Using AI to identify cancer in hematuria patients

The current methods of testing are comprehensive, expensive and may be uncomfortable and risky. The current criteria for the comprehensive tests cause both under- or misdiagnosis.

AIM

By using AI, the projcect will analyse patient journals in order to identify different bleeding symptoms and to link them with the risk of cancer. This knowledge combined with other patient data will make it possible to identify which individuals to examine and thereby diagnose in time.

The project will provide valuable information about risk factors for hematuria and the prognosis for hematuria and urinal tract cancer. It will also provide knowledge on how to treat patients with a bleeding risk.

The aims of the project are:

1. Developing an AI algorithm for diagnosing urinal tract cancer among hematuria patients.
2. Examining risk factors, clinical courses and prognoses for urinal tract cancer and hematuria.
3. Building a model for calculating the risk for urinal tract cancer when a person experiences blood in the urine.
4. Validating a patient tool for self-reporting of bleeding symptoms. 

PARTNERS

Partners on the project are the Mærsk Mc-Kinney Møller Institute at the University of Southern Denmark, and the Department of Clinical Biochemistry and Pharmacology at Odense University Hospital.

The project is a PhD project by Rasmus Søgaard Hansen and is anchored at the Centre for Clinical Artificial Intelligence (CAI-X). Read more about the project at CAI-X.

The PhD project is part of the project the Intelligent Patient Record.

EXTERNAL FUNDING

The project has received financing from the SDU Scholarship and the PhD Fund in the Region of Southern Denmark.