Automated Detection of Venous Gas Emboli on Ultrasound Recordings (FY2019-000274-AS)

It is well established that bubbles are a necessary, but not sufficient, condition for the development of decompression sickness (DCS). In scuba divers, these bubbles are assessed as circulating venous gas emboli (VGE) in blood post-dive using ultrasound, with either subclavian Doppler ultrasound or trans-thoracic echocardiography. Recordings are evaluated in real-time and/or recorded for future (re-)analysis by expert human raters. Raters assign a semi-quantitative and non-linear severity grade and DCS risk statistically increases with increasing grade. Depending on the type of scale and recording used, these grading methods can be difficult and lengthy to learn, and inter-rater reliability can be low in less trained raters. Conversely, automating ultrasonic VGE assessment would allow for standardized, continuous, real-time decompression stress assessment, of particular relevance to the Navy for optimization of decompression procedures.The robustness of such a program, however, ultimately relies on a large and representative dataset of accurately labeled data for training and validation. Computer-automated VGE detection has been previously proposed. However, we believe that this proposal incorporates key aspects that will take this encouraging technology from a proof-of-principle demonstration to a usablereal-time system. Notably, we will ensure that our training data is both spatially and temporally labeled in both echocardiography and Doppler recordings (as opposed to only using an overall grade for each recording). This allows for more sophisticated algorithms to be used, but also for the final programs performance to be precisely assessed.