Publishing Service

Polishing & Checking

Journal of Zhejiang University SCIENCE B

ISSN 1673-1581(Print), 1862-1783(Online), Monthly

Single-cell RNA-sequencing-guided reactive oxygen species-scavenging hydrogel design for regeneration of osteoporotic bone

Abstract: The pathological microenvironment of osteoporosis poses a substantial clinical challenge for bone defect regeneration. Through single-cell RNA-sequencing (scRNA-seq) analysis, we identified a reactive oxygen species (ROS)-overloading osteoblast subpopulation as a critical pathological feature of osteoporotic niches. Guided by scRNA-seq analysis, we engineered a microenvironment-adaptive hydrogel system through precise integration of antioxidant curcumin-encapsulated zeolitic imidazolate framework-8 nanoparticles (CCM@ZIF-8 NPs) within photo-crosslinkable alginate methacrylate (AlgMA) hydrogel (AlgMA/CCM@ZIF-8). This engineered design exhibited dual functions: effectively scavenging ROS in bone marrow-derived mesenchymal stem cells (BMSCs) while simultaneously suppressing osteoclast differentiation. The osteo-regenerative superiority of the AlgMA/CCM@ZIF-8 nanocomposite hydrogel was conclusively demonstrated in bone defect models of osteoporotic mice. This scRNA-seq-informed engineering strategy offers a promising approach for developing pathophysiology-adapted biomaterials to promote regenerative repair in osteoporotic bone defects.

Key words: Osteoporosis; Single-cell RNA-sequencing (scRNA-seq); Reactive oxygen species (ROS); Curcumin (CCM); Bone regeneration

Chinese Summary  <53> 考虑电机与外力耦合的机器人动力学预测

物理信息神经网络
孙丰雨,吴双双,李志明,熊沛霖,陈雯柏
北京信息科技大学自动化学院,中国北京市,100192
摘要:近年来,物理信息神经网络(PINN)在刚体动力学保守系统建模中展现出显著潜力。然而,现有PINN框架在应用于机械臂实际交互任务(如零件装配和医疗操作)时,因缺乏有效的外部作用力建模机制,导致其在动态交互场景中的预测精度显著下降。此外,由于工业机器人(包括UR5和UR10e等型号)通常未配备关节扭矩传感器,获取精确动力学训练数据仍具挑战。为此,本研究提出两种融合电机动力学与外部作用力建模的增强型PINN模型。首先,引入两种数据驱动的雅可比矩阵估计方法以嵌入外部作用力:其一通过学习末端执行器速度与关节速度间映射关系以近似雅可比矩阵;其二先学习系统运动学行为,再通过正向运动学模型解析微分推导雅可比矩阵。其次,将电流-扭矩映射作为物理先验知识嵌入模型,以建立系统运动状态与电机电流的直接关联。在两种不同机械臂上的实验结果表明,所提模型皆无需关节扭矩传感器即可在复杂外部作用力场景下实现高精度扭矩估计。与现有先进方法相比,所提模型在多种复杂场景下整体建模精度平均提高31.12%与37.07%,同时关节轨迹跟踪误差分别降低40.31%与51.79%。

关键词组:动力学建模;物理信息神经网络;电机动力学;外力建模;运动学


Share this article to: More

Go to Contents

References:

<Show All>

Open peer comments: Debate/Discuss/Question/Opinion

<1>

Please provide your name, email address and a comment





DOI:

10.1631/jzus.B2500254

CLC number:

Download Full Text:

Click Here

Downloaded:

687

Download summary:

<Click Here> 

Downloaded:

325

Clicked:

759

Cited:

0

On-line Access:

2025-12-31

Received:

2025-05-14

Revision Accepted:

2025-09-21

Crosschecked:

2025-12-31

Journal of Zhejiang University-SCIENCE, 38 Zheda Road, Hangzhou 310027, China
Tel: +86-571-87952276; Fax: +86-571-87952331; E-mail: jzus@zju.edu.cn
Copyright © 2000~ Journal of Zhejiang University-SCIENCE