ESOC 2024 | Ensuring medical data security in machine learning models

Jonathan Wenstrup, MD, PhD Candidate, Herlev Hospital, Herlev, Denmark, shares his thoughts on the ethical and regulatory challenges surrounding medical data security. His work aims to develop AI tools to aid emergency call handlers in identifying stroke, with early results demonstrating the value of this technology as a supportive decision-making tool. Dr Wenstrup emphasizes that, with a secure infrastructure, the ethical use of data for developing AI tools is achievable, as only the machine sees the data. He acknowledges the difficulties in obtaining permits and data from registries but stresses that when handled properly, there are no significant ethical issues. Dr Wenstrup discusses the need to adapt machine learning models to specific geographic regions, and thus the importance of committing resources to ethically sound data collection and ongoing system improvement through clinical use. This interview took place at the 10th European Stroke Organisation Conference (ESOC) 2024 in Basel, Switzerland.

These works are owned by Magdalen Medical Publishing (MMP) and are protected by copyright laws and treaties around the world. All rights are reserved.