STTR Phase I: Applying real-time data streams to predict operating room resource allocation with neural networks

The broader impact/commercial potential of this Small Business Technology Transfer (STTR) Phase I project is to reduce the cost of surgical operations. Instrument tracking enables hospitals to optimize the supply chain, with a potential annual benefit to the US healthcare system of $8.5 B. Predictive scheduling can save $500k per operating room (OR) by closing gap times between procedures. Similarly, instrument prediction assistance can save an OR an estimated $14 per minute. The project will gather procedure and tool data from the operating room and apply artificial intelligence to optimize OR processes. This project has the potential to improve the overall function of the surgical team by anticipating surgical instrument needs. This Small Business Technology Transfer (STTR) Phase I project advances the fields of medicine and artificial intelligence by leveraging intraoperative data gathered by surgical instrument tracking. This unique data stream offers one of the first quantitative windows into a surgical operation. The objective of this project is to create computational tools to improve operating room scheduling and instrument supply, and test them with real clinical data. Transformer networks, commonly used in natural language processing tasks, will be adapted for this application and leveraged as an autoregressive tool to predict parameters of interest. The system will generate data regarding variations among surgeons, procedures, and patients. 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.