The broader impact/commercial potential of this Small Business Technology Transfer (STTR) Phase I project relates to the development of a novel system capable of measuring medical equipment utilization with high accuracy and scalability. This innovation will arm healthcare technology managers with the insights needed to optimize inventory size and composition according to actual patient needs, thereby saving hospitals an estimated $23.3 billion annually in equipment-related costs, in addition to making possible usage-based predictive maintenance that can effectively prevent dangerous equipment failures. Beyond these core value propositions, comprehensive medical equipment utilization insights may be leveraged to facilitate strategic resource management in public health emergencies, increase energy efficiency of healthcare facilities, and improve regulatory surveillance of emerging equipment safety issues. The results of this project will form the basis for a hardware-enabled service and clear the path towards development of deployable products, clinical pilots, and early sales. Through commercialization under a sustainable business model, the envisioned product will substantially increase the economic competitiveness of US hospitals, which comprises one of the largest sectors of the American economy.The project will also advance the health and welfare of the American public through improved medical device safety and management. _x000D_ _x000D_ This Small Business Technology Transfer (STTR) Phase I project will establish technical and commercial feasibility for an innovative, asset-agnostic, medical equipment utilization tracking system which will integrate state-of-the-art techniques for non-intrusive load monitoring, deep learning, and edge computing in order to overcome previously insurmountable asset monitoring challenges posed by the heterogeneity and churn of hospital equipment inventories. Key technical hurdles to be addressed relate to the capture and characterization of medical equipment electrical load data, real-time translation of this data into accurate usage statistics suitable for hospital decision-making, and distributed implementation of this process through non-invasive sensor modules that are broadly compatible with sundry medical equipment. The proposed research will overcome these hurdles through (i) systematic collection and analysis of power consumption data from a representative group of medical equipment under various operational states, (ii) formulation, training, and validation of adaptive artificial neural networks that predict usage from power data, (iii) construction of a proof-of-concept intelligent sensor module, and (iv) system performance testing in a simulated clinical environment. Through completion of these objectives, this project will advance knowledge in the fields of hospital asset management and industrial Internet-of-Things._x000D_ _x000D_ This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.
