Cognitive impairment is a major concern in the aging population. The advent of a number of disease-modifying treatments for dementia of the Alzheimer’s type that need to be used at an early stage has emphasized the importance of making a correct diagnosis. The overlap among clinical phenotypes associated with cognitive impairment has led to the use of biomarkers to improve disease classification. Biomarkers can come from neuroimaging, cerebrospinal fluid, blood, and neuropsychological testing. Initially, the main biomarkers were amyloid and tau proteins in the CSF and on PET studies. Recently, the use of CSF and PET has been replaced with plasma-based biomarkers as measured by ultra-low quantities of these proteins in the blood. These measurements can be done with commercially available instruments such as the Quanterix HD-X Simoa and the MesoScale Discovery platform. Simoa can be used to measure neurofilament light (NfL) to detect axonal injury and glial fibrillary acidic protein (GFAP) for astrocyte activation as a measure of inflammation. By bypassing an invasive lumbar puncture or an expensive PET study, these blood measurements open the possibility of large-scale screening for early dementia populations that would otherwise be inaccessible. Adding neuroimaging of the white matter with diffusion tensor imaging indicates vascular damage. Combining blood-based biomarkers with imaging using a clustering machine learning approach expands the classification process to include vascular cognitive impairment and multi-etiology dementia. The use of biomarkers for early detection will improve the selection of patients for disease-modifying therapies and make them accessible to underrepresented populations.