An in-silico study of the effect of non-linear skin dynamics on skin-mounted accelerometer inference of skull motion. (September 2021)

• Record Type: Journal Article
• Authors: Wright, Frederick; Docherty, Paul D.; Williams, Elisabeth; Greybe, Desney; Arora, Hari; Kabaliuk, Natalia
• Is Part Of: Biomedical signal processing and control. Volume 70(2021)
• Publication Date: 2021-09
• Subjects: Soft tissue artefact -- Skin modelling -- Head impact -- Skin-mounted sensor -- Accelerometer; Signal processing -- Periodicals; Biomedical engineering -- Periodicals; Signal Processing, Computer-Assisted -- Periodicals; Image Processing, Computer-Assisted -- Periodicals; Biomedical Engineering -- Periodicals; 610.28
• DOI: 10.1016/j.bspc.2021.102986
• Languages: English
• ISSNs: 1746-8094

Abstract

Highlights: An accurate, nonlinear, second-order skull-skin-sensor system model was developed. A linear comparator assessed the effects of soft tissue artefact in impact response. The non-linear model drastically overestimated head acceleration during impact. Thus, skin mounted sensors are likely to overestimate acceleration of heavy impacts. Abstract: Accurate and precise analysis of head impact telemetry data is important for development of biomechanical models and methodologies to decrease the risk of traumatic brain injury. Systematic review suggests that much existing data lacks verification. Soft tissue artefact is a common problem that is not frequently addressed. This paper outlines a method of modelling the coupled, non-linear, skull-skin-sensor system. The model is based on a second order underdamped spring mass damper system that incorporates non-linear values to account for the complex dynamic nature of skin. MATLAB was used to simulate the estimated movement of a sensor mounted to the skin relative to measurements collected via a mouthguard sensor. The non-linear elastic and damping models were developed from descriptions in literature. The model assumed a sensor of 8 g, mounted behind the ear. Results were compared to a typical linear system. In small impacts, the linear and non-linear models provided similar accelerations to the skull. However, in large impacts, the acceleration of the sensor was estimated to be 158% greater than the skull acceleration when modelled non-linearly, while a linear model showed only a 0.7% increase. This implies that for small impacts, the nonlinearity of skin-skull dynamics is not an important characteristic for modelling. However, in large impacts, the non-linearity of the skin-skull dynamic can lead to drastic over-estimates of skull acceleration when using skin mounted accelerometers. In Brief: The system of an accelerometric sensor mounted to the skin is often used to assess the accelerations experienced by athletes during impacts. By developing a non-linear model of the skull-skin-sensor model, the importance of accounting for the non-linear nature of skin while taking telemetry measurements of head acceleration during impact is demonstrated.