Adaptive 4D Volume Based Medical Imaging Analysis

Songan Mao, Purdue University

Abstract

Medical image processing is an exciting and dynamic part of computer vision and image processing techniques. The significant growth of applications in the past years, is evident in areas such as healthcare, radiotherapy technologies, and technical diagnostics, which are the developing areas. Medical imaging plays a vital role in acknowledging these features and helps in the assessment of the disease. Four-dimensional computed tomography is one of the most significant innovations represents the next step in medical imaging takes images that show both the location of a tumor and movement in the body. Besides, it reveals the movement other body organs with time. The prevalence of more developed imaging instruments characterized by high resolution and more refined output is one of the motivating factors is medical imaging. I will illustrate my work on the fiducial marker 3D localization by combining the use of kilovoltage (kV) image and Megavoltage (MV) image, and then describe the target tumor volume prediction based on four-dimensional computed tomography in image guided treatment system. Finally, I will introduce a tumor segmentation method and an approach to analyze the correlation between imaging phenotype and genomic information, by combing cancer patient image data and the associated gene expression.

Degree

Ph.D.

Advisors

Chien, Purdue University.

Subject Area

Computer Engineering|Engineering|Medical imaging|Computer science

Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server
.

Share

COinS