Design of ultrasound transducer arrays for medical imaging

Two emerging areas of development within medical ultrasound are real-time 3-D imaging and high-frequency imaging (>30 MHz). Advances in both areas have been limited by difficulties in developing transducer arrays that provide the imaging performance required for these applications. Arrays are required for high-frequency imaging to overcome the trade off between resolution and depth of field in the current single-element high-frequency transducers. The problem encountered when developing arrays for high-frequency imaging is the small size of the elements (30 mum for a 50 MHz linear array), and the corresponding difficulty in machining kerfs to separate the array elements. Two-dimensional arrays are required for real-time 3-D imaging to electronically steer and focus an ultrasound beam throughout a volume. The problem encountered with developing 2-D (grid) arrays is the huge number of elements in the array, and the large electrical impedance of the elements due to the small element area.;The purpose of this thesis is to investigate and evaluate novel designs of arrays for high-frequency imaging and real-time 3-D imaging that overcome the problems associated with conventional designs of arrays. The designs are investigated using theoretical models based on array geometry and realistic finite-element models to predict the imaging performance of the arrays. The design of kerfless high-frequency annular arrays and linear arrays is investigated. The elements in the kerfless arrays are defined by an electrode pattern on the surface of the piezoelectric ceramic instead of by machined kerfs. The design of a new type of array, called a crossed-electrode array, that can image a volume of tissue in real time is also investigated. The array has the same imaging performance as a conventional 2-D array, but requires far fewer elements, and much larger elements. It is feasible to develop these arrays because the designs have been simplified compared to conventional arrays by changing, in the case of the high-frequency arrays, how the array is fabricated and, in the case of the two-dimensional array, the pattern of the array elements. The kerfless high-frequency arrays and the crossed-electrode array are shown to have comparable image quality to arrays with conventional designs.