SBIR Phase I: Rapid, pre-symptomatic detection of COVID-19 and unknown viruses using a novel biosensor chip, computer vision and machine learning.

这个小企业创新研究第一阶段项目的更广泛的影响/商业潜力是开发一种新的设备，在出现症状前的环境中检测新冠肺炎、其衍生毒株和新的未知病毒。目前的方法依赖于了解病毒结构和/或产生针对特定已知病毒蛋白的抗体。该项目将使用标准咽喉或鼻拭子以快速护理点形式(20分钟)提供症状前病毒检测，以便先发制人地检测已知和未知来源的新出现的病毒感染。该项目将机器学习与新设备相结合。这可能会缓解社会疏远的担忧，特别是在养老院和学校等特殊环境中。这个小型企业创新研究(SBIR)第一阶段项目使用细胞表型来检测细胞的病毒感染。细胞的表型是特定细胞的形态、功能和其他可观察到的特征，这些特征是由其目前的基因组和蛋白质表达的特定组合产生的。该项目将一种新的计算机视觉算法与光刻制备的、高度特征化的微芯片基板相结合，向细胞提供特定的、可量化的物理和化学线索。这使得能够在体外观察它们的反应，定量地表征细胞表型，并检测由于细胞病毒感染的开始而导致的有意义的细胞行为偏差或异常。这个项目将展示：1)机器视觉图像分析算法提取最相关数据集的能力；以及2)由于感染开始而检测有意义的特征向量的能力。预期的技术成果是通过体外研究来表征细胞表型，展示利用感染细胞表型特征来用传统的光学显微镜模式诊断病毒感染，并描述感染SARS-CoV-2类冠状病毒的细胞系的表型。该奖项反映了NSF的法定使命，并通过使用基金会的智力优势和更广泛的影响审查标准进行评估，被认为值得支持。

The broader impact/commercial potential of this Small Business Innovation Research (SBIR) Phase I project is to develop a novel device to detect COVID-19, its derivative strains, and new, unknown viruses in pre-symptomatic environments. Current methods depend on knowing the viral structure and/or producing antibodies to specific, known viral proteins. This project will provide pre-symptomatic virus detection using a standard throat or nasal swab in a rapid point-of-care format (20 minutes) for preemptive detection of emerging viral infections of both known and unknown origin. The project integrates machine learning with novel devices. This could potentially alleviate social distancing concerns, particularly in special environments, such as nursing homes and schools. This Small Business Innovation Research (SBIR) Phase I project employs cellular phenotyping to detect viral infection of cells. A cell’s phenotype is the particular cell’s morphology, functionality and otherwise observable characteristics which result from the specific combination of its present genomic and protein expression. This project combines a novel computer vision algorithm with lithographically-prepared, highly-characterized microchip substrates that impart specific, quantifiable, physical, and chemical cues to cells. This enables in-vitro observation of their responses and to quantitatively characterize cell phenotype and detection of meaningful cell behavior deviations or anomalies due to the onset of cellular viral infection. This project will demonstrate: 1) the capability of a machine vision image analysis algorithm to extract the most relevant datasets; and 2) the ability to detect meaningful feature vectors due to infection onset. The anticipated technical results are to characterize cell phenotype through in-vitro studies, demonstrate the use of infected cell phenotype features to diagnose viral infections with traditional modes of light microscopy, and describe the phenotyping of cell lines infected with a SARS-CoV-2-like coronavirus.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.