From Narrow to Wide: Autoencoding Transformers for Ultrasound Bandwidth Recovery

TL;DR

This paper presents a data-driven autoencoder approach that enhances narrowband ultrasound signals to broadband quality, significantly improving image resolution and detail without hardware changes.

Contribution

It introduces a novel variation of the Tiny Vision Transformer auto-encoder trained on simulation data to recover broadband information from narrowband ultrasound signals.

Findings

90% reduction in image-domain MSE

6.7 dB increase in PSNR

SSIM improved to 0.965

Abstract

Conventional pulse-echo ultrasound suffers when low-cost probes deliver only narrow fractional bandwidths, elongating pulses and erasing high-frequency detail. We address this limitation by learning a data-driven mapping from band-limited to broadband spectrogram of radio-frequency (RF) lines. To this end, a variation of Tiny Vision Transform (ViT) auto-encoder is trained on simulation data using a curriculum-weighted loss. On heterogeneous speckle-cyst phantoms, the network reduces image-domain MSE by 90 percent, boosts PSNR by 6.7 dB, and raises SSIM to 0.965 compared with the narrow-band input. It also sharpens point-target rows in a completely unseen resolution phantom, demonstrating strong out-of-distribution generalisation without sacrificing frame rate or phase information. These results indicate that a purely software upgrade can endow installed narrow-band probes with broadband-like performance, potentially widening access to high-resolution ultrasound in resource-constrained settings.