Imitating Radiological Scrolling: A Global-Local Attention Model for 3D Chest CT Volumes Multi-Label Anomaly Classification
cs.CV
Submitted: 2025-03-26
Updated: 2026-09-26
Comments: 13 pages, 4 figures. Accepted for publication at MIDL 2025
License: http://creativecommons.org/licenses/by/4.0/
The gist: The rapid increase in the number of Computed Tomography (CT) scan examinations has created an urgent need for automated tools, such as organ segmentation, anomaly classification, and report
Terminology
Abstract
The rapid increase in the number of Computed Tomography (CT) scan examinations has created an urgent need for automated tools, such as organ segmentation, anomaly classification, and report generation, to assist radiologists with their growing workload. Multi-label classification of Three-Dimensional (3D) CT scans is a challenging task due to the volumetric nature of the data and the variety of anomalies to be detected. Existing deep learning methods based on Convolutional Neural Networks (CNNs) struggle to capture long-range dependencies effectively, while Vision Transformers require extensive pre-training, posing challenges for practical use. Additionally, these existing methods do not explicitly model the radiologist's navigational behavior while scrolling through CT scan slices, which requires both global context understanding and local detail awareness. In this study, we present CT-Scroll, a novel global-local attention model specifically designed to emulate the scrolling behavior of radiologists during the analysis of 3D CT scans. Our approach is evaluated on two public datasets, demonstrating its efficacy through comprehensive experiments and an ablation study that highlights the contribution of each model component.
Sources
- Layer Normalization
- Longformer: The Long-Document Transformer
- ViT-V-Net: Vision Transformer for Unsupervised Volumetric Medical Image Registration
- Generating Long Sequences with Sparse Transformers
- BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding
- An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale
- ConvFormer: Combining CNN and Transformer for Medical Image Segmentation
- Generalist Foundation Models from a Multimodal Dataset for 3D Computed Tomography
- CT2Rep: Automated Radiology Report Generation for 3D Medical Imaging
- Deep Residual Learning for Image Recognition
- Swin Transformer: Hierarchical Vision Transformer using Shifted Windows
- Effective Approaches to Attention-based Neural Machine Translation
- U-Mamba: Enhancing Long-range Dependency for Biomedical Image Segmentation
- ImageNet Large Scale Visual Recognition Challenge
- A comprehensive overview of deep learning techniques for 3D point cloud classification and semantic segmentation
- Grad-CAM: Visual Explanations from Deep Networks via Gradient-based Localization
- GLU Variants Improve Transformer
- 3D Deep Learning on Medical Images: A Review
- Gemma 2: Improving Open Language Models at a Practical Size
- LLaMA: Open and Efficient Foundation Language Models
Related papers
- Loss Knows Best: Detecting Annotation Errors in Videos via Loss Trajectories
- AnchorWeave: World-Consistent Video Generation with Retrieved Local Spatial Memories
- Benchmarking the Robustness of Foundation Models for Mammography under Domain Shift
- MambaX-Net: Dual-Input Mamba-Enhanced Cross-Attention Network for Longitudinal MRI Segmentation
- TeleOCR: Navigating Document Parsing Across Digital and Camera-Captured Documents
- A Survey on Efficient Vision-Language-Action Models