testing AI models were developed. The remote monitoring pilot for Parkinson's disease implemented a proof-of-concept, which continues as a clinical study to gather additional evidence needed for clinical use.

 

"Simply developing datasets and machine learning is not enough to bring AI into hospitals. A comprehensive IT infrastructure is needed to collect and store health data, develop algorithms, and present results in a form that healthcare professionals and patients can understand. The AICCELERATE project focused on solving this overall challenge," notes Laura Mäkitie, the neurologist from HUS who led the pilot.

 

In the pilot led by Oulu University Hospital, AI-assisted planning for emergency and non-urgent surgeries was developed to optimize the operation of operating rooms from both the patient and staff resource perspectives.

 

The third pilot, conducted in children's hospitals in Barcelona and Rome, sought solutions to improve the care pathway and home care of severely ill pediatric patients using interactive robotics and AI-assisted risk assessment tools.

 

Technology Officer Pekka Kahri at HUS concludes: "The rapid development of technology requires a deep understanding of the opportunities and challenges brought by AI from both hospitals and technical partners. The AICCELERATE project brought together experts from various fields to consider the use of AI in healthcare not only from a technical perspective but also from ethical and legal viewpoints."

 

More information:

www.aiccelerate.eu

https://www.youtube.com/@AICCELERATE_EU

 

AICCELERATE consortium partners: HUS Helsinki University Hospital, Oulu University Hospital, Ospedale Pediatrico Bambino Gesù (Italy), Barcelona Children's Hospital (Spain), University Hospital Università degli Studi di Padova (Italia), Erasmus University Rotterdam (the Netherlands), RTO Fundació Eurecat (Spain), TICBioMed (Spain) and companies Chino (Italy), Symptoma (Austria), Nuromedia (Germany), SRDC (Türkiye), Evondos (Finland), NeuroPath (Belgium), NEC Laboratories Europe (Germany) and Innofactor (Finland).