Department of Information Management, National Taipei University of Nursing and Health Sciences, Taipei, Taiwan.
Department of Biomedical Engineering, National Yang Ming Chiao Tung University, Taipei, Taiwan.
J Digit Imaging. 2023 Apr;36(2):753-763. doi: 10.1007/s10278-022-00746-0. Epub 2022 Dec 20.
Recently, WebGL has been widely used in numerous web-based medical image viewers to present advanced imaging visualization. However, in the scenario of medical imaging, there are many challenges of computation time and memory consumption that limit the use of advanced image renderings, such as volume rendering and multiplanar reformation/reconstruction, in low-cost mobile devices. In this study, we propose a client-side rendering low-cost computation algorithm for common two- and three-dimensional medical imaging visualization implemented by pure JavaScript. Particularly, we used the functions of cascading style sheet transform and combinate with Digital Imaging and Communications in Medicine (DICOM)-related imaging to replace the application programming interface with high computation to reduce the computation time and save memory consumption while launching medical imaging interpretation on web browsers. The results show the proposed algorithm significantly reduced the consumption of central and graphics processing units on various web browsers. The proposed algorithm was implemented in an open-source web-based DICOM viewer BlueLight; the results show that it has sufficient rendering performance to display 3D medical images with DICOM-compliant annotations and has the ability to connect to image archive via DICOMweb as well.Keywords: WebGL, DICOMweb, Multiplanar reconstruction, Volume rendering, DICOM, JavaScript, Zero-footprint.
最近,WebGL 在众多基于网络的医学图像查看器中得到了广泛应用,用于呈现高级成像可视化效果。然而,在医学成像场景中,由于计算时间和内存消耗方面存在许多挑战,限制了在低成本移动设备中使用高级图像渲染技术,如体绘制和多平面重建/重组。在这项研究中,我们提出了一种客户端渲染的低成本计算算法,用于通过纯 JavaScript 实现常见的二维和三维医学成像可视化。特别是,我们使用级联样式表转换的功能,并结合与医学数字成像和通信(DICOM)相关的成像,以替代具有高计算要求的应用程序编程接口,从而减少计算时间并节省内存消耗,同时在网络浏览器上启动医学图像解释。结果表明,所提出的算法显著降低了各种网络浏览器上中央处理器和图形处理器的消耗。所提出的算法已在开源的基于网络的 DICOM 查看器 BlueLight 中实现;结果表明,它具有足够的渲染性能来显示符合 DICOM 标准的注释的 3D 医学图像,并且能够通过 DICOMweb 连接到图像存档。
WebGL、DICOMweb、多平面重建、体绘制、DICOM、JavaScript、零足迹。