Contact person
Pernille Just Vinholt, Associate Professor, Consultant, MD, PhD
pernille.vinholt@rsyd.dk; https://portal.findresearcher.sdu.dk/en/persons/pvinholt
Research aim
To explore and support clinical decisions. In this regard we disclose relevant aspects regarding clinical decisions. We develop and investigate algorithms and tools, which can help with complex decisions made in screening, diagnosis and monitoring of patients.
Improved decision support can benefit diagnostics and monitoring, leading to increased patient safety and better care, e.g. by suggesting an intervention. Further, decision support may lead to cost reduction, e.g. physician time for reviewing medical records is saved - time that can be used with the patient. Moreover, the systems can provide health statistics and optimize planning and resource consumption in the health care system.
Project examples
“The Intelligent Patient Medical Record” is a portfolio of projects. Based on natural language processing, we develop algorithms that finds critical information in the medical record and presents it to the physician and/or patient for decision support. For more information on the team and projects, please consult this page: https://ipj.nu/
Porphyria diagnostic. The development and evaluation of a digital decision-support tool for porphyria diagnostics based on clinical-biochemistry results.
Collaborators
-
The Mærsk McKinney Møller Institute, SDU
Thiusius Rajeeth Savarimuthu, Professor of Medical Robotics, PhD
-
Department of Medicine, OUH Svendborg Hospital,
Søren Andreas Just, Specialist Consultant, PhD
-
Dataenheden OUH, The Data Unit at Odense University Hospital
-
Rasmus Søgaard Hansen PhD Student, MD
-
Martin Sundahl Laursen PhD Student, MSc, Engineer
-
Jannik Skyttegaard Pedersen PhD Student, MSc, Engineer
Publication example
Pedersen JS, Laursen MS, Rajeeth Savarimuthu T, Hansen RS, Alnor AB, Bjerre KV, Kjær IM, Gils C, Thorsen AF, Andersen ES, Nielsen CB, Andersen LC, Just SA, Vinholt PJ. Deep learning detects and visualizes bleeding events in electronic health records. Res Pract Thromb Haemost. 2021 May 5;5(4):e12505. doi: 10.1002/rth2.12505. PMID: 34013150