Cloud-computer-aided diagnostic imaging decision support system

描述（申请人提供）：高性能云计算（HPCC）将高性能计算（HPC）与云计算相结合，为高质量、多维诊断图像的处理和移动显示提供所需的超级计算信息学基础设施。这种基础设施将促进高通量癌症筛查，如虚拟结肠镜检查（VC），也称为计算机断层扫描结肠镜检查（CTC）。本提案的短期目标是开发和验证新型移动下一代hpc - vc决策支持系统（hpc - vc2）的临床效益，该系统集成了(1)用于重建超低剂量（ULD）超分辨率（SR）多能无轻泻剂CTC （ME-lfCTC）图像的高性能迭代重建技术（hpIRT）；(2)用于清除残留粪便物质的高性能计算机辅助肠准备（hpCABP）方案和ME-lfCTC中息肉样和扁平非息肉样病变的高性能计算机辅助检测（hpCADe）方案。具体而言，我们假设hpc - vc2系统将(1)以1 mSy或更低的ULD提供ME-lfCTC检查，(2)对结直肠病变具有较高的检测性能，(3)在5分钟内准备好CTC检查，(4)提高阅读器从移动显示设备上显示的lfCTC图像中检测结直肠病变的性能。具体目标是：(1)开发动态压缩方案，实现CTC原始数据快速上传到HPCC；(2)建立了ULD SR ME-CTC并行hpIRT方法；(3)针对hpc - vc2开发并行hpCABP和hpCADe方案；(4)评价hpc - vc2系统的临床疗效。hpc - vc2系统的成功开发将证明该平台在改善诊断成像和促进准确、高通量结肠癌筛查方面的临床益处，这是患者高度可接受的。从长远来看，广泛采用和使用hpc - vc2系统将有助于早期和准确的诊断，从而降低结肠癌的死亡率，结肠癌目前是美国癌症死亡的第三大原因。

DESCRIPTION (provided by applicant): High-performance cloud computing (HPCC) integrates high-performance computing (HPC) with cloud computing to provide the supercomputing informatics infrastructure needed for the processing and mobile display of high quality, multi-dimensional diagnostic images. Such infrastructure would advance high-throughput cancer screening like virtual colonoscopy (VC), also known as computed tomography colonography (CTC). The short-term goals of this proposal are to develop and validate the clinical benefits of a novel mobile next-generation HPCC-VC decision support system (HPCC-VC2) that integrates (1) a high-performance iterative reconstruction techniques (hpIRT) for reconstructing ultra-low-dose (ULD) super-resolution (SR) multi-energy laxative-free CTC (ME-lfCTC) images, (2) a high-performance computer-assisted bowel preparation (hpCABP) scheme for removal of residual fecal materials and a high-performance computer-aided detection (hpCADe) scheme and detection of polypoid and flat non-polypoid lesions in ME-lfCTC. Specifically, we hypothesize that the HPCC-VC2 system will (1) provide an ME-lfCTC examination at an ULD of 1 mSy or less, (2) achieve high detection performance for colorectal lesions, (3) prepare a CTC examination within 5 minutes, and (4) improve readers' performance in the detection of colorectal lesions from lfCTC images shown on mobile display devices. The specific aims are: (1) develop dynamic compression scheme for rapid uploading of raw CTC data to the HPCC; (2) develop parallel hpIRT method for ULD SR ME-CTC; (3) develop parallel hpCABP and hpCADe schemes for the HPCC-VC2; and (4) evaluate the clinical benefit of the HPCC-VC2 system. Successful development of the HPCC-VC2 system will demonstrate the clinical benefit of the platform for improved diagnostic imaging and facilitation of accurate, high-throughput colon cancer screening that is highly acceptable to patients. In the longer term, broad adoption and use of the HPCC-VC2 system will facilitate early and accurate diagnoses, and thus will reduce mortality from colon cancer that is currently the third leading cause of cancer deaths in the United States.