Date of Completion

6-7-2013

Embargo Period

6-7-2015

Keywords

Ultrasonography, Photoacoustic Imaging, FPGA Processing, Cancer Detection, Image Recognition

Major Advisor

Quing Zhu

Associate Advisor

Rajeev Bansal

Associate Advisor

John Chandy

Field of Study

Biomedical Engineering

Degree

Doctor of Philosophy

Open Access

Campus Access

Abstract

Combining ultrasound and optical imaging modalities has shown promising results for improving the detection of cancer. Ultrasound maps the anatomical structure while optical imaging modalities can provide contrast related to the vasculature density or tumor angiogenesis. In this work, we developed a new technology that allows live tissue characterization with real-time co-registered ultrasound (US) and photoacoustic (PA) imaging. A field programmable gate array (FPGA) based reconfigurable processor is specifically designed to allow real-time switching between the two modalities by adjusting its structure for optimum performance of each. Furthermore, we investigated various image processing and recognition techniques to improve the probability of detecting malignant ovaries from the co-registered US and PA images. This work also includes theoretical contribution to diffuse optical tomography by introducing a novel idea of estimating closed-form solutions of optical fluence inside a turbid medium using the gradient descent optimization, which can be applied to any boundary shape for any given source location. This numerical approach provides new means of faster imaging reconstruction. The applications include accurate tumor characterization and better tracking of tumor chemotherapy response.

Movie1.wmv (5739 kB)
Movie2.wmv (5747 kB)

Share

COinS