Metraipolis › archive › dc_0dab773feec7
Wrong Organ, Right Physics: Transferring Echocardiography Pretraining to Lung Ultrasound for Tuberculosis Screening — arXiv:2610.03290
- Source
- arXiv
- Original
- https://arxiv.org/abs/2610.03290
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-05T23:10:50.027Z
- SHA-256 of the archived text
58ecebaac9b27c9267dd5982018d1518ca7d42f41698782d11908305e33980a9- Stable URL of this page
https://metraipolis.com/documento/dc_0dab773feec7
How to cite
Wrong Organ, Right Physics: Transferring Echocardiography Pretraining to Lung Ultrasound for Tuberculosis Screening — arXiv:2610.03290 — arXiv, https://arxiv.org/abs/2610.03290 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_0dab773feec7 (Metraipolis archive dc_0dab773feec7, sha256 58ecebaac9b27c9267dd5982018d1518ca7d42f41698782d11908305e33980a9)
BibTeX
@misc{dc0dab773feec7,
title = {Wrong Organ, Right Physics: Transferring Echocardiography Pretraining to Lung Ultrasound for Tuberculosis Screening — arXiv:2610.03290},
howpublished = {arXiv},
url = {https://arxiv.org/abs/2610.03290},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_0dab773feec7; sha256 58ecebaac9b27c9267dd5982018d1518ca7d42f41698782d11908305e33980a9},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv, https://arxiv.org/abs/2610.03290. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
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.
Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.LG, nuovi lavori del giorno).
Categorie arXiv: Image and Video Processing (eess.IV) ; Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2610.03290 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-05T23:01:38.768Z dalla pagina https://arxiv.org/list/cs.LG/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).