AIR Lund Chest pain - More efficient and equal emergency care with advanced medical decision support tools

Estimates suggest that patients seeking emergency care due to chest pain lead to avoidable costs of the order of 100 MSEK annually in Sweden. Clinical investigations of these patients aim at ruling in or out acute coronary syndrome (myocardial infarction or unstable angina pectoris).

The overarching aim of our project is to develop a medical decision support tool in order to improve efficiency and increase equity in the treatment of patients with chest pain while maintaining highest security. The proposed technical solution is based on artificial intelligence and deep learning methods in order to collect, systematize and categorize patient data from medical journals and extensive health care registers.

A new and innovative prediction model will be developed based on extensive data regarding e.g. symptoms, biomarkers and disease history in order to improve diagnostic accuracy and treatment. The solution will be directly beneficial for chest pain patients due to the improved accuracy in the diagnoses, but also through decreased anxiety when calming information can be given earlier and with increased confidence. The solution will also be indirectly beneficial for other patient groups through shorter lead times and reallocation of resources. The proposed solution is novel from an international perspective as it combines data extracted from medical records with register data, a possibility that is to a large extent lacking internationally. Our pioneer work has great potential in the long run to be used also for other health seeking causes besides chest pain and also for visits to other health care services, e.g. in primary health care.