Collaborative digital health innovations for clinical needs

CleverHealth Network‘s multidisciplinary collaboration between Helsinki University Hospital, companies, and research institutions generates new digital solutions for identified clinical needs by harnessing health data, AI, and machine learning, resulting in impactful treatments that improve the quality of life for patients and raise the efficiency of health care processes.

Results: New digital solutions for acute leukemia, gestational diabetes and dialysis

Three research and development projects have now reached the commercialisation stage:

A clinical tumour board application has been developed to automate cancer diagnostics and ease the election to suitable treatment for patients suffering from acute leukemia. The app, developed through collaboration of HUS, TietoEvry, Productivity Leap and Hematoscope, will be used in weekly oncological meetings as well as in research.

An automated end-to-end service solution to provide a safe home-based dialysis for dialysis patients, thus easing their daily life, has been created by HUS, Aalto University, University of Helsinki and Gillie. Medical staff has access to a user interface providing real-time data.

Mobile application to treat gestational diabetes improves the glycaemic control for expecting mothers by bringing real-time data on glycaemic levels and the lifestyle choices of mothers-to-be. The app will soon be piloted on a wider scale in maternity and child health clinics. The project partners are HUS, Aalto University, University of Helsinki Fujitsu and Elisa.

AI algorithm identifying life-threatening cerebrovascular disorder

More promising results by CleverHealth Network comes from AI Head Analysis project, a collaboration of HUS, CGI and Planmeca developing an AI-based tool set for analysing head computed tomography (CT) scans to identify and localise a life-threatening cerebrovascular disorder, subarachnoid hemorrhage.

New research results were recently published in the prestigious Neurology journal, proving that the accuracy of the created algorithm is excellent and has the potential to significantly improve the timely diagnosis of patients. In addition, the AI model has been made accessible to the entire research community.