Title: Magnetic resonance brain volume property-based accelerate medical image algorithms using graphics processing unit

Authors: P. Sriramakrishnan; T. Kalaiselvi

Addresses: Department of Computer Applications, School of Computing, Kalasalingam Academy of Research and Education (Deemed to be University), Krishnankoil, Tamil Nadu 626126, India ' Department of Computer Science and Applications, The Gandhigram Rural Institute (Deemed to be University), Gandhigram, Tamil Nadu 624302, India

Abstract: This paper elaborates the design and implementation details of parallel image processing techniques that are used to accelerate the medical image algorithms with compute unified device architecture (CUDA) supported graphics processing unit (GPU). The algorithms are chosen from denoising, morphology, clustering, and segmentation. Three parallel computing models are developed based on the properties of algorithms and magnetic resonance imaging (MRI) principles. The acceleration of parallel algorithms is compared with that of sequential central processing unit (CPU) implementation measured in terms of speedup folds (×).

Keywords: graphics processing unit; GPU; compute unified device architecture; CUDA; parallel processing; medical imaging; brain volume; bilateral filter; non-local means; NLM; K-means clustering; FCM.

DOI: 10.1504/IJBET.2022.120866

International Journal of Biomedical Engineering and Technology, 2022 Vol.38 No.2, pp.128 - 150

Received: 16 Oct 2018
Accepted: 18 Dec 2018
Published online: 15 Feb 2022 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article