Characterization and evaluation of sparse array transducers for small particle detection

Li, Xiaotong and Gachagan, Anthony and Murray, Paul (2019) Characterization and evaluation of sparse array transducers for small particle detection. In: 2019 IEEE International Ultrasonics Symposium, 2019-10-06 - 2019-10-09, SECC. (

[thumbnail of Li-etal-IUS-2019-Characterization-and-evaluation-of-sparse-array-transducers-for-small-particle-detection]
Text. Filename: Li_etal_IUS_2019_Characterization_and_evaluation_of_sparse_array_transducers_for_small_particle_detection.pdf
Accepted Author Manuscript

Download (529kB)| Preview


Background, Motivation and Objective -- A log spiral transducer array pattern has been designed [1] which aims to reduce the operator dependency of current transcranial ultrasound devices by offering a flexible alternative to conventional transducers. This paper describes the fabrication, characterization and evaluation of two prototype arrays and describes the adopted image processing method to automatically analyze ultrasonic images. Statement of Contribution/Methods -- Two prototype sparse array transducers have been fabricated, one using the fiber Composite Element Composite Array Transducer (CECAT) structure [2], the other using a conventional 1-3 composite (C1-3). Three measurements including the impedance response, inter-element cross-talk and pulse-echo response have been acquired for both transducers to evaluate and compare their performance. Both designs have also been used to acquire ultrasonic Total Focusing Method (TFM) images of a tube submersed in water containing artificial anomalies (solid particles in range 1 mm – 2 mm) to synthesize our intended application of detecting anomalies through ultrasonic transcranial imaging. An algorithm combining difference imaging and the Hough Transform has been designed to detect and estimate the size of these particles in the images acquired using both transducer designs. Results/Discussion -- Impedance response tests show that the averaged coupling coefficient is 0.64 for the fiber CECAT and 0.6 for the C1-3 device, with the CECAT array also demonstrating a 10dB improvement in mechanical cross-talk level between neighboring array elements. As shown in Figure 1, the CECAT transducer has a broader bandwidth than C1-3 (47.44% to 30.95%). Moreover, Figure 2 shows the results of applying our image analysis algorithm on data collected using the CECAT transducer through which two ball bearings (1.5mm; 2mm) have been successfully detected with 3.3% and 0% sizing errors, respectively.