Background: Driveline infection (DLI) is a critical complication in left ventricular assist device (LVAD) therapy. Following the first total artificial heart (TAH) with HeartMate 6 (HM6) implantation in Asia, performed in Taiwan, China, the complexity of mechanical circulatory support necessitates advanced infection prevention. This study presents a nurse-led Artificial Intelligence (AI) image-tracking prototype designed for asynchronous DLI surveillance. Methods: We developed a digital prototype utilizing an AI-driven framework to analyze driveline exit-site images. The system features image-based measurement calibration to quantify erythema expansion and classify exudate characteristics (e.g., serous, purulent). ICU nurses upload images daily to the backend for continuous, standardized tracking. Results: The prototype enables automated calculation of the erythema radius and detection of abnormal exudation patterns. If clinical thresholds are exceeded, the system issues a “High-Risk Infection Alert”. This proof-of-concept effectively bridges the gap between subjective visual assessment and quantifiable digital tracking. Conclusion: This nurse-led AI prototype for HM6 management enhances early infection detection and optimizes patient safety by transforming subjective clinical observations into objective, continuous data. Future phases will integrate machine learning to validate clinical efficacy further. *Correspondence concerning this article should be addressed to Yih-Sharng Chen, Division of Cardiovascular Surgery, National Taiwan University Hospital, Taipei, Taiwan, China.
Huang et al. (Mon,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: