The cutting-edge technologies in eHealthcare include telemedicine and artificial intelligence. A digital health framework has been introduced by the Indian government to promote research in these new fields in order to solve the issue of access to competent medical care. This would allow for cutting-edge telemedicine installations. In these computer-assisted systems, MRI diagnostics is essential since it serves as the foundation for making diagnoses of disorders. The study presented here offers bilateral filter as the best suitable denoising method to produce high-quality MR images. However, the filter's computational complexity makes it difficult to analyse huge images quickly enough. The study presented here presents a fast bilateral filter using a GPU-based parallel method for MR image denoising, taking into account the real-time demands of remote healthcare. The GPU onchip shared memory and constant cache are explored while implementing the denoising algorithm utilising a novel memory optimization approach, which results in a shorter execution time. The difficulties that bilateral filters for MR image denoising present are thoroughly discussed in this study. In order to achieve higher speedup, it covers the new memory optimization strategy suggested for GPU-based implementation of the filter. It also addresses the suggested fix to improve the filter's denoising effectiveness for noisy MR pictures

Denoising, Filter, MR Image, Healthcare, Diagnostics and GPU.