Novel Methods in Medical Image Computing

Allen Tannenbaum
Comprehensive Cancer Center/Electrical and Computer Engineering, University of Alabama
Wachman 447
Wednesday, February 20, 2013 - 11:00
In this talk, we will describe some novel approaches to medical mage computing, including segmentation and registration. Segmentation is the process of extracting key features from imagery. We will describe statistical methods for doing this, especially the extraction of various tumor types from a number of modalities including MRI and CT. This will also include new methods for white matter tractography. Very importantly, we will describe some ideas from feedback control that may be used to close the loop around and robustify both open-loop segmentation and registration algorithms in
computer vision.
In addition to segmentation, the second key component is registration. Registration is the process of establishing a common geometric reference frame between two or more data sets obtained by
possibly different imaging modalities. The registration problem (especially in the deformable case) is still one of the great challenges in vision and medical image processing. Registration has a substantial literature devoted to it, with numerous approaches ranging from optical flow to computational fluid dynamics. For this purpose, we propose using ideas from optimal mass transport. We will show how the information gleaned from this may be used to drive certain tumor growth models.
We will demonstrate our techniques on a wide variety of data sets from various medical imaging modalities. The talk is designed to be accessible to a broad audience of computer scientists, medical researchers, clinicians, and engineers.