Imitating Radiological Scrolling: A Glocal-Lobal Attention Model for 3D Chest CT Volumes Multi-Label Anomaly Classification
Keywords: 3D Computed Tomography, Multi-label classification, Attention Mechanism, Convolutional Neural Network
TL;DR: We propose CT-Scroll, a global-local attention model for 3D CT scans that emulates radiologists’ scrolling-based navigation along axial slices for multi-label anomaly classification.
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 classification methods, relying on standard Convolutional Neural Networks or Vision Transformers, do not explicitly model the radiologist's navigational behavior while scrolling through CT scan slices. In this study, we present CT-Scroll, a novel glocal-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.
Primary Subject Area: Detection and Diagnosis
Secondary Subject Area: Application: Radiology
Paper Type: Methodological Development
Registration Requirement: Yes
Submission Number: 113
Loading