How AI can benefit healthcare

At the Open Day on 29 March, researchers talked about how artificial intelligence can be used to improve patient care and make the health services more efficient.

AI was the topic on the webinar — AI holds great promise for healthcare, but we need to understand both its benefits and limitations

This event series is hosted by our centre in collaboration with the Polish Ministry of Health. It is part of the Norway Grants-funded project «Tackling social inequalities in health with the use of e-health and telemedicine solutions.»

Professor Marcin Moniuszko from the Medical University of Bialystok, Poland, said that the quality and standardisation of health data is important. By collecting many types of patient data, including genome analysis, clinicians can find out what causes certain diseases and improve preventive care and personalised medicine. AI tools can be built on the collected genomic data. The crucial role of biobanks as the foundation for building AI tools, is often underestimated.

Senior researcher Maryam Tayefi at E-health Research presented highlights from an AI report she and her colleagues have written. Why is it so difficult to implement AI in the health services? Several obstacles exist. For example, lack of trust in AI and a lack of knowledge, organisational and regulatory challenges, and problems with infrastructure and data quality. Some recommendations to improve adoption are spreading knowledge about AI and help organisations to be ready. Financial incentives can boost uptake, as can guidelines, standardisation and public-private partnerships.

Senior researcher Phuong Dinh Ngo from E-health Research talked about the IM-PACT method – how multimorbid patients with impactable risk profiles can be identified by using AI. Health services could be better tailored to accomodate individuals with complex, long-term needs. Using predictive tools makes it possible to identify people in the risk groups who may benefit from preventive measures. The aim is to improve care and reduce costs for elderly patients with complex needs.

Gro-Hilde Severinsen, senior researcher at E-health Research, shared information on a project where they follow the procurement and implementation of AI software in a radiology department in a Norwegian health trust. The AI will help doctors make the diagnosis. So far, CE-marked AI solutions are only slowly adopted in healthcare. The researchers explore expectations and organisational processes, how the AI system is integrated, and study the impacts of the system, including unanticipated consequences.

The recorded presentations can be viewed here. (YouTube playlist: Open Days)