# Assessing the generalization performance of SAM for ureteroscopy scene understanding

arXiv: https://arxiv.org/abs/2505.17210 · HTML: https://mlsanity.com/paper/2505.17210/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 8 of its 18 references that are in the map.
- First posted on arXiv 22 May 2025.

## Builds on (most cited first, 8 of 8)

- [U-Net: Convolutional Networks for Biomedical Image Segmentation](https://mlsanity.com/paper/1505.04597/) (2015)
- [Segment Anything](https://mlsanity.com/paper/2304.02643/) (2023)
- [Assessing deep learning methods for the identification of kidney stones in endoscopic images](https://mlsanity.com/paper/2103.01146/) (2021)
- [Towards Automatic Recognition of Pure & Mixed Stones using Intraoperative Endoscopic Digital Images](https://mlsanity.com/paper/2105.10686/) (2021)
- [On the in vivo recognition of kidney stones using machine learning](https://mlsanity.com/paper/2201.08865/) (2022)
- [On the generalization capabilities of FSL methods through domain adaptation: a case study in endoscopic kidney stone image classification](https://mlsanity.com/paper/2205.00895/) (2022)
- [Multi-class motion-based semantic segmentation for ureteroscopy and laser lithotripsy](https://mlsanity.com/paper/2104.01268/) (2021)
- [Deep morphological recognition of kidney stones using intra-operative endoscopic digital videos](https://mlsanity.com/paper/2205.06093/) (2022)

## Abstract (arXiv)

The segmentation of kidney stones is regarded as a critical preliminary step to enable the identification of urinary stone types through machine- or deep-learning-based approaches. In urology, manual segmentation is considered tedious and impractical due to the typically large scale of image databases and the continuous generation of new data. In this study, the potential of the Segment Anything Model (SAM) -- a state-of-the-art deep learning framework -- is investigated for the automation of kidney stone segmentation. The performance of SAM is evaluated in comparison to traditional models, including U-Net, Residual U-Net, and Attention U-Net, which, despite their efficiency, frequently exhibit limitations in generalizing to unseen datasets. The findings highlight SAM's superior adaptability and efficiency. While SAM achieves comparable performance to U-Net on in-distribution data (Accuracy: 97.68 + 3.04; Dice: 97.78 + 2.47; IoU: 95.76 + 4.18), it demonstrates significantly enhanced generalization capabilities on out-of-distribution data, surpassing all U-Net variants by margins of up to 23 percent.