The Dutch healthcare system is under increasing pressure. The population is ageing, more people are living with multiple conditions at the same time, and staff shortages continue to grow. By 2040, it is estimated that one in four working people would need to work in healthcare to meet demand.
Artificial intelligence can help make healthcare smarter and more efficient. Yet many AI applications developed for healthcare do not make it into daily clinical practice. Trinitas HORIZON aims to bridge that gap by developing AI that is useful, reliable and applicable in the context of general practice.
Complex diagnostic decisions
Trinitas HORIZON focuses on one of the most challenging parts of healthcare: the first diagnostic assessment by the general practitioner. Patients are getting older and often present with multiple, sometimes vague symptoms. At the same time, more diagnostic tests and data sources are becoming available.
This means GPs increasingly need to combine large amounts of information while deciding whether a patient should be referred, whether further testing is needed, or whether a rare disease may be involved. This complexity can lead to repeated tests, referrals that later prove unnecessary, uncertainty for patients and additional pressure on the healthcare system.
First-time-right diagnosis
The consortium will develop an AI-supported platform that brings diagnostic information together and helps GPs make better-informed decisions. The aim is a ‘first-time-right’ diagnosis: choosing the right direction as early as possible, so patients receive clarity and appropriate care sooner.
“Together, we support GPs in answering important diagnostic questions: should someone be referred to hospital, and when is that not necessary? And are we overlooking a rare disease?” says consortium leader Saskia Haitjema of UMC Utrecht. “In this way, we help ensure that patients follow the right diagnostic path more quickly and receive the right treatment sooner.”
AI for uncertain and complex data
Peter Bosman, EI group leader, participates in Trinitas HORIZON as a co-applicant, contributing expertise in artificial intelligence, data science and algorithmic methods. The project requires AI that can deal with uncertainty, incomplete information and multiple data sources, precisely because the general practice setting differs from the hospital environment. Symptoms may still be unclear, and diagnostic reasoning often involves several possible explanations at once.
Within the consortium, Bosman will work with medical, technical and societal partners on AI methods that can support this kind of diagnostic reasoning in practice. The focus is not only on developing advanced technology, but also on making sure that the resulting tools fit the needs and working routines of GPs.
Broad national collaboration
Trinitas HORIZON is led by UMC Utrecht and brings together a broad national network. Dutch university medical centres and their departments of general practice are involved, alongside GP organizations, hospitals, first-line care organizations, patient organizations, universities, technology partners and experts in social, ethical and legal aspects of healthcare innovation.
This broad collaboration is intended to ensure that the technology is developed with the practice of healthcare in mind from the start. GPs, specialists, researchers, data experts and societal partners will work together to design, test and evaluate solutions that can be used in real-world settings.
About the funding
Trinitas HORIZON is co-funded through the NWO-KIC programme line Strategy. Through this programme, NWO supports ten-year public-private partnerships that address major societal challenges. Consortia define their own research themes, provided these align with the Dutch government’s Knowledge and Innovation Agendas.
This article is based on a news item published by UMC Utrecht.
Header photo: Shutterstock
