Searcharxiv⌕ Search
Home/Archive/arXiv

arXiv · 2610.03290

Wrong Organ, Right Physics: Transferring Echocardiography Pretraining to Lung Ultrasound for Tuberculosis Screening

Christiaan M. Geldenhuys·Joshua M. Jansen van Vüren·Véronique Suttels·Trevor Brokowski·Ablo P. Wachinou·Mary-Anne Hartley·Rensu P. Theart·Grant Theron·Thomas R. Niesler
View original source ↗Read source document ↗Export metadata ↓

Abstract

Lung ultrasound (LUS) is attractive for tuberculosis (TB) screening at primary-care level, but labelled cohorts are small. Echocardiography carries no such constraint, while sharing the same underlying ultrasound imaging physics, signal processing and B-mode appearance as LUS. We ask whether an encoder pretrained on that high-resource ultrasound domain carries representations that remain usable in the low-resource one. Only the encoder varies, across seventeen encoders spanning three architecture families. Among them, a latent-predictive video encoder pretrained on generic video (V-JEPA2-L) and its echocardiography counterpart (EchoJEPA-L) differ in pretraining corpus alone. The choice among these encoders does not resolve the classification, the whole family spanning 2.50 percentage points against a measurement resolution of 2.71. What moves the task instead is feature conditioning. Standardising the features between the encoder and the classifier improves all seventeen encoders by a mean of +1.23 percentage points at $p=1.5\times10^{-5}$. On the held-out test set every encoder selected on the development folds stands above the baseline system by up to +2.57 percentage points of area under the receiver operating characteristic curve (AUROC), and specificity at 90% sensitivity reaches 79.3% against 60.3%. The contrast specified in advance, EchoJEPA-L against V-JEPA2-L, measures -0.16 percentage points at $p=0.926$. We therefore find no evidence that shared ultrasonic physics alone makes echocardiography a more productive pretraining corpus than generic video, and any advantage, if present, is smaller than this cohort can resolve. The video encoders receive replicated still images, however, so whether this absence of an effect reflects the pretraining domain or a video encoder applied to static frames cannot be separated. The limiting factor is the labelled cohort rather than the encoder.

Explore related subjects

eess.IVcs.CVcs.LG

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

Christiaan M. Geldenhuys, Joshua M. Jansen van Vüren, Véronique Suttels, Trevor Brokowski, Ablo P. Wachinou, Mary-Anne Hartley, Rensu P. Theart, Grant Theron, Thomas R. Niesler. 2026-10-02. Wrong Organ, Right Physics: Transferring Echocardiography Pretraining to Lung Ultrasound for Tuberculosis Screening. https://arxiv.org/abs/2610.03290

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related papers

arXiv2026

A Unified Deep Learning Framework for Motion Correction in Medical Imaging

Deep learning has shown significant value in medical image registration for motion correction; however, current techniques are either limited by the type and range of motion they can handle or require iterative inference and/or retraining for new imaging data. To address these limitations, we introduce UniMo, a Unified Motion Correction framework that uses deep neural networks to correct various types of motion in medical imaging. UniMo uses an alternating optimization scheme with a unified loss function to train an integrated model of 1) an equivariant neural network for global rigid motion correction and 2) an encoder-decoder network for local deformations. It features a geometric deformation augmenter that 1) enhances the robustness of global motion correction by addressing local deformations, whether caused by non-rigid motion or geometric distortions, and 2) generates augmented data to improve training. As a hybrid model that uses both image intensities and shapes, UniMo is robust to appearance variations and generalizes to various imaging modalities without retraining. We trained and tested UniMo for motion tracking in fetal magnetic resonance imaging, which is challenging due to 1) both large rigid and non-rigid motion and 2) large variations in image appearance. We then tested the trained model, without retraining, on three public datasets: MedMNIST, lung CT, and BraTS. UniMo surpassed existing motion correction methods in accuracy and, notably, enabled one-time training on a single modality while maintaining high stability and adaptability across multiple unseen imaging datasets. By offering a unified solution to motion correction, UniMo marks a significant advance in challenging applications with a mixture of bulk motion and local deformations. Code is available at https://github.com/IntelligentImaging/UNIMO

eess.IV↗
arXiv2026

MedForj: An open, large-scale foundational generative prior for high-resolution 3D brain MRI

This work introduces MedForj, a suite of 3D foundational generative priors based on diffusion models. The MedForj models were trained on $72{,}659$ 1~mm isotropic 3D $T_1$-weighted MRI human brain image volumes from $38{,}174$ subjects, drawn from a curated corpus of $80{,}675$ volumes from $42{,}506$ subjects spanning $38$ publicly available datasets. These training images were manually inspected to exclude those with poor quality and excessive pathology, and otherwise were minimally processed. The models include six different diffusion training strategies: rectified flow, latent diffusion rectified flow, flow matching, velocity prediction, clean prediction, and noise prediction. Image samples produced by each of these models were compared to each other and against real, ground truth data under downstream segmentation distributions, FID, five inverse problems, and blind human inspection in an observer study. Flow matching was the strongest strategy overall, achieving the best inverse problem solving results at $28.80$~dB PSNR and $0.874$ SSIM averaged over the five forward problems, the highest rate of reconstructions judged real by blind human raters at $72.6\%$, and the closest per-structure match to real segmented anatomy in a permutation test. It was not best everywhere: rectified flow produced the most convincing unconditional samples in the observer study and the best FID, and the latent rectified-flow model achieved the smallest joint distributional distance to real anatomy. No other strategy, however, performed consistently well across all four evaluations. We therefore recommend flow matching as the default MedForj prior, while releasing every strategy so that the choice can be revisited per application. All model weights and corresponding code are publicly available at https://github.com/piksl-research/medforj.

eess.IV↗
arXiv2026

Gaussian Surrogates for Poisson Imaging: Some Theoretical and Empirical Results

In imaging inverse problems with Poisson-distributed measurements, it is common to use objectives derived from the Poisson likelihood. But performance is often evaluated by mean squared error (MSE), which raises a practical question: how much does a Poisson objective matter for MSE, even at low dose? We analyze the MSE of Poisson and Gaussian surrogate reconstruction objectives under Poisson noise. In a stylized diagonal model, we show that the unregularized Poisson maximum-likelihood estimator can incur large MSE at low dose, while Poisson MAP mitigates this instability through regularization. We then study two Gaussian surrogate objectives: a heteroscedastic quadratic objective motivated by the normal approximation of Poisson data, and a homoscedastic quadratic objective that yields a simple linear estimator. We show that both surrogates can achieve MSE comparable to Poisson MAP in the low-dose regime, despite departing from the Poisson likelihood. Numerical computed tomography experiments indicate that these conclusions extend beyond the stylized setting of our theoretical analysis.

eess.IV↗