0
30. 6. 2018.
Evaluation of medical image algorithms on multicore processors
Introduction : In recent time medical image processing and analysis become an essential component in clinical practice. Medical images have a huge data to process due to increased image resolution. These tasks are inherently parallel in nature, so will fit to naturally to parallel processors like Central Processing Unit (CPU) and Graphics Processing Unit (GPU). Methods : In this work several common used image processing algorithms for 2D and 3D were evaluated regarding the computation performance increase using the GPUs and CPUs on a Personal Computer (PC). Results : For tested algorithms GPU leads to decreasing running times from 1.1 to 422 times.