MATHEMATICAL ALGORITHM FOR MAGNETIC RESONANCE IMAGING

Abstract

In this study, we investigated the noise sources used in random number generators. For the random number generators, conventional methods use noise sources that depend on the hardware and noise sources provided by a specific operating system (OS) as seeds. However, our proposed algorithm presents a new method for generating noise sources independent of the hardware and OS. The noise source presented in this study was used to verify the random number distribution using time series and frequency distribution data and to evaluate the random number using the National Institute of Standards and Technology (NIST) SP800-90B random number tool. The data was randomized using the excessive noise generated. By testing the performance of the conventional and proposed mathematical algorithms, we observed a similar degree of randomness with magnetic resonance imaging data. Therefore, we can confirm that we can use the proposed mathematical algorithm as a noise source in low-end systems, such as microprocessor-based microcomputers, where it is difficult to collect noise sources based on the abundant resources provided by hardware and OSs.