Wearable Data Synthesis for Portable Biomechanics

 Biomechanical analysis and musculoskeletal simulation techniques have been  developed in laboratory settings to help us understand the body and the  mechanisms of motion with enough precision to evaluate health and suggest  rehabilitative treatments that aim to improve musculoskeletal function. To achieve  this same result outside of a laboratory, portable sensors must be made available  that can monitor the motion of the body, kinematics, the contact interactions  between the body and the surrounding environment, kinetics, and the forces  generated by the body that enable locomotion, muscle dynamics. While many  portable sensing options have been developed in recent years, the accuracy of  portable biomechanics monitoring techniques has yet to match the accuracy of  traditional laboratory-based tools and remains insufficient for many rehabilitative  applications. One cause for insufficient accuracy is that portable sensing techniques  are often explored in isolation and unable to overcome their unique limitations.  Another cause is that many possible alternative portable sensing approaches  developed outside of the biomechanics field have yet to be fully investigated for their  potential to serve as biomechanics monitoring tools in combination with existing  portable techniques.  

Here, I show how developments in inertial sensing and computer vision  techniques can be intelligently synthesized using either kinematics equations of  motion or rigid body dynamics equations of motion to enable more accurate portable  predictions of body kinematics than approaches which utilize only inertial sensors or  computer vision (Chapter 2). I also show how nuance exists in the choice of fusion  approach depending on the quality of inertial sensing data and computer vision  estimates. A prominent trade-off exists when adding rigid body dynamics into the  synthesis paradigm, and adding dynamics is helpful so long as the dynamics  equations provide more accurate estimates of angular velocities and accelerations  than inertial sensing data, which is likely to occur during real-world applications due  to soft-tissue motion artifacts.  

Next, I show how capacitive sensing, a sensing technique that has been  understudied in biomechanics, can be adapted for use as a customizable,  v  comfortable, lightweight, and sensitive biomechanics monitoring wearable sensor  that enables muscle-activity measurements with the fidelity of gold-standard  laboratory-based techniques (Chapter 3). Capacitive sensing muscle-activity  measurements can then be synthesized with inertial sensors to enable full-body  kinematics, kinetics, and muscle dynamics predictions with comparable accuracy to  that of marker-based motion capture.   

Altogether, these findings show the importance of extensively validating and  incorporating new sensing approaches into biomechanics monitoring tools that  seamlessly integrate with other sensors to cover their weaknesses. I suggest that  future biomechanics monitoring emphasize more nuanced applications, where  multiple sensing modalities are fused intelligently and optimized for specific  applications to maximize monitoring accuracy and intervention efficacy in each local  domain, rather than sacrificing local accuracy to reach for a one-size-fits-all solution.  In the future, I envision the development of a list of locally optimized biomechanics  monitoring best practices, where specific sensor combinations with precise  placements and parameterized computational algorithms are tuned to maximize  monitoring accuracy for use on specific clinical populations and pathologies. I  believe this more nuanced approach to biomechanics monitoring will enable the  development of the next generation of rehabilitative strategies to improve and  sustain more widespread musculoskeletal well-being.