Reverberant wavefield characterization and directional filtering for penile vibro-elastography: a pilot feasibility study in post-prostatectomy erectile dysfunction.

Objective. Penile vibro-elastography (PVE) is complicated by reverberant shear wavefields from the cylindrical geometry of the corpus cavernosum (CC), which bias shear wave speed (SWS) estimates. This study evaluated local phase velocity imaging with wavefield-based filtering (LPVI-WBF) for CC stiffness assessment in post-prostatectomy erectile dysfunction (ED), compared against phase gradient (PG) and time-of-flight (TOF) methods.Approach. Six patients with post-prostatectomy ED underwent PVE at four longitudinal visits (902 acquisitions). SWS was estimated in flaccid and erect states at 100, 150, and 200 Hz using LPVI-WBF, PG, and TOF. Wavefield composition was quantified via the standing wave index. Simulations evaluated accuracy and frequency stability under clean and reverberant conditions.Main results. Reverberant components dominated penile acquisitions, with traveling wave energy absent in 59% of acquisitions and mean standing wave fraction of 57 ± 17%. LPVI-WBF was the only method producing physiologically consistent estimates, with erect SWS exceeding flaccid (3.59 ± 0.48 vs 3.25 ± 0.58 m s-1,p< 0.0001). PG and TOF yielded reversed results (E<F, p< 0.0001). The dispersion exponent was near-zero for LPVI-WBF (α= 0.048,p= 0.21), consistent with weakly viscoelastic tissue, while PG (α= 0.184) and TOF (α= 0.321) showed significant frequency dependence from reverberant contamination. Simulations confirm PG accuracy in clean wavefields but collapse under reverberant ones.Significance. LPVI-WBF produced SWS estimates consistent with expected CC biomechanics across all patients and frequencies, a criterion PG and TOF failed to satisfy. Phantom validation and multi-directional wavefield simulation are needed to establish accuracy.