UltraBench 2 is a new benchmark designed to evaluate vision foundation models on ultrasound images, addressing the lack of standardized tests in this area. It covers a wide range of anatomical structures and tasks, emphasizing reproducibility and ease of use. The authors compare existing models, finding that ultrasound-specific pretraining still outperforms on classification, while general-purpose models have matched performance on segmentation.
By Ashwath Radhachandran, Adam Tupper, Christian Gagn\'e, William Speier
US-JEPA introduces a self‑supervised framework for ultrasound imaging that predicts masked latent representations instead of raw pixels, using a frozen, domain‑specific teacher to provide stable targets. This approach avoids the hyperparameter sensitivity and computational cost of traditional online teachers, enabling the student model to build upon the teacher’s semantic priors. The authors benchmark US‑JEPA against all publicly available ultrasound foundation models on UltraBench, showing competitive or superior performance across multiple organs and pathological conditions under linear probing.
By Ashwath Radhachandran, Vedrana Ivezi\'c, Shreeram Athreya, Corey W. Arnold, William Speier
Risk‑Routed Implicit Boundary Refinement (RIBR) is a compact framework designed to improve boundary delineation in medical ultrasound images. It refines convolutional segmentation by applying spatially weighted implicit residuals, where a coordinate‑conditioned implicit branch predicts logit corrections and a learned gate modulates their influence based on predictive uncertainty and local appearance. Geometry‑ and speckle‑aware regularization provides contour supervision, and RIBR outperforms ten baselines across nine datasets, achieving a macro‑average Dice of 76.67% and the lowest Hausdorff distance on all datasets.
By Jingguo Qu, Xinyang Han, Xiang Wang, Yuqi Yang, Tonghuan Xiao, Sheng Ning, Jing Qin, Ann Dorothy King, Winnie Chiu-Wing Chu, Jing Cai, Michael Ying
The paper introduces Segment Anything Small (SAS), a data‑augmentation method that improves deep‑learning segmentation of small anatomical structures in ultrasound images. SAS uses two transformations: resizing and embedding organ thumbnails into a black background to vary organ scale, and adding noise to regions of interest to mimic tissue texture variability. Experiments on one internal and five external datasets show Dice score gains up to 0.35, with an average improvement of 0.16, and demonstrate that SAS enhances model robustness and generalizability without adding hallucinations or artifacts.
By Danielle L. Ferreira, Ahana Gangopadhyay, Hsi-Ming Chang, Ravi Soni, Gopal Avinash
arXiv:2608. 04766v1 Announce Type: cross Abstract: A large number of infants with congenital anomalies are born each year globally, especially in areas with underdeveloped medical resources.
By Bin Pu, Jiewen Yang, Liwen Wang, Ying Tan, Guannan He, Xingbo Dong, Qika Lin, Jiarong Guo, Lixian Yang, Zuozhu Liu, Shengli Li, Kenli Li
UltraDiff is a modular framework that extends differentiable rendering to medical ultrasound by modeling image formation as a path‑space integral gated by travel time. It derives a Monte Carlo estimator for both the forward ultrasound model and its gradients, enabling gradient‑based optimization of scene parameters. The authors demonstrate the method on inverse geometry estimation, recovering vertebral surfaces from simulated and real B‑mode sweeps without requiring pre‑segmented images, achieving competitive geometric accuracy.
By Felix Duelmer, Magdalena Wysocki, Nassir Navab, Mohammad Farid Azampour