Wavelet Analysis of Head Acceleration Response Under Dirac Excitation for Early Oedema Detection

Abstract

The present work deals with the application of an innovative in-house developed wavelet-based methodology for the analysis of the acceleration responses of a human head complex model as a simulated diffused oedema progresses. The human head complex has been modeled as a structure consisting of three confocal prolate spheroids, whereas the three defined regions by the system of spheroids, from the outside to the inside, represent the scull, the region of cerebrospinal fluid, and the brain tissue. A Dirac-like pulse has been used to excite the human head complex model and the acceleration response of the system has been calculated and analyzed via the wavelet-based methodology. For the purpose of the present analysis, a wave propagation commercial finite element code, LS-DYNA 3D, has been used. The progressive diffused oedema was modeled via consecutive increases in brain volume accompanied by a decrease in brain density. It was shown that even a small increase in brain volume (at the level of 0.5%) can be identified by the effect it has on the vibration characteristics of the human head complex. More precisely, it was found that for some of the wavelet decomposition levels, the energy content changes monotonically as the brain volume increases, thus providing a useful index of monitoring an oncoming brain oedema before any brain damage appears due to uncontrolled intracranial hypertension. For the purpose of the present work and for the levels of brain volume increase considered in the present analysis, no pressure increase was assumed into the cranial vault and, associatively, no brain compliance variation.

The combined use of vibration, acoustic emission and oil debris on-line monitoring towards a more effective condition monitoring of rotating machinery

“…The DWT-based methodology used in this work was introduced and described elsewhere [23] by the authors. Fig.…”

Fatigue damage monitoring in carbon fiber reinforced polymers using the acousto‐ultrasonics technique

“…The same procedure can be repeated and the same set of filters can be applied to the approximation at every level for the desirable number of analysis levels. The discrete wavelet‐based signal processing methodology, as developed in‐house [25, 26] is depicted schematically in Fig. 3.…”

“…Thus 11 parameters (since a 10‐level decomposition was used) namely ED1‐ED10 (for the details) and EA10 (for the approximation) were resulted. The methodology was introduced and applied in [25, 26] and realized in Matlab workspace.…”