Evaluation of the Effects of Folic Acid Combined with Atorvastatin on the Poststroke Cognitive Impairment by Low-Rank Matrix Denoising Algorithm-Based MRI Imaging

Abstract

This research aimed to study the optimization effects of the low-rank matrix denoising (LRMD) algorithm based on the Gaussian mixture model (GMM) on MRI images of stroke patients, aiming to evaluate the effects of atorvastatin combined with folic acid on poststroke cognitive impairment (PSCI) in patients with ischemic stroke. First, the GMM-based low-rank matrix denoising (LRMD) algorithm was constructed and applied to process MRI images of 64 patients with ischemic stroke. Then, the MRI images before and after processing were compared for the denoising degree and quality. An image with 5% noise was not as clear as an MRI image with 1% noise, and the effects of atorvastatin combined with folic acid on PSCI in patients with ischemic stroke were discussed. It was found that the denoising degree of MRI images processed by the GMM-based LRMD algorithm was significantly improved, the image quality was significantly enhanced ( $P<0.05$ ), and the diagnosis accuracy and efficiency of stroke patients were heightened. Atorvastatin combined with folic acid reduce the homocysteine (HCY) and total cholesterol (TC) levels, as well as Montreal Cognitive Scale (MOCA) scores of PSCI patients ( $P<0.05$ ). In conclusion, the MRI images processed by the LRMD algorithm have good quality. Folic acid combined with atorvastatin can effectively reduce HCY and TC levels, thereby alleviating PSCI of stroke patients.