Computational Core

Dartmouth College, Hanover NH

Investigators

Linked publications & trials

Abstract

The Computational Core is the structure through which the Program oversees and manages the software tasks associated with developing and evaluating the alternative breast exams. It serves the four Project components on an equal basis while also interfacing with the Clinical Core on critical data management and analysis issues pertaining to the clinical imaging sessions. The Core provides the expertise and effort necessary to (i) ensure that a state-of-the-art computing environment exists within the Program, (ii) maintain and evolve image reconstruction methods and manage, in collaboration with the Clinical Core, the data generated by the clinical activity associated with the Program, and (iii) develop the algorithmic advances common to model-based image reconstruction, image coregistration, image analysis and image display required to execute the technical and clinical strategic plans identified as the basis for Program renewal. Major accomplishments during the current funding period include (i) realization of software, modules and platforms for standardization of image reconstruction, visualization and analysis, (ii) implementation of parameter reduction/deployment techniques and parallelization to improve reconstruction efficiency, accuracy and throughput, (iii) development of parametric imaging concepts from multi-frequency/multispectral datasets and (iv) incorporation of statistical models and methods as tools for image uncertainty determination and image quality assessment. Continuation of the Computational Core is underpinned by a set of specific aims to maintain, enhance and advance these software development activities with particular emphasis on executing the cross-modality image coregistration component of the Program's technical strategic plan including the realization of a universal platform for multi-modality image visualization, registration and analysis. The Core will play a lead role in realizing the software for image registration where it will serve to integrate contributions from two subcontractors with the Program - (1) Genex Technologies, Inc., which will supply custom-designed optical tracking instrumentation and surface rendering software for Projects II-IV and (2) Worcester Polytechnique Institute (WPI) which will provide an image visualization/analysis platform that includes image registration capabilities.