STTR Phase I: Development of a Machine Learning Platform to Predict Surgical Complications

The broader impact/commercial potential of this Small Business Technology Transfer (STTR) Phase I project is to create a personalized, precision-based practice paradigm for surgery that improves surgical outcomes and reduces the cost of healthcare. This paradigm utilizes individual patient characteristics with machine-learning algorithms to accurately predict the risk of post-surgical complications. Additionally, it offers the possibility of enacting impactful interventions among high-risk patients, while reducing unnecessary therapies among low-risk patients, thereby improving surgical outcomes, maximizing the efficiency of healthcare, and minimizing cost. In a value-based care model, this paradigm aligns the goals of health systems, surgeons, and patients. The proposed project combines individual patient data with machine-learning algorithms to effectively predict surgical complication risk and improve surgical clinical outcomes. Currently, 13% of 50 million surgical procedures performed in the United States annually result in a surgical complication, half which are potentially avoidable. A primary cause of avoidable complications include significant variable in risk assessment and standardized preventative practice. Therefore, the principal objective of this proposal is to develop machine-learning predictive models of various surgical complications (wound, cardiac, respiratory, renal, etc.), which provides an objective risk assessment for surgeons. Additionally, this risk assessment platform will allow stratification of patients into high- vs. low-risk categories and link patients with risk-appropriate preventative interventions at the point-of-care.