IQ-JEPA: A Joint-Embedding Predictive Architecture with a Hermitian Vision Transformer for Sound Speed and Attenuation Estimation from Ultrasound IQ Data

2026-07-24

AI Summary

Accurate estimation of tissue sound speed and attenuation from raw ultrasound IQ data is challenged by a nonlinear inverse problem and scarce annotations. This work proposes the IQ-JEPA framework, which introduces the Hermitian Vision Transformer to ultrasound modeling for the first time, incorporating phase-equivariant attention and conjugate-product feedforward modules that inherently align with the phase-shift characteristics induced by sound speed variations. Combining self-supervised pretraining via masked region prediction with fine-tuning on limited simulated labels, the method achieves a sound speed estimation error of 15.60 m/s using only 10,000 labeled samples—more than three times more label-efficient than purely supervised approaches. With full labeling, the error further reduces to 8.71 m/s, and the model successfully generalizes to attenuation estimation and cross-distribution scenarios.

Abstract

The speed of sound in tissue is a prerequisite for well-focused imaging and has diagnostic value, but recovering it from raw pulse-echo channel data is fundamentally a nonlinear inverse problem. Learned solvers are fast yet label hungry. Simulated sound-speed labels are expensive, while abundant real channel data is unlabeled. We propose IQ-JEPA to exploit both data types. An encoder is pretrained without labels to predict the latent representation of masked in-phase and quadrature (IQ) regions from visible context, then fine-tuned on simulated maps. Sound speed appears in the IQ signal as a phase difference, invariant to the constant phase offset. The encoder is a Hermitian vision transformer that operates on the complex signal directly. Its attention is equivariant to that phase and its conjugate-product feed-forward is invariant to it, so the encoder reads a quantity analogous to the one classical coherence methods use. On 79,293 Fullwave 2.5 simulations at 2.5 MHz, pretraining on the 63,435 unlabeled acquisitions reaches 15.60 m/s at 10,000 labels. This is a roughly threefold gain in label efficiency over supervised training, growing to over fourfold at 1,000 labels. It is about 2.2x below an InversionNet baseline, and 8.71 m/s at full labels. The gain still grows with more unlabeled pretraining data. Our comparisons point to self-supervision as the dominant factor. The same encoder transfers. Its frozen features expose sound speed and attenuation, and cross-distribution pretraining between layered and abdominal phantoms costs little accuracy. We see this as a first step toward a foundation model for quantitative ultrasound.

Problem

• sound speed estimation
• ultrasound IQ data
• attenuation estimation
• label efficiency
• quantitative ultrasound

Innovation

• Hermitian Vision Transformer
• self-supervised pretraining
• ultrasound IQ data
• sound speed estimation
• label efficiency