This article investigates late-onset cognitive impairment using neuroimaging and genetics biomarkers for subjects participating in the Alzheimer’s Disease Neuroimaging Initiative (ADNI). Eight hundred and eight ADNI subjects were identified and divided into three groups: those with Alzheimer’s Disease (AD), those with mild cognitive impairment (MCI), and asymptomatic normal control (NC) group. Two hundred of the subjects qualified for AD diagnosis at the baseline; three hundred and eighty-three had MCI; and 225 were included in the NC group. The structural magnetic resonance imaging (MRI) data were parcellated using BrainParser, and the 80 most important neuroimaging biomarkers were extracted using the Global Shape Analysis (GSA) Pipeline workflow. We obtained 80 SNPs using Plink analysis via the Pipeline environment. In the AD cohort, rs2137962 was significantly associated with changes in left and right hippocampi and bilaterally in parahippocampal gyri, and rs1498853, rs288503, and rs288496 were significantly associated with hippocampi bilaterally, the right parahippocampal gyrus, and left inferior temporal gyrus. In the MCI cohort, rs17028008 and rs17027976 were significantly associated with right caudate and right fusiform gyrus, and rs2075650 (TOMM40) was significantly associated with right caudate, rs1334496 and rs4829605 were significantly associated with right inferior temporal gyrus. In the NC cohort, Chromosome 15 [rs734854 (STOML1), rs11072463 (PML), rs4886844 (PML) and rs1052242 (PML)] was significantly associated with the both hippocampi and both insular cortex and rs4899412 (RGS6) was significantly associated with caudate related biomarkers. We observed significant correlations between the SNPs and the neuroimaging phenotypes in the 808 subjects in terms of neuroimaging genetics. These results illustrate some of the neuroimaging-genetics associations between the AD, MCI and NC cohorts.