Three-dimensional ultrasound imaging and its applications in combined near-infrared optical imaging and co-registered photoacoustic imaging

Date of Completion

January 2007


Engineering, Biomedical




This dissertation focuses on three-dimensional (3-D) ultrasound and its applications as an adjunct modality for biomedical optical imaging. ^ Near infrared diffusive optical tomography (NIR-DOT) has demonstrated outstanding functional imaging potential, especially in breast cancer detection. Due to the high scattering of light in tissue, however, it suffers from poor spatial localization. Ultrasound has the capability to provide accurate geometric information and excellent spatial resolution. We have therefore combined NIR and ultrasound imaging to take advantage of both methodologies. A 64-channel linear array ultrasound imaging system has been developed and integrated with a NIR-DOT system to provide registered a priori lesion information of lesion to guide the optical reconstruction. Clinical applications benefit from three-dimensional localization of tumors and for this purpose a 1280-channel volumetric-scanning ultrasound system has been developed using a 1.75D array. 1.75D acoustic arrays are well suited for this application due to their ability to scan volumes rapidly and accurately while maintaining a reasonable system complexity and cost. Simulations of the system performance in terms of resolution, transmission/reception patterns, and field of view have been performed and confirmed with experimental measurements. ^ Photoacoustic (PAT) imaging is another promising non-invasive imaging technology due to its ability to combine the enhanced contrast of optical absorption with the spatial resolution of acoustic imaging. Co-registration of ultrasound and photoacoustic images takes advantage of both modalities to allow simultaneous visualization of tissue structural and functional information. The developed 1.75D ultrasound system has been characterized for photoacoustic operation and compared with theoretical and simulation predictions. Co-registered images of various phantom targets have been successfully obtained, demonstrating excellent potential for clinical applications. ^