SCH: Al-driven Flexible Electronics for Cardiac Organoid Maturation

控制和监测人诱导多能干细胞(HiPSC)来源的成熟的能力 组织是组织工程、再生医学、药理学和合成生物学的关键。这 提案提出了一个由人工智能(Al)驱动的、集成了类似组织的“半机械组织”平台 灵活的电子传感器和执行器，具有正在开发的组织，并提供多模式记录和 控制力。将建立基于机器学习的数学模型来整合数据和组织 成熟状态通过原位单细胞RNA测序读出。这个闭环系统将 控制组织范围内的分布式电动驱动，以促进组织发育。其目的是使用 HiPSC衍生的心脏器官作为一个模型系统来证明这个铝驱动的半机械组织 Platform可促进患者特异性HiPSC来源组织的成熟并消除变异 样本。 具体地说，柔性和可伸展的网状纳米电子设备具有微型多功能传感器和 电刺激器将被完全植入、集成并分布在整个三维 (3D)用于连续、多路传感和致动的有机物体积。此外，现场 电子测序将用于将空间分辨的单细胞分子表型与 电子设备的功能读数。将开发用于建模的统计学习体系结构， 测试和解释多模式电活动、机械收缩、基因调控和 决定有机体功能成熟度的信号网络。最后，反馈控制 系统将实现实时的实验设计增强、电刺激 优化，模型改进，以提高心脏器官的功能成熟。 这项工作的成功可能会提供一个更好的机制理解如何遗传， 分子、电和机械过程调节HiPSC来源的心脏的成熟 有机化合物和建立铝控制的生物电子系统来感知和控制功能 用于各种再生医学和药理学应用的HiPSC来源的心脏器官类化合物的成熟。这项技术很可能是可推广的，以帮助科学家了解成熟度和 几乎任何一种发育中的组织和器官系统，甚至是活体系统的功能。这 拟议的研究将结合人工智能、机器学习、计算生物学、生物医学信息学和 多模式细胞数据，以推进干细胞成熟并实现新的数据驱动的发现，这与 国家医学图书馆的使命。

The ability to control and monitor the maturation of human-induced pluripotent stem cell (hiPSC)-derived tissues is critical for tissue engineering, regenerative medicine, pharmacology, and synthetic biology. This proposal presents an artificial intelligence (Al)-driven "cyborg tissue" platform that integrates tissue-like flexible electronic sensors and actuators with developing tissues and provides multimodal recording and control. Machine learning-based mathematical models will be built to integrate the data and tissue maturation status readout through the in situ single-cell RNA sequencing. This closed-loop system will control the tissue-wide distributed electrical actuations to promote tissue development. The aim is to use hiPSC-derived cardiac organoids as a model system to demonstrate that this Al-driven cyborg tissue platform can improve the maturation and eliminate the variations in patient-specific hiPSC-derived tissue samples. Specifically, flexible and stretchable mesh nanoelectronics with miniaturized multifunctional sensors and electrical stimulators will be fully implanted, integrated, and distributed across the entire three-dimensional (3D) volume of organoids for continuous, multiplexed sensing and actuation. Additionally, in situ electro-sequencing will be used to combine spatially resolved single-cell molecular phenotypes with the functional readouts from the electronics. A statistical learning architecture will be developed for modeling, testing, and interpreting multimodal electrical activities, mechanical contractile, gene regulatory, and signaling networks to determine the functional maturation of the organoids. Finally, a feedback control system will be implemented for real-time experimental design enhancement, electrical stimulation optimization, and model refinement to improve the functional maturation of cardiac organoids. The success of this work will potentially provide an improved mechanistic understanding of how genetic, molecular, electrical, and mechanical processes regulate the maturation of the hiPSC-derived cardiac organoids and establish an Al-controlled bioelectronics system to sense and control the functional maturation of hiPSC-derived cardiac organoids for various regenerative medicine and pharmacological applications. The technology is likely to be generalizable to help scientists understand the maturation and functions of virtually any kind of developing tissue and organoid systems and even in vivo systems. This proposed research will combine AI, machine learning, computational biology, biomedical informatics and multimodal cell data to advance stem cell maturation and enable new data-driven discovery, which aligns with the mission of the National Library of Medicine.