Implementation of a nonlinear gradient adaptive filter for processing of large-size medical sequences on general-purpose hardware

Abstract

To achieve significant noise reduction in medical images while at the same time preserving fine structures of diagnostic value, a non-linear filter called the multi-resolution gradient adaptive filter (MRGAF) was developed. Though the algorithm is well suited for its task of noise reduction in medical images, it is still limited to the application of offline processing in medical workstations due to its computational complexity. The aim of our study is to reach real-time processing of data from low-cost x-ray systems on a standard PC without additional hardware. One major drawback of the original MRGAF procedure is its irregular memory access behavior caused by the intermediate multi-resolution representation of the image (Laplacian pyramid). This is addressed by completely re-arranging the computation. The image is divided into super-lines carrying all relevant information of all pyramidal levels, which allow to apply the complete MRGAF procedure in a single pass. This way, the cache utilization is improved considerably, the total number of memory accesses is reduced, and the use of super-scalar processing capabilities of current processors is facilitated. The current implementation allows applying advanced multi-resolution non-linear noise reduction to images of 768 x 564 pixels at a rate of more than 30 frames per second on a workstation. This shows that high-quality real-time image enhancement is feasible from a technical as well as from an economical point of view.

Real-Time Image and Video Processing: From Research to Reality

“…This organization involved dividing an image into partitions called "super-lines," where each partition contained all the necessary information from each level of the multiresolution representation to allow the application of the algorithm in one pass. In [44], the results showed that after application of the super-line approach, only 0.2% of the memory accesses were to the external memory, which helped to achieve a processing rate of 43.6 fps for 512 × 512 images on a desktop GPP. Also, in [101], the results showed that the super-line approach was able to reduce the miss rate from 99% to 0.8%, achieving a processing rate of 44 fps for 555 × 382 images on a desktop GPP.…”

“…In the never ending quest for a perfect picture, research in developing fast, high-quality algorithms for processing pictures/videos captured by consumer digital cameras or cell-phone cameras [80] is expected to continue well into the future. Of course, the developments in industrial inspection [25,34,67,135,147] and medical imaging systems [18,23,24,44,136,143,145] will continue to progress. The use of color image data [8,85,107,109], or in some cases, multispectral image data [139] in real-time image/video processing systems is also becoming an important area of research.…”

“…In contrast to the traditional block-and strip-partitioning schemes often used in the reorganization of computations, a new data organization called "super-line" processing was developed in [44,58,101] specifically addressing the issue of cache-misses in algorithms that use multiresolution representations of image data. This organization involved dividing an image into partitions called "super-lines," where each partition contained all the necessary information from each level of the multiresolution representation to allow the application of the algorithm in one pass.…”