aiqbench analysis
The paper applies a learned contrastive model to ultrasound imaging, separating received signals into a coherent anatomy component and a noise component, and trains at test time from a single acquisition. The reported gains (69.3% SNR improvement in simulation, 84.8% in vivo) come from the authors' own experiments and have not been independently validated. For ML readers, the useful part is the test-time training approach, which avoids pretraining and domain-shift problems in medical imaging.
Coherence-Guided Pyramid Self-Contrastive Disentanglement for Ultrasound Imaging via Single-Shot Test-Time Training
arXiv:2605.12567v3 Announce Type: replace-cross Abstract: The inherent electronic noise, side lobes, and speckle noise complicate clinical interpretation of ultrasound images. Conventional denoising methods rely on explicit noise assumptions that may underperform in composite noise conditions. Learning-based methods are usually pretrained in a limited image domain and face inevitable domain shift in complex \textit{in vivo} environments. This study proposes a computational ultrasound imaging pipeline with a Pyramid Self-Contrastive Disentanglement (PSCD) model. We generalize the explicit IQ signal coherence to a learnable latent space and formulate it as a contrastive critic to implicitly disentangle the received IQ signals into a coherent anatomy space and an incoherent noise space. Clean IQ signals for B-mode image formation are estimated from the anatomy space exclusively. We first apply PSCD to synthetic aperture ultrasound, where an Aperture-to-Aperture proxy task constrains the selective reconstruction in dual domains. This framework can be trained at test time using single-shot imaging data. Simulation experiments, including noise levels from 0 to 30 dB with simple to complex geometries, demonstrated improvements of 69.3$\%$ in SNR and 34.4$\%$ in CNR. The \textit{in vivo} experiments, including human hearts in six echocardiographic views and other abdominal organs, showed 84.8$\%$ SNR and 25.7$\%$ CNR gains using only two apertures. Our PSCD also achieved higher image resolution and boundary sharpness, while preserving the high-frequency details. It can naturally generalize to other ultrasound imaging pipelines involving multiple transmit-receive events, enabling precise anatomical visualization without domain shift and pretraining costs.
Abstract:The inherent electronic noise, side lobes, and speckle noise complicate clinical interpretation of ultrasound images. Conventional denoising methods rely on explicit noise assumptions that may underperform in composite noise conditions. Learning-based methods are usually pretrained in a limited image domain and face inevitable domain shift in complex \textit{in vivo} environments. This study proposes a computational ultrasound imaging pipeline with a Pyramid Self-Contrastive Disentanglement (PSCD) model. We generalize the explicit IQ signal coherence to a learnable latent space and formulate it as a contrastive critic to implicitly disentangle the received IQ signals into a coherent anatomy space and an incoherent noise space. Clean IQ signals for B-mode image formation are estimated from the anatomy space exclusively. We first apply PSCD to synthetic aperture ultrasound, where an Aperture-to-Aperture proxy task constrains the selective reconstruction in dual domains. This framework can be trained at test time using single-shot imaging data. Simulation experiments, including noise levels from 0 to 30 dB with simple to complex geometries, demonstrated improvements of 69.3$\%$ in SNR and 34.4$\%$ in CNR. The \textit{in vivo} experiments, including human hearts in six echocardiographic views and other abdominal organs, showed 84.8$\%$ SNR and 25.7$\%$ CNR gains using only two apertures. Our PSCD also achieved higher image resolution and boundary sharpness, while preserving the high-frequency details. It can naturally generalize to other ultrasound imaging pipelines involving multiple transmit-receive events, enabling precise anatomical visualization without domain shift and pretraining costs.
Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
arXiv:2605.12567 [cs.CV]
(or arXiv:2605.12567v3 [cs.CV] for this version)
https://doi.org/10.48550/arXiv.2605.12567 Focus to learn more arXiv-issued DOI via DataCite
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 334 words