SBIR Phase I: Development of AI Software to Capture and Identify Circulating Rare Cells in Lung Patients

这个SBIR第一期NSF项目的更广泛的影响/商业潜力是开发人工智能(AI)软件，以高效和准确地识别循环中的肺癌相关细胞。据估计，2020年美国将有超过20万新的肺癌病例，推动成本超过1660亿美元。目前的护理标准要求密切监测这些患者，每6周进行一次胸部计算机断层扫描(CT)。如果患者的癌症被确定为晚期，也会同时接受骨盆CT扫描。这项拟议的技术将为临床医生提供额外的数据，以便在临床实验室环境中通过简单的抽血来早期检测肺癌，以便立即向患者和临床医生反馈，从而避免更具侵入性的程序和辐射暴露。拟议的项目将推动液体活检技术在研发临床环境中的应用。该项目新颖的成像系统识别荧光肿瘤来源细胞的能力将为检测早期疾病和区分惰性肺癌和侵袭性肺癌提供更灵敏和可靠的方法，并有可能进一步整合到肺癌筛查项目中。利用先进的人工智能(AI)算法和世界级的光学免疫荧光检测方法，该项目的荧光显微镜将是一个人工智能驱动的图像处理系统。该项目通过结合高分辨率多通道光学成像、专有荧光标记和分析以及最先进的人工智能分割和分类技术，为在临床环境中检测低水平的稀有细胞提供了前所未有的解决方案。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。

This broader impacts/commercial potential of this SBIR Phase I NSF project is to develop Artificial Intelligence (AI) software to identify circulating lung cancer related cells efficiently and accurately. It is estimated that there will be over 200,000 new cases of lung cancer in the US in 2020, driving the cost above $166 billion. The current standard of care requires close monitoring of these patients, with chest Computed Tomography (CT) scans taken every 6 weeks. Patients also undergo pelvis CT scans concurrently if their cancer is determined to be at later stages. The proposed technology will provide the clinician additional data for early detection of lung cancer with a simple blood draw in a clinical laboratory setting for immediate feedback to the patient and clinician, thus avoiding more invasive procedures and radiation exposures. The proposed project will advance liquid biopsy techniques in R&D clinical settings. This project’s novel imaging system’s ability to identify the fluorescent tumor-derived cells will provide a more sensitive and reliable methodology to detect early-stage disease and differentiate indolent from aggressive lung cancer, with further potential to be integrated into lung cancer screening programs. Utilizing advanced Artificial Intelligence (AI) algorithms and world-class optical immunofluorescent detection methods, this project’s fluorescent microscope will be an AI-driven image processing system. This project provides an unprecedented solution for detecting low levels of rare cells in a clinical setting through the combination of high resolution multichannel optical imaging, proprietary fluorescent taggants and assays, and state-of-the-art AI segmentation and classification techniques.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.