Project Grant 2551975

The National Science Foundation Division of Information and Intelligent Systems awarded the University of Notre Dame DU Lac $349,281 on August 1, 2026, under the Computer and Information Science and Engineering program (CFDA 47.070) for collaborative research on algorithm-design-technology co-optimization for next-generation self-evolving implantable devices.

The project develops a cross-layer co-design framework that jointly optimizes applications, learning algorithms, hardware architectures, and system-level operation to enable implantable medical devices—including pacemakers, implantable cardioverter defibrillators, deep brain stimulators, cochlear implants, and sleep apnea implants—to automatically adapt to individual patients' changing physiological needs while operating within strict constraints on size, power, and battery life. The research is organized into five integrated tasks across two phases: the first phase explores device requirements, develops lightweight state-space models for physiological signal analysis, introduces uncertainty-aware safety mechanisms, and designs energy-efficient in-memory-computing architectures; the second phase advances adaptive model updating, uncertainty-triggered evolution, hardware reconfiguration, and algorithm-architecture co-evolution. The project investigates structured state-space models, adaptive quantization, Bayesian confidence estimation, processing-in-memory architectures, fault tolerance techniques, and reinforcement learning–based optimization.

Performance occurs at Notre Dame, Indiana, with an ultimate completion date of July 31, 2030.