Project Summary This work will develop an innovative infusion device that leverages new electronics and sensor capabilities to produce a medical device that improves intravenous (IV) infusion delivery. IV infusions are among most common healthcare procedures administered worldwide; an estimated 80% of hospital patients receive some IV therapy. Infusions are also increasingly delivered outside of hospital settings, in homes, long-term care facilities, surge settings and other alternate site care (ASC) environments. ASC infusions, in particular, are a critical healthcare trend because they increase access to care by eliminating obstacles such as patient transportation, pathogen exposure for the immunocompromised, and inpatient costs. While infusion is critical infrastructure for healthcare delivery, the technologies to deliver it have not kept up with changing needs in the sector. This proposal builds on a clinically-validated infusion monitoring system, the DripAssist Infusion Rate Monitor, to create a closed-loop, digitally-connected infusion platform capable of administering even potentially complex drug regimes in both acute and ASC settings. This technology-enabled gravity infusion delivery system with automatic flow adjustment, what we call DripAssist Control, will use low cost sensors to accurately measure the volume of fluid in a drop. This is an experimental study that will collect data in silico across a predetermined set of dependent and independent variables. Large data sets will be collected to which both conventional statistical analysis and machine learning methods will be applied. Our research will utilize rich information confirmed to be present in outputs of an existing preliminary benchtop system to establish new data processing methods for measuring drop volume. Once refined, these methods will be optimized into a closed-loop flow control system, which will be developed into a form and fit device and tested within clinical settings with realistic constraints and variability. We have three Aims in this work. Aim 1: establishing two benchtop systems that can broadly identify the drip volume rating of an infusion set through a novel detection process. Aim 2: optimize our sensing and actuating algorithms to clinically acceptable precision and accuracy. Aim 3: perform closed-loop control system integration and preclinical development work to attain a working prototype device that can be tested for usability and established regulatory standards. We believe our work can create a bold shift in clinical practice by introducing novel instrumentation and sensors and producing a simpler infusion delivery device that matches the precision and accuracy of expensive pump platforms.