SUMMARY

Prolonged exposure to repetitive lifting and twisting under mechanical loading can lead to strain in soft tissues and degradation in bones, especially at the joints. Thus, leaving workers prone to prominent chronic ailments such as lower back pain and osteoarthritis which continue to plague the workforce, with about 50% of reported injuries stemming from the back or knee. Exoskeletons have been shown to reduce net joint loading from external forces and muscle activity in manual tasks. Additionally, wearable sensors (i.e., IMUs, insoles, or EMG) have been shown to be effective at estimating various biomechanical properties such as joint kinetics. Therefore, there is a critical gap to not only understand how internal joint loading manifests across various tasks, but also how wearable technology may affect and inform acute and chronic loading inside of the joint. My central hypotheses were that 1. exoskeletons can offload the biological muscle and joint effort to reduce joint contact forces in manual tasks, and 2. machine learning technologies informed by wearable sensing systems can reliably estimate joint contact forces across manual and dynamic tasks. I leveraged EMG to measure muscle activity and neuromusculoskeletal modeling tools to estimate joint contact forces without exoskeleton assistance and with active knee and passive back exoskeletons. Additionally, I used information from IMUs, insoles, and EMG to estimate joint contact forces across various tasks using machine learning. The outcome from these aims provides better understanding of how the use of wearable technologies can modify and inform internal joint loading. Completion of the aims opens the door for future studies to explore how intervention technologies affect external and internal biomechanical and physiological states. Additionally, this work can influence the future design of exoskeletons, controllers, and real-time biofeedback systems.