Ethical Framework for Responsible Foundational Models in Medical Imaging

Debesh Jha ¹ Gorkem Durak ¹ Abhijit Das ¹ Jasmer Sanjotra ¹ Onkar Susladkar ¹ Suramyaa Sarkar ¹ Ashish Rauniyar ¹ Nikhil Kumar Tomar ¹ Linkai Peng ¹ Sirui Li ¹ Koushik Biswas ¹ Ertugrul Aktas ¹ Elif KELES ¹ Matthew Antalek ¹ Zheyuan Zhang ¹ Bin Wang ¹ Xin Zhu ^1,2 Hongyi Pan ¹ Deniz Seyithanoglu ^1,3 Alpay Medetalibeyoglu ³ Vanshali Sharma ¹ Vedat Çiçek ¹ Amir Ali Rahsapar ¹ Rutger Hendrix ^1,4 A. Enis Cetin ² Bulent Aydogan ⁵ Mohamed Abazeed ¹ Frank H Miller ¹ Rajesh N Keswani ¹ Hatice Savas ¹ Sachin Jambawalikar ⁶ Daniela P Ladner ¹ Amir Ali Borhani ¹ Concetto Spampinato ⁴ Michael B Wallace ⁷ Ulas Bagci ^1*

• ¹ Northwestern Medicine, Chicago, Illinois, United States
• ² University of Illinois Chicago, Chicago, Illinois, United States
• ³ Faculty of Medicine, Istanbul University, Istanbul, Istanbul, Türkiye
• ⁴ University of Catania, Catania, Sicily, Italy
• ⁵ The University of Chicago, Chicago, Illinois, United States
• ⁶ Columbia University, New York City, New York, United States
• ⁷ Mayo Clinic Florida, Jacksonville, Florida, United States

The emergence of foundational models represents a paradigm shift in medical imaging, offering extraordinary capabilities in disease detection, diagnosis, and treatment planning. These large-scale artificial intelligence systems, trained on extensive multimodal and multi-center datasets, demonstrate remarkable versatility across diverse medical applications. However, their integration into clinical practice presents complex ethical challenges that extend beyond technical performance metrics. This study examines the critical ethical considerations at the intersection 1 Debesh Jha et al.of healthcare and artificial intelligence. Patient data privacy remains a fundamental concern, particularly given these models' requirement for extensive training data and their potential to inadvertently memorize sensitive information. Algorithmic bias poses a significant challenge in healthcare, as historical disparities in medical data collection may perpetuate or exacerbate existing healthcare inequities across demographic groups. The complexity of foundational models presents significant challenges regarding transparency and explainability in medical decisionmaking. We propose a comprehensive ethical framework that addresses these challenges while promoting responsible innovation. This framework emphasizes robust privacy safeguards, systematic bias detection and mitigation strategies, and mechanisms for maintaining meaningful human oversight. By establishing clear guidelines for development and deployment, we aim to harness the transformative potential of foundational models while preserving the fundamental principles of medical ethics and patient-centered care.