Title: "3D Medical Volume Reconstruction: Complexity and Medical Community Infrastructure Support"

Slide set [PDF]

Abstract: We address the problem of optimal registration decisions during 3D medical volume reconstruction and their impact on (a) anticipated accuracy of aligned images, (b) uncertainty of obtained results, (c) repeatability of alignment, and (d) computational requirements. The registration decisions include (1) image size used for registration, (2) transformation model, (3) invariant registration feature (intensity or morphology), (4) automation level, (5) evaluations of registration results (multiple metrics and methods for establishing ground truth), and (6) assessment of resources (geographically local or distributed computational resources and human expertise). Our goal is to provide data-driven mechanisms for evaluating the tradeoffs between accuracy of 3D volume reconstructions and registration variables. First, we present links between registration decisions and 3D reconstruction results in terms of accuracy, uncertainty, consistency and computational complexity characteristics. Second, we have built software tools that enable geographically distributed researchers to optimize their data-driven registration decisions by using web services and high performance computing (HPC) resources. The support developed for registration decisions about 3D volume reconstruction is available to the general community with the access to the NCSA HPC resources. Next, we illustrate the performance of our registration decision support system by considering 3D volume reconstruction of blood vessels in histological sections of uveal melanoma from serial fluorescent labeled paraffin sections labeled with antibodies to CD34 and laminin. The specimens are studied by fluorescence confocal laser scanning microscopy (CLSM) images. Finally, we discuss the complexity of building a web-enabled, web services based, data-driven, registration decision support system for 3D volume reconstruction.

Background: The content of the talk will be based on five journal papers (Journal of Microscopy, International Journal of Web Services Research, and EURASIP Journal on Applied Signal Processing) and four conference papers (SPIE on Medical Imaging and IEEE International Conference on Web Services). The talk will provide an overview of our three years of NIH funded work jointly with UIC.

Bio: Peter Bajcsy is a research scientist at NCSA, UIUC, and an adjunct assistant professor in the CS and ECE departments at the University of Illinois. He is interested in X-informatics problems, where X stands for biology, medical, health care, hydrology, sensor and instrumentation domain specific information processing issues (see http://www.ncsa.uiuc.edu/people/pbajcsy/).