Towards Physical-Interface Pressure Monitoring Outside the Laboratory with Lower-Limb Electronic Clothing and Robust Optimization

Project Summary The overall goal of this project is to develop lower-limb clothing and optimization-based algorithms that merge information from electronic textiles, inertial sensors, portable load cells, and biomechanical models to characterize the pressure sources at the physical interface between residual lower-limbs and prosthetic sockets. The project will enable analysis of pressure for more than 16 continuous hours outside the laboratory with automatic calibration of pressure magnitude and its location relative to the human body. The health outcomes of prostheses and orthoses depend on the physical interface between compliant human tissue and the rigid or padded structures of the device. Quantifying this pressure can inform clinicians on how to prevent ulcers or other side effects of poorly managed physical interfaces, particularly for users with reduced sensation (e.g., due to diabetes or spinal cord injury). However, existing knowledge to relate interface pressure and health outcomes is inconclusive, in part, because 1) available data do not capture the long-term effects of interface pressures and 2) measurements represent mostly laboratory conditions. Pressure recordings depend on expert supervision for sensor calibration and placement — to the extent that there is no user-driven method to accurately measure a complete day of pressure activity. The state-of-the-art focuses on sensor design to improve accuracy and ease of use, but accurate pressure readings are only a portion of the challenge; unsupervised pressure measurements require biomechanical context for calibration and interpretation. For example, pressure on the distal end of a prosthetic socket can increase when there is a transition from swing to stance or a volume change of the residual limb. This project will enable interface pressure and gait-event information for a complete day of activity outside the clinic and the laboratory without expert supervision. The central hypothesis is that, by combining information from inertial sensors, external forces, pressure sensors, and biomechanical models, it is possible to generate accurate user-driven pressure measurements without the need for expert calibration or placement. The specific aims of this project are to 1) create an optimization framework to calibrate pressure sensors with partially unknown placement, 2) develop a clothing paradigm for pressure and inertial measurements in residual limbs, and 3) demonstrate the feasibility of one day of unsupervised pressure measurements outside the laboratory. Successful completion of these specific aims will establish a generalizable wearable sensing platform, which enables the analysis of long-term body exposure to lower-limb device pressures. The project will apply existing knowledge in biomechanics, robotics, mechanics and materials for stretchable e-textiles, and robust optimization to accomplish these aims. Subsequent studies might focus on reducing the risks associated with asymmetric loading through pressure monitoring, or the design of feedback control laws for lower-limb exosuits that maximize comfort through pressure feedback.