Gespeichert in:
| Hauptverfasser: | , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2025
|
| Schlagworte: | |
| Online-Zugang: | https://arxiv.org/abs/2512.16685 |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866909969025073152 |
|---|---|
| author | Alves, Gonçalo Gaspar Zadeh, Shekoufeh Gorgi Husch, Andreas Bausch, Ben |
| author_facet | Alves, Gonçalo Gaspar Zadeh, Shekoufeh Gorgi Husch, Andreas Bausch, Ben |
| contents | Combining open-source datasets can introduce data leakage if the same subject appears in multiple sets, leading to inflated model performance. To address this, we explore subject fingerprinting, mapping all images of a subject to a distinct region in latent space, to enable subject re-identification via similarity matching. Using a ResNet-50 trained with triplet margin loss, we evaluate few-shot fingerprinting on 3D MRI and 2D X-ray data in both standard (20-way 1-shot) and challenging (1000-way 1-shot) scenarios. The model achieves high Mean- Recall-@-K scores: 99.10% (20-way 1-shot) and 90.06% (500-way 5-shot) on ChestXray-14; 99.20% (20-way 1-shot) and 98.86% (100-way 3-shot) on BraTS- 2021. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2512_16685 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Few-Shot Fingerprinting Subject Re-Identification in 3D-MRI and 2D-X-Ray Alves, Gonçalo Gaspar Zadeh, Shekoufeh Gorgi Husch, Andreas Bausch, Ben Computer Vision and Pattern Recognition Artificial Intelligence Combining open-source datasets can introduce data leakage if the same subject appears in multiple sets, leading to inflated model performance. To address this, we explore subject fingerprinting, mapping all images of a subject to a distinct region in latent space, to enable subject re-identification via similarity matching. Using a ResNet-50 trained with triplet margin loss, we evaluate few-shot fingerprinting on 3D MRI and 2D X-ray data in both standard (20-way 1-shot) and challenging (1000-way 1-shot) scenarios. The model achieves high Mean- Recall-@-K scores: 99.10% (20-way 1-shot) and 90.06% (500-way 5-shot) on ChestXray-14; 99.20% (20-way 1-shot) and 98.86% (100-way 3-shot) on BraTS- 2021. |
| title | Few-Shot Fingerprinting Subject Re-Identification in 3D-MRI and 2D-X-Ray |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence |
| url | https://arxiv.org/abs/2512.16685 |