Edge Computing versus Cloud Computing: Impact on Retinal Image Pre-processing

With the emergence of new intelligent applications, there has been a revolution in information management, mainly including processing routines, storage, and computing capacity. Cloud computing is one of the most well-known and popular paradigms. However, there are still some barriers to cloud adoption. Edge computing serves as a decentralized extension and provides solutions that facilitate data processing at the generation source. This paper discusses the development of a retinal image pre-processing application for efficient use in screening systems and the impact that pre-processing has on network interconnection. Results show that the parallel version has reduced application execution time by up to ≈ 73%, decreasing by ≈11.5× the bandwidth used, achieving throughput above 5 images/second with Edge pre-processing, 2.57× higher than Cloud.