Wearable Nanocomposite Sensor System for Motion Phenotyping Chronic Low Back Pain: A BACPAC Technology Research Site.

Chronic low back pain (cLBP) is a prevalent and multifactorial ailment. No single treatment has been shown to dramatically improve outcomes for all cLBP patients, and current techniques of linking a patient with their most effective treatment lack validation. It has long been recognized that spinal pathology alters motion. Therefore, one potential method to identify optimal treatments is to evaluate patient movement patterns (ie, motion-based phenotypes). Biomechanists, physical therapists, and surgeons each utilize a variety of tools and techniques to qualitatively assess movement as a critical element in their treatment paradigms. However, objectively characterizing and communicating this information is challenging due to the lack of economical, objective, and accurate clinical tools. In response to that need, we have developed a wearable array of nanocomposite stretch sensors that accurately capture the lumbar spinal kinematics, the SPINE Sense System. Data collected from this device are used to identify movement-based phenotypes and analyze correlations between spinal kinematics and patient-reported outcomes. The purpose of this paper is twofold: first, to describe the design and validity of the SPINE Sense System; and second, to describe the protocol and data analysis toward the application of this equipment to enhance understanding of the relationship between spinal movement patterns and patient metrics, which will facilitate the identification of optimal treatment paradigms for cLBP.

Accounting for Viscoelasticity When Interpreting Nano-Composite High-Deflection Strain Gauges.

High-deflection strain gauges show potential as economical and user-friendly sensors for capturing large deformations. The interpretation of these sensors is much more complex than that of conventional strain gauges due to the viscoelastic nature of strain gauges. This research endeavor developed and tested a model for interpreting sensor outputs that includes the time-dependent nature of strain gauges. A model that captures the effect of quasi-static strains was determined by using a conventional approach of fitting an equation to observed data. The dynamic relationship between the strain and the resistance was incorporated by superimposing dynamic components onto the quasi-static model to account for spikes in resistances that accompany each change in sensor strain and subsequent exponential decays. It was shown that the model can be calibrated for a given sensor by taking two data points at known strains. The resulting sensor-specific model was able to interpret strain-gauge electrical signals during a cyclical load to predict strain with an average mean absolute error (MAE) of 1.4% strain, and to determine the strain rate with an average MAE of 0.036 mm/s. The resulting model and tuning procedure may be used in a wide range of applications, such as biomechanical monitoring and analysis.