FMSG:BIO: Integrating Artificial Intelligence with Bioprinting for Future Manufacturing of Organoids

人类多能干细胞(HPSC)衍生的有机化合物在疾病治疗和治疗开发中具有很高的价值。将有机化合物转化为实际应用将需要可扩展和可重复使用的方法来生产高质量的有机化合物。该项目将制定策略，利用三维(3D)生物打印可伸缩的有机物制造。机器学习(ML)/人工智能(AI)技术将被开发来从显微镜图像中快速评估有机物的质量。其目标是通过增强用于3D生物打印的ML/AI技术，为大规模制造定制和患者特定的有机体提供一条途径。这项研究将为学员提供多学科技能集，包括生物打印、有机类工程、成像和人工智能方法。此外，该项目将把全球和社会推广方面整合到工程课程中，提供一种全面的教育途径，以培养未来、具有全球意识的美国生物制造劳动力，使其在革命性的制造概念方面处于世界领先地位。这个概念验证未来制造种子赠款(FMSG)项目旨在通过ML/AI辅助的3D生物打印，为有机化合物制造整合有益的机械信号途径。在执行这一方法时，目标是开发非侵入性分析来测量有机物的表型，以满足制造流水线的关键质量属性。中心假设是，通过包裹施加细胞限制将诱导有利的机械转导途径，这将通过使肌动蛋白细丝失活来促进类器官分化。肌动蛋白重塑所产生的细胞骨架修饰可以被用来提取有关细胞特征的信息，以预测表型。这一假设将通过开发以下技术来验证：(I)使用3D生物打印的细胞限制驱动的有机类化合物衍生；以及(Ii)使用人工智能(AI)模型预测细胞骨架修饰的细胞命运，以识别亮场图像中有机类物质的发育特征。这个FMSG项目展示了将人工智能与3D生物打印集成在有机类化合物生成管道中的可行性，长期目标是实现具有严格在线质量控制的可扩展的有机类化合物制造。该项目由化学、生物工程、环境和运输系统司以及生物科学局的工程教育和中心司联合资助。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。

Human pluripotent stem cell (hPSC) derived organoids are highly valuable for disease treatment and therapeutic development. Translating organoids to practical use will require scalable and reproducible methods to produce high-quality organoids. This project will devise strategies exploiting three-dimensional (3D) bioprinting for scalable organoid manufacturing. Machine Learning (ML)/Artificial Intelligence (AI) techniques will be developed to rapidly assess organoid quality from microscopy images. The goal is to offer a pathway for large-scale manufacturing of tailored and patient-specific organoids by augmenting ML/AI techniques for 3D bioprinting. The research will provide trainees with multidisciplinary skill sets, encompassing bioprinting, organoid engineering, imaging, and AI methods. In addition, the project will integrate global and social outreach aspects into engineering curriculum, offering a holistic educational pathway to train the future, globally aware U.S. biomanufacturing workforce to lead the world in revolutionary manufacturing concepts.This proof-of-concept Future Manufacturing Seed Grant (FMSG) project aims to integrate beneficial mechano-signaling pathways for organoid manufacturing via ML/AI aided 3D bioprinting. In executing this approach, the goal is to develop non-invasive assays to measure organoid phenotype to meet the critical quality attributes of the manufacturing pipeline. The central hypothesis is that imposing cell confinement via encapsulation will induce favorable mechano-transduction pathways, which will enhance organoid differentiation by deactivating actin filaments. The resulting cytoskeletal modification from actin remodeling can be exploited to extract information about cell characteristics to predict phenotype. This hypothesis will be tested by developing techniques for (i) cell-confinement driven organoid derivation using 3D bioprinting; and (ii) predicting cell fate from cytoskeletal modifications using Artificial Intelligence (AI) models to identify features of organoid development in bright-field images. This FMSG project demonstrate the feasibility of integrating AI with 3D bioprinting in the organoid generation pipeline, with the long-term goal of enabling scalable manufacturing of organoids with stringent in-line quality control.This project is jointly funded by the Division of Chemical, Bioengineering, Environmental, and Transport Systems and the Division of Engineering Education and Centers in the Directorate for Engineering and the Division of Molecular and Cellular Biosciences in the Directorate for Biological Sciences.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.