Accelerating biomedical image processing using massively parallel processors

DESCRIPTION provided by applicant During the last decade the quantity of bioimaging data has grown tremendously Current estimates indicate that the average hospital in the USA houses some TB of data of which approximately is composed of unstructured image data from CT MRI and X ray machines This huge quantity of data is expected to grow at a rate of annually meaning hospitals could generate a total of one exabyte of new biomedical imaging data this year The last decade has also seen the development of several new computing platforms In particular multi core and massively parallel processors are ubiquitous Of these new platforms the sheer computational power in modern Graphical Processing Units GPUs have created a computing era where it is feasible for a developer to purchase a personal supercomputer with teraflops of processing ability for less than $ One of the most popular components of modern biomedical imaging software the Insight ToolKit ITK could benefit greatly from GPU computing There have been two attempts to implement ITKandapos s functionality on the GPU and although there were impressive results accelerations between x both projects were ultimately abandoned As it stands our GPU accelerated ArrayFire library already contains about of ITKandapos s core functionality more than any competing software Within the context of this proposal we seek to expand ArrayFireandapos s support of ITKandapos s functionality and create tools that will help developers use ArrayFire to leverage the massively parallel computing capabilities of GPUs from their ITK applications PUBLIC HEALTH RELEVANCE We seek to accelerate image processing algorithms in one of bioimagingandapos s most popular libraries the Insight ToolKit ITK by porting a subset of this libraryto Graphical Processing Units GPUs Previous work has shown that ITKandapos s functionality can be accelerated by times Such acceleration will dramatically reduce the time spent processing images and enable new approaches to medical image analysis