GPU-Driven Optimization of Web-Based Volume Rendering in Peripheral Artery Disease CT Imaging

This paper presents significant advancements in GPU-driven optimization for web-based volume rendering, which is specifically applied to peripheral artery disease (PAD) CT imaging. The proposed method enhances rendering efficiency and image quality, addressing the critical need for real-time, high-quality visualization of complex anatomical structures of PAD. Key improvements in the preprocessing pipeline, such as efficient chunking of large datasets, contribute to better GPU performance. The results demonstrate a substantial improvement over existing tools, with 95.37% reduction in render time, and 96.87% decrease in GPU memory usage compared with BlueLight, and significant memory optimization over Glance. These enhancements facilitate the detailed and interactive 3D visualization of vascular structures, which is crucial for accurate diagnosis and surgical planning. This paper highlights the potential of the proposed method to transform medical imaging practices, improving clinical outcomes for patients with PAD.