Addressing Data Scarcity in 3D Trauma Detection through Self-Supervised and Semi-Supervised Learning with Vertex Relative Position Encoding

• URL: http://arxiv.org/abs/2603.12514v1
• Date: Thu, 12 Mar 2026 23:17:17 GMT
• Title: Addressing Data Scarcity in 3D Trauma Detection through Self-Supervised and Semi-Supervised Learning with Vertex Relative Position Encoding
• Abstract summary: This paper presents a label-efficient approach combining self-supervised pre-training with semi-supervised detection for 3D medical image analysis.<n>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.<n>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.<n>For classification, expanding to 2,244 labeled samples yields 94.07% test accuracy across seven injury
• Score: 3.9641557356685166
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: 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.

Related papers

• Disease-Centric Vision-Language Pretraining with Hybrid Visual Encoding for 3D Computed Tomography [75.7789001619107]
  Vision-language pre-training holds great promise for general-purpose medical AI.<n>We propose a tailored framework featuring three key components.<n>A diagnosis-aware prompt strategy employs real clinical phrases and aggregated disease prototypes to bridge the pre-training-inference gap.
  arXiv  Detail & Related papers  (2026-06-24T08:24:45Z)
• DinoAtten3D: Slice-Level Attention Aggregation of DinoV2 for 3D Brain MRI Anomaly Classification [2.731729370870452]
  Anomaly detection and classification in medical imaging are critical for early diagnosis but remain challenging due to limited annotated data, class imbalance, and the high cost of expert labeling.<n>We propose an attention-based global aggregation framework tailored specifically for 3D medical image anomaly classification.
  arXiv  Detail & Related papers  (2025-09-15T23:31:40Z)
• 3D Universal Lesion Detection and Tagging in CT with Self-Training [3.68620908362189]
  We propose a self-training pipeline to detect 3D lesions and tag them according to the body part they occur in.<n>To our knowledge, we are the first to jointly detect lesions in 3D and tag them according to the body part label.
  arXiv  Detail & Related papers  (2025-04-07T15:50:27Z)
• Anatomically-guided masked autoencoder pre-training for aneurysm detection [6.025753055139489]
  Intracranial aneurysms are a major cause of morbidity and mortality worldwide.<n>We propose a novel pre-training strategy using unannotated head CT scan data to pre-train a 3D Vision Transformer model.<n>Compared with SOTA aneurysm detection models, our approach gains +4-8% absolute Sensitivity at a false positive rate of 0.5.
  arXiv  Detail & Related papers  (2025-02-28T17:13:58Z)
• Detection of Intracranial Hemorrhage for Trauma Patients [1.0074894923170512]
  We propose a novel Voxel-Complete IoU (VC-IoU) loss that encourages the network to learn the 3D aspect ratios of bounding boxes. We extensively experiment on brain bleeding detection using a publicly available dataset, and validate it on a private cohort.
  arXiv  Detail & Related papers  (2024-08-20T12:03:59Z)
• Deep Omni-supervised Learning for Rib Fracture Detection from Chest Radiology Images [41.62893318123283]
  Deep learning (DL)-based rib fracture detection has shown promise of playing an important role in preventing mortality and improving patient outcome. DL-based object detection models requires a huge amount of bounding box annotation. Annotating medical data is time-consuming and expertise-demanding, making obtaining a large amount of fine-grained annotations extremely infeasible. We present a novel omni-supervised object detection network, ORF-Netv2, to leverage as much available supervision as possible.
  arXiv  Detail & Related papers  (2023-06-23T05:36:03Z)
• Self-supervised contrastive learning of echocardiogram videos enables label-efficient cardiac disease diagnosis [48.64462717254158]
  We developed a self-supervised contrastive learning approach, EchoCLR, to catered to echocardiogram videos. When fine-tuned on small portions of labeled data, EchoCLR pretraining significantly improved classification performance for left ventricular hypertrophy (LVH) and aortic stenosis (AS) EchoCLR is unique in its ability to learn representations of medical videos and demonstrates that SSL can enable label-efficient disease classification from small, labeled datasets.
  arXiv  Detail & Related papers  (2022-07-23T19:17:26Z)
• 3D unsupervised anomaly detection and localization through virtual multi-view projection and reconstruction: Clinical validation on low-dose chest computed tomography [2.2302915692528367]
  We propose a method based on a deep neural network for computer-aided diagnosis called virtual multi-view projection and reconstruction. The proposed method improves the patient-level anomaly detection by 10% compared with a gold standard based on supervised learning. It localizes the anomaly region with 93% accuracy, demonstrating its high performance.
  arXiv  Detail & Related papers  (2022-06-18T13:22:00Z)
• Self-Supervised Learning as a Means To Reduce the Need for Labeled Data in Medical Image Analysis [64.4093648042484]
  We use a dataset of chest X-ray images with bounding box labels for 13 different classes of anomalies. We show that it is possible to achieve similar performance to a fully supervised model in terms of mean average precision and accuracy with only 60% of the labeled data.
  arXiv  Detail & Related papers  (2022-06-01T09:20:30Z)
• Unsupervised Anomaly Detection in 3D Brain MRI using Deep Learning with impured training data [53.122045119395594]
  We study how unhealthy samples within the training data affect anomaly detection performance for brain MRI-scans. We evaluate a method to identify falsely labeled samples directly during training based on the reconstruction error of the AE.
  arXiv  Detail & Related papers  (2022-04-12T13:05:18Z)
• Revisiting 3D Context Modeling with Supervised Pre-training for Universal Lesion Detection in CT Slices [48.85784310158493]
  We propose a Modified Pseudo-3D Feature Pyramid Network (MP3D FPN) to efficiently extract 3D context enhanced 2D features for universal lesion detection in CT slices. With the novel pre-training method, the proposed MP3D FPN achieves state-of-the-art detection performance on the DeepLesion dataset. The proposed 3D pre-trained weights can potentially be used to boost the performance of other 3D medical image analysis tasks.
  arXiv  Detail & Related papers  (2020-12-16T07:11:16Z)
• Deep Q-Network-Driven Catheter Segmentation in 3D US by Hybrid Constrained Semi-Supervised Learning and Dual-UNet [74.22397862400177]
  We propose a novel catheter segmentation approach, which requests fewer annotations than the supervised learning method. Our scheme considers a deep Q learning as the pre-localization step, which avoids voxel-level annotation. With the detected catheter, patch-based Dual-UNet is applied to segment the catheter in 3D volumetric data.
  arXiv  Detail & Related papers  (2020-06-25T21:10:04Z)
• Volumetric Attention for 3D Medical Image Segmentation and Detection [53.041572035020344]
  A volumetric attention(VA) module for 3D medical image segmentation and detection is proposed. VA attention is inspired by recent advances in video processing, enables 2.5D networks to leverage context information along the z direction. Its integration in the Mask R-CNN is shown to enable state-of-the-art performance on the Liver Tumor (LiTS) Challenge.
  arXiv  Detail & Related papers  (2020-04-04T18:55:06Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.