Beyond Initial Design: AI Advancement of Portable Ultrasound Technology for Brain Perfusion Imaging

This paper presents a case study of SONAS®, a novel medical device utilizing low-power, low-frequency ultrasound with a microbubble contrast agent to visualize brain blood perfusion. By detecting nonlinear harmonic responses from microbubble oscillations, this portable, cost-effective device enables cerebrovascular anomaly detection. Regulatory milestones demonstrated strong correlation between SONAS harmonic signals (analyzed using heuristic algorithms) and conventional 3-D MRI/CT imaging. Building on AI’s success in traditional imaging analysis, we propose enhancing SONAS through a progression from basic machine learning to sophisticated diagnostic assistance that integrates expert knowledge with generative AI. Because this device operates with limited network connectivity, we also demonstrate the value and importance of a software architecture that decouples model training from inference, enabling efficient on-device deployment. The augmentation of an existing system like this represents an opportunity for AI to amplify capabilities of medical technologies beyond their initial design parameters. We demonstrate the technology steps in this evolution from basic diagnostic tool to AI-enhanced clinical decision support system. These steps can significantly improve diagnostic accuracy, workflow, and outcomes—establishing that AI integration is both advantageous and inevitable in contemporary medical technology adoption.