EAGER: Establishing Near-Ultraviolet Coherent Anti-Stokes Raman Scattering Microscopy for Highly Sensitive Imaging of Native Biomolecules

在脑肿瘤切除过程中，经常咨询病理学家来帮助指导手术决定。然而，目前的外科病理程序是费时费力的，包括组织冷冻切片和化学染色。相干拉曼显微镜是一种很有前途的技术，可以提供无标记的病理样组织图像。当前的相干拉曼成像使用可见光或近红外（NIR）光，并且在成像细胞核中的天然DNA分子时灵敏度有限。该项目将利用近紫外（NUV）光和相干反斯托克斯拉曼散射（CARS）成像来突破灵敏度障碍。NUV-CARS成像系统将用于开发脑肿瘤诊断的无标签数字病理学。这项技术具有推进外科病理学和图像引导手术领域的潜力。该项目将为不同背景的本科生和研究生提供生物光子学和数据科学方面的培训机会。这个EAGER项目的目标是建立NUV-CARS显微镜，用于高分辨率、高灵敏度的天然生物分子成像，并应用这种成像工具开发无标签的数字病理学，用于脑肿瘤诊断。虽然NIR-CARS能够对活细胞和组织进行快速化学成像，但在对天然生物分子进行成像时，其检测灵敏度较低。NUV-CARS将利用电预共振效应，显著提高对天然生物分子化学键成像的灵敏度。本文将采用石英光学的非线性光学策略来控制非谐振背景和紫外光损伤。NUV-CARS可以获得DNA分子的真实化学对比，绘制出与苏木精和伊红（H&amp；E）染色相似的病理图像，用于基于新鲜组织的脑肿瘤诊断。该研究将推动NUV相干拉曼和多光子荧光领域的知识，为科学知识和技术应用的突破性进展铺平道路。结合机器学习算法的NUV-CARS成像将为脑肿瘤诊断提供新的基于组织的数字病理学，这可以很容易地应用于其他类型的癌症手术。该奖项反映了美国国家科学基金会的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

During brain tumor resection, pathologists are often consulted to help guide surgical decisions. However, current surgical pathology procedures are time-consuming and labor-intensive, involving tissue cryosectioning and chemical staining. Coherent Raman microscopy is a promising technology to render label-free pathology-like images of tissue. Current coherent Raman imaging uses visible or near-infrared (NIR) light and has limited sensitivity when imaging native DNA molecules in the nucleus of a cell. This project will leverage near-ultraviolet (NUV) light and coherent anti-Stokes Raman scattering (CARS) imaging to break the sensitivity hurdle. The NUV-CARS imaging system will be applied to develop label-free digital pathology for brain tumor diagnosis. This technology holds the potential to advance the fields of surgical pathology and image-guided surgery. This project will provide training opportunities in biophotonics and data science for undergraduate and graduate students with diverse backgrounds. The goal of this EAGER project is to establish NUV-CARS microscopy for high-resolution, highly sensitive imaging of native biomolecules and apply this imaging tool to develop label-free digital pathology for brain tumor diagnosis. While NIR-CARS is able to perform rapid chemical imaging of live cells and tissue, it suffers from a low detection sensitivity when imaging native biomolecules. NUV-CARS will leverage the electric pre-resonance effect to achieve a significantly enhanced sensitivity for imaging chemical bonds in native biological molecules. Novel nonlinear optical strategies with quartz optics will be implemented to manage the non-resonant background and UV photodamage. With NUV-CARS, the true chemical contrast of DNA molecules will be obtained to render pathological images similar to the images with hematoxylin and eosin (H&E) staining, which will be used for fresh tissue-based brain tumor diagnosis. This research will advance the knowledge in the fields of NUV coherent Raman and multiphoton fluorescence, paving the way for groundbreaking advancements in scientific knowledge and technological applications. NUV-CARS imaging with machine learning algorithms will enable fresh tissue-based digital pathology for brain tumor diagnosis, which can be readily applied to other types of cancer surgeries.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.