Accuracy of the resonance ultrasound method in determination of the acoustic phase shifts in plastic and bone

The accuracy of a resonance ultrasound method is determined when used towards resolving the acoustic phase shifts in single- and three-layer plastic plates, and in pig skull samples ex vivo at oblique angles. The investigation is conducted with a single, broadband nonfocused transducer at frequency range from 0.45 to 1.45 MHz while the angle of incidence of the sound beam is varied by rotating the studied plate by 0-15/spl deg/. Acoustic phase shifts are calculated using the resonance frequencies of the plates resolved from the reflected spectrum. These phase shifts are compared to those measured directly with hydrophone. Materials include 20 pieces of acrylic, six pieces of polycarbonate plastic (PC), and two pig's skull bone samples ex vivo. Ten acrylic and three PC plates are bent to model the shape of the human skull. Further, the four thinnest acrylic plates and all PC plates are used in three-layer stacks to model the layered structure of skull bone. Results show that with flat single-layer plates, 70% of the measured phase shifts (of altogether 416) deviate by less than 30/spl deg/ from those measured with the hydrophone; after bending the similar performance is about 61% (of 208), and about 58% (of 192) with the flat three-layer phantoms. Few measurements show deviation of higher than 90/spl deg/ between the methods. With curved three-layer plates and bone samples the discrepancy increased, showing a difference of less than 30/spl deg/ in 23% (of 192) for curved three-layered plates, and in 17% (of 64) for bone samples, respectively. In general the increase in the discrepancy is due to the destruction of the resonance pattern at higher angles of incidence as sound beam is reflected away from the transducer and uncertainty in resolving the resonance frequencies increases.