A deep learning system using in-ear audio from commodity earbuds estimated stroke volume with a mean absolute error of 5.24 ml and a Pearson correlation of 0.94 compared to a clinical device.
Observational (n=23)
No
Does a deep learning system using in-ear audio from earbuds accurately estimate stroke volume compared to clinically validated devices in healthy participants?
In-ear audio sensing using commodity earbuds combined with deep learning can accurately estimate cardiac stroke volume, offering a potential scalable method for out-of-clinic cardiovascular monitoring.
Effect estimate: r=0.94
p-value: p=<0.001
Abstract Stroke volume, the volume of blood ejected by the left ventricle during a contraction, is a key metric of cardiovascular health. Currently, stroke volume is measured in clinic with specialised equipment. While purpose-made wearables exist to measure stroke volume, no solution relies solely on commodity devices. We present a deep learning system for stroke volume estimation from in-ear audio of earbuds. We combine generative self-supervised/transfer learning, a transformer-based autoencoder, to predict average stroke volume in unseen subjects. With data from 23 healthy participants, we compare our estimations to clinically validated device estimations. We achieve a mean absolute error of 5.24 ml, a Pearson correlation of r=0.94 between average predicted stroke volume and average true stroke volume, and a Percentage Error in the limits of agreement of 11.05% (within clinical range for stroke volume measurement devices). These findings open the doors to longitudinal, scalable and affordable cardiovascular measurement out of clinic.
Published July 17, 2026; high novelty in wearable tech; discussed in cardiology social media for non-invasive monitoring potential.
Butkow et al. (Fri,) conducted a observational in Healthy (n=23). In-ear audio sensing (CardiAural) vs. Finapres NOVA (clinically validated device) was evaluated on Average stroke volume estimation accuracy (r=0.94, p=<0.001). A deep learning system using in-ear audio from commodity earbuds estimated stroke volume with a mean absolute error of 5.24 ml and a Pearson correlation of 0.94 compared to a clinical device.