AI-enhanced prognostication in cardiac resynchronization therapy using displacement encoding with stimulated echoes (DENSE) MRI; Data Privacy and Security in AI-driven Healthcare 8 views

How can machine learning systems improve patient care while preserving privacy, interpretability, and trust? In healthcare, these goals converge on the question of legitimacy, or whether algorithmic tools can be accepted as reliable and transparent partners in clinical decision-making. The union of the two projects therefore concerns the alignment between data-driven innovation and the social systems that govern its ethical use. The technical project developed a deep learning framework to predict patient response to cardiac resynchronization therapy (CRT) for heart failure using displacement encoding with stimulated echoes (DENSE) cardiac MRI. By training a three-dimensional convolutional autoencoder on voxel-level myocardial motion data, the study sought to identify mechanical patterns that distinguish responders from non-responders. The model achieved an area under the receiver operating characteristic curve of about 0.7, showing that DENSE MRI encodes detailed biomechanical information capable of guiding precision CRT planning. The sociotechnical research examined how patients, clinicians, developers, and regulators negotiate privacy, security, and trust in AI-enabled healthcare. It found that data protection measures and regulatory frameworks succeed only when they align with patient expectations and clinical workflow. Technical safeguards such as federated learning or differential privacy gain meaning through governance structures that ensure transparency and accountability. The analysis concludes that sustainable AI in medicine depends on this convergence: legitimacy arises when design architecture, institutional oversight, and stakeholder values reinforce one another, allowing innovation to advance without compromising human dignity or autonomy.

BS (Bachelor of Science)

DENSE MRI; Heart Failure with Reduced Ejection Fraction (HFrEF); Cardiac Resychronization Therapy (CRT); Deep Learning; Artificial Intelligence (AI)

School of Engineering and Applied Science Bachelor of Science in Biomedical Engineering Technical Advisor(s): Dr. Frederick Epstein, Dr. Sona Ghadimi STS Advisor(s): Peter Norton, Travis Elliott Technical Team Members: Sean Mullaghy

English

All rights reserved by the author (no additional license for public reuse)

2025-12-11