Image Processing and Optimization Based on Nvidia RTX Family Graphics Processors

Auccahuasi W.; Romani-Allende F.; Olaya-Cotera S.; Cosme M.; Castro-Mejia P.; Meza S. · 2025 6th International Conference on Data Intelligence and Cognitive Informatics, ICDICI 2025 (2025) · DOI: 10.1109/icdici66477.2025.11135311

Resumen

Graphics processors allow greater performance and are required, in processes where computational capacity is necessary, Nvidia's new RTX architecture provides certain innovations in its architecture, which can be differentiated from its previous version, this difference can be observed in video games, where it presents greater sharpness of the image, greater realism of the scenes, which provide a better gaming experience. We bring these benefits to the visualization of medical images, through the use of RTX family GPUs, allowing a better experience in the analysis of medical images, providing shorter processing time with better visualization, the sharpness of the images is improved. They perceive the best detail of the images. In the results we present the use and demonstration through the use of fundus images, which has a series of difficulties in its analysis, with the use of the RTX GPU, certain artifacts can be observed, which are soft exudates, which They are like a kind of yellowish cloud spots, which is very difficult to observe with the use of conventional computers.

Instituciones: Universidad ESAN; Universidad Científica del Sur; Universidad San Ignacio de Loyola; Universidad Tecnológica del Perú; Universidad César Vallejo; Universidad Autónoma de Ica