A Miniaturized Flexible Surface Electromyography Sensor With an Integrated Localization Concept

Recent developments in sensor technology have allowed researchers to investigate the potential of using human movement measurements for unobtrusive diagnosis and monitoring of prevalent neuromuscular disorders such as Parkinson’s disease (PD) or joint osteoarthritis. Disease diagnosis and monitoring require both information about the movement itself (kinematics) and the muscle forces that drive it (kinetics). Although the movement reflects the impairment, the forces reveal its underlying cause <xref ref-type="bibr" rid="ref1" xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">[1]</xref>. The millimeter-level tracking accuracy achieved in our previous work within the Collaborative Research Center (CRC) “Empathokinaesthetic Sensor Technology—Sensor Techniques and Data Analysis Methods for Empathokinaesthetic Modeling and Condition monitoring” (CRC 1483 “EmpkinS”) <xref ref-type="bibr" rid="ref2" xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">[2]</xref> enables the detection of small movement differences. Additionally, recent advances in optimal control <xref ref-type="bibr" rid="ref3" xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">[3]</xref>, <xref ref-type="bibr" rid="ref4" xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">[4]</xref>, <xref ref-type="bibr" rid="ref5" xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">[5]</xref> and machine learning <xref ref-type="bibr" rid="ref6" xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">[6]</xref> have enabled the accurate estimation of kinematic and kinetic movement parameters outside of laboratory environments and across various sensor types. These optimal control problems are solved by minimizing an objective related to energy minimization, which replicates the central nervous system’s approach to movement planning <xref ref-type="bibr" rid="ref7" xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">[7]</xref>. However, this method has been shown to inaccurately predict muscle activity <xref ref-type="bibr" rid="ref8" xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">[8]</xref>. This inaccuracy is thought to arise from the failure to account for uncertainty and randomness as it is known that optimal muscle contraction differs in environments with randomness compared to those without <xref ref-type="bibr" rid="ref9" xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">[9]</xref>. Furthermore, the energy optimality of movement might not be a valid assumption in patients with neuromuscular disease, for example, in PD <xref ref-type="bibr" rid="ref10" xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">[10]</xref>.