EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Fine-Tuned Segment Anything Model with Low-Rank Adaptation for Chest X-Ray Images

Saeed S Alahmari, Michael R Gardner iD, Fawaz Alqahtani, Tawfiq Salem

Diagnostics · 12.03.2026 · 16 (6) · 847

DOI: 10.3390/diagnostics16060847 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Diagnostics, online erschienen am 12.03.2026. Verfasst von Saeed S Alahmari und 3 weiteren Autor:innen. Beteiligte Einrichtung: Department of Computer Science, Najran University, Najran 66462, Saudi Arabia. Die Arbeit nennt 35 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Deanship of Graduate Studies and Scientific Research at Najran University, King Faisal University, Saudi Arabia.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Background

This paper investigates the use of the Segment Anything Model (SAM) for chest X-ray (CXR) image segmentation, with a focus on improving its performance using low-rank adaptation (LoRA).

Methods

We evaluate three versions of SAM: two zero-shot methods (using coordinate and bounding box prompts) and a fine-tuned SAM using LoRA. To support these approaches, we also trained two standard convolutional neural networks (CNNs), U-Net and DeepLabv3+, to generate draft lung segmentations that serve as input prompts for the SAM methods. Our fine-tuning approach uses LoRA to add lightweight trainable adapters within the Transformer blocks of the SAM, allowing only a small subset of parameters to be updated. The rest of the SAM remains frozen, helping preserve its pre-trained knowledge while reducing memory and computational needs. We tested all models on a dataset of CXR images labeled for COVID-19, viral pneumonia, and normal cases.

Results

Results show that fine-tuned SAM with LoRA outperforms zero-shot SAM methods and CNN baselines in terms of segmentation accuracy and efficiency.

Conclusions

This demonstrates the potential of combining LoRA with SAM for practical and effective medical image segmentation.

Schlagwörter

Stichwörter der Autor:innen

SAMchest X-raydeep learningsegmentation

Autor:innen und Institutionen

  • Department of Computer Science, Najran University, Najran 66462, Saudi Arabia.
  • Department of Biomedical Engineering, College of Engineering, King Faisal University, Al Ahsa 31982, Saudi Arabia.
  • Department of Radiological Sciences, College of Applied Medical Sciences, King Khalid University, Abha 61421, Saudi Arabia.
  • School of Applied and Creative Computing, Purdue University, West Lafayette, IN 47906, USA.

ORCID-Profile: Michael R Gardner

Förderung

  • Deanship of Graduate Studies and Scientific Research at Najran University
  • King Faisal University, Saudi Arabia

Zitieren

APA

Alahmari, S. S., Gardner, M. R., Alqahtani, F., & Salem, T. (2026). Fine-Tuned Segment Anything Model with Low-Rank Adaptation for Chest X-Ray Images. Diagnostics, 16(6), 847. https://doi.org/10.3390/diagnostics16060847

Vancouver

Alahmari SS, Gardner MR, Alqahtani F, Salem T. Fine-Tuned Segment Anything Model with Low-Rank Adaptation for Chest X-Ray Images. Diagnostics. 2026;16(6):847. doi:10.3390/diagnostics16060847

BibTeX
@article{Alahmari2026,
  author = {Alahmari, Saeed S and Gardner, Michael R and Alqahtani, Fawaz and Salem, Tawfiq},
  title = {Fine-Tuned Segment Anything Model with Low-Rank Adaptation for Chest X-Ray Images},
  journal = {Diagnostics},
  year = {2026},
  volume = {16},
  number = {6},
  pages = {847},
  doi = {10.3390/diagnostics16060847}
}
Kostenlos starten →