Biomedical Engineering

This study develops a deep ensemble method with uncertainty management to detect atrial fibrillation from short-term single-lead ECGs, achieving high accuracy and enhancing telehealth clinical decision support systems.

This study introduces the Integrated Difference Autocorrelation (IDA) approach, which accurately estimates shear wave speed in elastography despite compression wave interference, showing robustness across simulations, imaging modalities, and clinical scenarios.

We present a signal processing technique for beam-steering radar architectures enabling simultaneous 2-D localization and vital signs monitoring, including heart rate and respiration rate, in human subjects.

This study is the first to develop a self-supervised denoiser for featureless RF phase corrupted by non-Gaussian noise, utilizing a randomly initialized CNN prior that is optimized with an unbiased estimator.

We present a signal processing technique for beam-steering radar architectures enabling simultaneous 2-D localization and vital signs monitoring, including heart rate and respiration rate, in human subjects.