Objective MRI brain segmentation artificial intelligence imaging tool "U-AI" was used to quantitatively analyze the brain volume in order to explore the difference of atrophy in different brain regions in patients with moderate and severe carotid artery stenosis (CAS) and cerebral microvascular disease (CSVD), and reveal the effect of different vascular diameters on brain volume. Methods The data of patients hospitalized in the Affiliated Hospital of Yan'an University from November 2021 to December 2023 were collected retrospectively. 92 patients with carotid artery stenosis ≥50% and 88 patients with CSVD total load score ≥3 were selected. All patients completed carotid ultrasound and cranial MRI examination, and then were divided into two groups according to the degree of carotid stenosis and CSVD total load score. Patients with carotid stenosis ≥50% were classified as moderate to severe stenosis group, and patients with CSVD total load score ≥3 as moderate and severe load group. At the same time, the general clinical data and imaging data of all patients were collected. The brain volumes of all patients were measured automatically by AI software, and the quantitative data of brain volumes in different brain regions were obtained. χ2 test, t test or Mann-Whitney U test were used to compare the differences of various indexes between the two groups, and stepwise regression analysis was used to explore the influencing factors of gray matter and white matter of the whole brain. 〖WTHZ〗Results The proportion of coronary heart disease and hyperlipidemia in the moderate and severe stenosis group was significantly higher than that in the moderate and severe burden group (P<0.05), while the proportion of hypertension in the moderate to severe burden group was significantly higher than that in the moderate and severe stenosis group (P<0.05). The brain volume atrophy of whole brain gray matter, temporal lobe cortex (hippocampus, fusiform gyrus, middle temporal gyrus, temporal pole), frontal lobe cortex (superior frontal gyrus), parietal cortex (postcentral gyrus, inferior parietal lobule, supramarginal gyrus), cingulate gyrus, insular lobe and subcortical gray nucleus (globus pallidus, thalamus) in moderate and severe burden group were significantly higher than those in moderate and severe stenosis group (all P<0.05). Further analysis showed that age and CSVD total load score were the influencing factors of the relative volume of gray matter in the whole brain. Age was the influencing factor of the relative volume of white matter of the whole brain (P<0.05). Conclusion The degree of brain volume atrophy in different brain regions of patients with moderate and severe CSVD load was significantly higher than that of patients with moderate and severe CAS. This result clearly shows that the degree of brain atrophy in patients with cerebral small vascular disease is more prominent than that of large cerebral artery stenosis, which further indicates that cerebral small vascular disease has more serious clinical symptoms and harm