SBIR Phase I: Machine Learning for Early Detection of COVID-19 Plaques in Cells

这一小型企业创新研究(SBIR)第一阶段项目的更广泛影响/商业潜力是利用人工智能(AI)技术加快疫苗和抗病毒药物的开发。该项目将开发技术，以便在人工检测病毒感染细胞的变化之前几天或几周检测到这些变化。这将加速研究新的抗病毒化合物，以表征其对快速变异的病毒株的有效性，如流感和SARS-CoV-2。这将影响新冠肺炎的研究和一般病毒学。这个SBIR第一阶段项目将研究人工智能技术，以加快斑块分析中抗病毒剂的测试，用于疫苗和抗病毒药物的开发。这些检测方法通过观察感染对敏感细胞培养的影响来测量样本中具有感染性的病毒颗粒的数量。目前，化验需要2-14天，因为需要几轮感染才能确保准确的读数。该项目将推进人工智能技术，在显微镜图像中自动检测受感染的细胞，而无需人工干预或耗时的准备，从而提高这些分析的吞吐量。为了实现这一目标，该项目将：1)收集受感染细胞培养物的显微镜图像的时间进程，用于训练人工智能模型以自动测量大型细胞培养板上的病毒感染；2)调查显微镜图像获取方法是否易于整合到现有工作流程和图像质量；3)评估各种人工智能技术的适用性；4)确定检测准确性并将其与传统分析进行比较。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 accelerate the development of vaccines and anti-virals with artificial intelligence (AI) techniques. This project will develop technology to detect changes in virus-infected cells days or weeks before they can be detected manually. This will accelerate studies for novel anti-viral compounds characterizing their effectiveness on rapidly mutating viral strains, such as influenza and SARS-CoV-2. This will impact COVID-19 research and general virology.This SBIR Phase I project will investigate AI techniques to accelerate testing of anti-viral agents in plaque assays for the development of vaccines and anti-virals. These assays measure the number of infectious viral particles in a sample by observing the effects of infection on a culture of susceptible cells. Currently, the assay takes 2-14 days because several rounds of infection are necessary to ensure an accurate reading. This project will advance AI techniques to automatically detect infected cells in microscopy images without human intervention or time-consuming preparations, thereby increasing the throughput for these assays. To achieve this goal, this project will: 1) Collect a time-course of microscopy images of infected cell cultures for training an AI model to measure virus infections automatically on large cell culture plates; 2) Investigate microscopy image acquisition approaches with respect to ease of integration in existing workflows and image quality; 3) Evaluate the suitability of various AI techniques; 4) Determine the detection accuracy and compare it with traditional assays.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.