EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Segment anything small for ultrasound: Enhancing segmentation with non-generative augmentation

Danielle L Ferreira iD, Ahana Gangopadhyay, Hsi-Ming Chang, Ravi Soni, Gopal Avinash

PLOS Digital Health · 08.04.2026 · 5 (4) · e0001309

DOI: 10.1371/journal.pdig.0001309 ↗

Kurz gefasst

Fachartikel in der Zeitschrift PLOS Digital Health, online erschienen am 08.04.2026. Verfasst von Danielle L Ferreira und 4 weiteren Autor:innen. Beteiligte Einrichtung: Science and Technology Office, GE HealthCare, San Ramon, California, United States of America. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 34 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Accurate segmentation of anatomical structures in ultrasound (US) images, particularly small ones, is challenging due to noise and variability in imaging conditions (e.g., probe position, patient anatomy, tissue characteristics and pathology). To address this, we introduce Segment Anything Small (SAS), a simple yet effective scale- and texture-aware data augmentation technique designed to enhance the performance of deep learning models for segmenting small anatomical structures in ultrasound images. SAS employs a dual transformation strategy: (1) simulating diverse organ scales by resizing and embedding organ thumbnails into a black background, and (2) injecting noise into regions of interest to simulate varying tissue textures. These transformations generate realistic and diverse training data without introducing hallucinations or artifacts, improving the model’s robustness to noise and variability. We fine-tuned a promptable foundation model on a controlled organ-specific medical imaging dataset and evaluated its performance on one internal and five external datasets. Experimental results demonstrate significant improvements in segmentation performance, with Dice score gains of up to 0.35 and an average improvement of 0.16 [95% CI: 0.132,0.188]. Additionally, our iterative point prompts provide precise control and adaptive refinement, achieving performance comparable to bounding box prompts with just two points. SAS enhances model robustness and generalizability across diverse anatomical structures and imaging conditions, particularly for small structures, without compromising the accuracy of larger ones. By offering a computationally efficient solution that eliminates the need for extensive human labeling efforts, SAS enables performance improvements in medical image analysis, particularly in resource-constrained settings.

Autor:innen und Institutionen

  • Science and Technology Office, GE HealthCare, San Ramon, California, United States of America.

ORCID-Profile: Danielle L Ferreira

Zitieren

APA

Ferreira, D. L., Gangopadhyay, A., Chang, H. M., Soni, R., & Avinash, G. (2026). Segment anything small for ultrasound: Enhancing segmentation with non-generative augmentation. PLOS Digital Health, 5(4), e0001309. https://doi.org/10.1371/journal.pdig.0001309

Vancouver

Ferreira DL, Gangopadhyay A, Chang HM, Soni R, Avinash G. Segment anything small for ultrasound: Enhancing segmentation with non-generative augmentation. PLOS Digital Health. 2026;5(4):e0001309. doi:10.1371/journal.pdig.0001309

BibTeX
@article{Ferreira2026,
  author = {Ferreira, Danielle L and Gangopadhyay, Ahana and Chang, Hsi-Ming and Soni, Ravi and Avinash, Gopal},
  title = {Segment anything small for ultrasound: Enhancing segmentation with non-generative augmentation},
  journal = {PLOS Digital Health},
  year = {2026},
  volume = {5},
  number = {4},
  pages = {e0001309},
  doi = {10.1371/journal.pdig.0001309}
}
Kostenlos starten →