The Uptake of Federated Analytics in Dutch Healthcare: Advancing Breast Cancer Care

Abstract

As healthcare increasingly relies on data-driven insights, federated analytics offer a transformative approach to harnessing sensitive medical data while respecting privacy regulations. This presentation explores the integration of federated analytics into clinical research in Dutch healthcare on a national scale. The focus is on breast cancer care with an emphasis on breast reconstruction outcomes. Utilizing decentralized AI models, this approach enables collaboration across hospitals without centralizing patient data, addressing key barriers such as data security, patient confidentiality, and inter-institutional data sharing restrictions. This framework will be piloted across a network of Dutch hospitals and nationally operating organizations, including IKNL and DICA, improving clinical decision-making for breast reconstruction after cancer treatment. We expect the results of this study to demonstrate the feasibility and impact of federated analytics, highlighting an improved ability to answer novel nation-scale clinical research questions. The study will also reflect on the practical challenges encountered, including infrastructural limitations, algorithm standardization, data harmonization and stakeholder alignment, offering actionable lessons for scaling such innovations. This study underscores the critical role of federated analytics in bridging the gap between AI research and its application in healthcare, paving the way for broader adoption in other medical fields. By fostering collaboration between research institutions and healthcare providers, our work sets a precedent for leveraging AI to address pressing societal challenges in a secure and ethical manner. Our insights aim to inspire researchers, clinicians, managers, and policymakers to embrace federated analytics for enhancing patient outcomes while overcoming the barriers to data sharing in healthcare.

Ähnliche Arbeiten