Addressing Data Scarcity in 3D Trauma Detection through Self-Supervised and Semi-Supervised Learning with Vertex Relative Position Encoding
Fuente:
arXiv
Salvato in:
| Autori principali: | , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2026
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866914390363602944 |
|---|---|
| author | Chaudhary, Shivam Bhat, Sheethal Maier, Andreas |
| author_facet | Chaudhary, Shivam Bhat, Sheethal Maier, Andreas |
| contents | Accurate detection and localization of traumatic injuries in abdominal CT scans remains a critical challenge in emergency radiology, primarily due to severe scarcity of annotated medical data. This paper presents a label-efficient approach combining self-supervised pre-training with semi-supervised detection for 3D medical image analysis. We employ patch-based Masked Image Modeling (MIM) to pre-train a 3D U-Net encoder on 1,206 CT volumes without annotations, learning robust anatomical representations. The pretrained encoder enables two downstream clinical tasks: 3D injury detection using VDETR with Vertex Relative Position Encoding, and multi-label injury classification. For detection, semi-supervised learning with 2,000 unlabeled volumes and consistency regularization achieves 56.57% validation mAP@0.50 and 45.30% test mAP@0.50 with only 144 labeled training samples, representing a 115% improvement over supervised-only training. For classification, expanding to 2,244 labeled samples yields 94.07% test accuracy across seven injury categories using only a frozen encoder, demonstrating immediately transferable self-supervised features. Our results validate that self-supervised pre-training combined with semi-supervised learning effectively addresses label scarcity in medical imaging, enabling robust 3D object detection with limited annotations. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_12514 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Addressing Data Scarcity in 3D Trauma Detection through Self-Supervised and Semi-Supervised Learning with Vertex Relative Position Encoding Chaudhary, Shivam Bhat, Sheethal Maier, Andreas Computer Vision and Pattern Recognition Machine Learning Accurate detection and localization of traumatic injuries in abdominal CT scans remains a critical challenge in emergency radiology, primarily due to severe scarcity of annotated medical data. This paper presents a label-efficient approach combining self-supervised pre-training with semi-supervised detection for 3D medical image analysis. We employ patch-based Masked Image Modeling (MIM) to pre-train a 3D U-Net encoder on 1,206 CT volumes without annotations, learning robust anatomical representations. The pretrained encoder enables two downstream clinical tasks: 3D injury detection using VDETR with Vertex Relative Position Encoding, and multi-label injury classification. For detection, semi-supervised learning with 2,000 unlabeled volumes and consistency regularization achieves 56.57% validation mAP@0.50 and 45.30% test mAP@0.50 with only 144 labeled training samples, representing a 115% improvement over supervised-only training. For classification, expanding to 2,244 labeled samples yields 94.07% test accuracy across seven injury categories using only a frozen encoder, demonstrating immediately transferable self-supervised features. Our results validate that self-supervised pre-training combined with semi-supervised learning effectively addresses label scarcity in medical imaging, enabling robust 3D object detection with limited annotations. |
| title | Addressing Data Scarcity in 3D Trauma Detection through Self-Supervised and Semi-Supervised Learning with Vertex Relative Position Encoding |
| topic | Computer Vision and Pattern Recognition Machine Learning |
| url | https://arxiv.org/abs/2603.12514 |