AUTHOR=Orban Mostafa , Guo Kai , Luo Caijun , Yang Hongbo , Badr Karim , Elsamanty Mahmoud TITLE=Development and evaluation of a soft pneumatic muscle for elbow joint rehabilitation JOURNAL=Frontiers in Bioengineering and Biotechnology VOLUME=Volume 12 - 2024 YEAR=2024 URL=https://www.frontiersin.org/journals/bioengineering-and-biotechnology/articles/10.3389/fbioe.2024.1401686 DOI=10.3389/fbioe.2024.1401686 ISSN=2296-4185 ABSTRACT=Elbow joint rehabilitation presents a formidable challenge, underscored by the joint's complex biomechanics and high vulnerability to injuries. Despite the advancements in rehabilitative technology, current solutions such as rigid exoskeletons often fall in provide precision, flexibility, and customization .Although traditional robotic aids, such as rigid exoskeletons, help recover, they lack in providing sufficient flexibility,and easy customization with no need for complicated calculation and complex design considerations. The introduction of soft muscles marks a significant development in the rehabilitation field, offering advantages and challenges compared to conventional rigid systems. These flexible actuators mimic the elasticity of biological tissues, improving safety and interaction between humans and machines. Designed for individualized therapy, its versatility allows application in various rehabilitation scenarios, from clinical settings to home settings. novelty of this approach lies in the development of biomechanically compliant soft pneumatic muscles optimized for precise rotational control of the elbow joint, with an advanced deep learning-based motion tracking system. This design overcomes limitations in force control, stability, and pressure requirements found in exist systems, improving the safety and efficacy of elbow rehabilitation. In this study, the design, fabrication and systematic evaluation of a soft pneumatic muscle for elbow rehabilitation are presented. The device is designed to simulate the complex biomechanical movements of the elbow, with focus on the rotational motions essential for controlling flexion and extension. Through the integration of precise geometric parameters, the actuator is capable of controlled flexion and extension, reflecting elbowkinematics. Employing a rigorous methodology, the research integrates finite element with empirical testing to refine the actuator's performance. Under varying air pressures, the muscle demonstrated deformation along the X axis (10 to 150 mm) and the Y axis, indicative of its symmetrical rotational behavior, while maintaining minimal elongation along the Z axis (0.003 mm max), and proper force under a maximum wight of 470gm. A specialized experimental apparatus comprising a 3D environment, a pneumatic circuit, a LabVIEW-controller , and a deep learning algorithm was developed for accurate position estimation. The algorithm achieved a high accuracy of 99.8% in coordination tracking, indicating precision of the system in monitoring and controlling the actuator's motion.