Publications Repository - Gdańsk University of Technology

ML algorithms are very effective tools for medical data analyzing, especially at image recognition. Although they cannot be considered as a stand-alone diagnostic tool, because it is a black-box, it can certainly be a medical support that minimize negative effect of human-factors. In high-risk domains, not only the correct diagnosis is important, but also the reasoning behind it. Therefore, it is important to focus on trustworthiness which is a concept that includes fairness, data security, ethics, privacy, and the ability to explain model decisions, either post-hoc or during the development. One of the interesting examples of a medical applications is automatic SVD diagnostics. A complete diagnosis of this disease requires a fusion of results for different lesions. This paper presents preliminary results related to the automatic recognition of SVD, more specifically the detection of CMB and WMH. The results achieved are presented in the context of trustworthy AI-based systems.

Authors

• mgr inż. Maria Ferlin link open in new tab ,
• Zuzanna Klawikowska link open in new tab ,
• mgr Julia Niemierko,
• mgr Małgorzata Grzywińska,
• mgr inż. Arkadiusz Kwasigroch link open in new tab ,
• prof. dr hab. Edyta Szurowska,
• dr hab. inż. Michał Grochowski link open in new tab