Quantitative estimation of blood loss volume using laser speckle flow index signals from a wireless wearable device (Conference Presentation)

Postpartum hemorrhage (PPH) is the leading and most preventable cause of maternal mortality. There is an urgent need for a technology that can monitor visible and occult blood loss in real-time to detect PPH in early stages. A compact, low-cost (<$150), wrist worn wearable device that continuously monitors perfusion via laser speckle flow index (LSFI) was previously developed to detect blood loss. The device was used to captures pulsatile LSFI waveforms from swine subjects during blood removal. Collected waveforms were used to extract a set of features that describe the waveform shape, and the features were used to train a multivariate regression model to predict blood loss volume. The average absolute error was 137 mL, where the correlation between true and predicted blood loss volume was r=0.86. When hemoglobin concentration was also included as a feature in the model, the average error was reduced to 60.35 mL and r=0.93.