Ultraportable Stroke CT Based on Stationary Carbon Nanotube X-ray Source and Deep Learning Image Formation

点击翻译按钮获取中文摘要

PROJECT SUMMARY The goal of this 12-month SBIR phase 1 project is developing an imaging device that will enable highly efficient and cost-effective stroke imaging for patients suffered trauma events. The expected technical outcome will be a proof-of-concept head CT imaging system with sub- second imaging speed, sub-mSv radiation dose, and high-quality images. Imaging systems with such capabilities will address the current deficiency in timely diagnosis of stroke patients, and benefit society with higher efficiency and lower cost. Such imaging device will make a strong economic impact on the global head CT imaging market, which is estimated to be about $36 billion in the U.S. In a longer term, the novel imaging technology could be translated into markets for security screening, industry inspection, and dental imaging. This project is based on the recent research results by Dr. Cao's team under the support of an NSF CAREER award (PI Dr. Cao, 08/01/2014-07/31/2019, $400,000) and a Dr. Cao's Commonwealth Research Commercialization Fund (CRCF) award (Title: “Computed Tomography Without Moving Parts for Fast and Portable Biomedical Imaging”, 07/01/2017- 06/30/2019, $100,000). The project has a strong footing in intellectual property. The technology is protected by a few patent applications at the Virginia Tech, including US. 62/316649, “Ultrafast Micro-CT for Imaging Free-Moving Animals” (Inventor: Guohua Cao), and PCT/US2013/061049 and US14/429835, “System and Method of Stationary Source Computed Tomography” (Inventor: Guohua Cao, et. al.). In this project, the team will design and build a proof-of-concept head CT platform to test the feasibility of ultraportable and compact head CT based on the stationary carbon nanotube x-ray source design and deep learning image reconstruction algorithm. The expected outcomes from this project include demonstration of the feasibility for ultrafast, low dose, and diagnostic imaging capabilities for intracerebral hemorrhage in head phantoms (ICH), with potential automatic ICH identification through deep learning algorithm. A ready-to- prototype head CT device will be optimized and designed, and a corresponding business plan will be developed.