Statistical classification of scatterer ultrasonic signature in bio-mechanical media and spatial localization of digital data using nonlinear time reversal

Abstract

Various nonlinear methods have been investigated in recent years offering improvements of over conventional B-mode in medical imaging and non-destructive testing (NDT) [1], including Nonlinear Elastic Wave Spectroscopy (NEWS) methods. Furthermore, new optimized excitations such as Nonlinear Time Reversal (TR) based methods are classified within TR-NEWS methods [2]. In this context, one of the strategic plan of the international NDT community is to define standards for developing advanced and digitalized automated NDT set-up in modern mass production, defining the concept of NDT 4.0 or NDE 4.0 [3]. In the typical NDE4.0 TR-NEWS equipment, an optimized chirp-coded signal in the range of 0.6–3 MHz is used as a compression coding. The signal classification is performed using the Fuzzy Classification (FC) method and the Divergence Decision Tree (DDT) algorithm using specific $\phi$-divergence spectral measures extracted from the received ultrasonic response containing acoustic nonlinearities. Finally, experimental setup and scatterer classification procedures have been further validated in ondotology with tooth echodentographic experiment where the ultrasonic responses to chirp-coded signals were monitored with a standard Polytec vibrometer, showing the efficiency of TR-NEWS methods for understanding of ultrasonic wave propagation in complex media.