This is the authors' abstract. We don't add key points for this paper.
Magnetic resonance (MR) images are often influenced by undesirable noise that is combined with the relevant signal. Therefore, the resulting images may not provide the quality needed for their interpretation. This study proposes the periodogram as a nonparametric method for processing and reconstructing noise-affected MR images without prior knowledge of their spectral characteristics. Here, we simulate real-world image noise by artificially adding Gaussian noise. By transforming the corrupted MR image into the frequency domain, applying the Daniell periodogram, and performing an inverse transformation, the noise impact is reduced. The authors aim to evaluate the method’s potential applications and limitations for future use in MRI noise elimination, using the structural similarity index measure (SSIM), peak signal-to-noise ratio (PSNR), and fractal dimension (FD) metrics. The values of the said metrics of the denoised images largely correspond to the visual evaluation of the images.
Answers are grounded in this paper and cite their sources.
No takes yet. Share an insight, caveat, or question.
Mamotenko et al. (2025) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: