The participant in this study includes Alzheimer's disease (AD including familial AD and sporadic AD) patients, amnestic mild cognitive impairment (aMCI) patients, non-AD dementia patients and cognitively normal control.

The purpose of this study is to establish the best cut-off value of cerebrospinal fluid (CSF) and blood β-amyloid (Aβ) 42/40, total tau (t-tau) , phosphorylated tau ,inflammatory factors, etc. in diagnosis of Alzheimer's disease (AD).

1. Incorporating patients who meet the criteria from Capital Medical University Xuanwu Hospital including AD, aMCI, non-AD dementia and cognitively normal control.
2. Collect the information of clinical and neuropsychological assessment (mini-mental state examination, MMSE, Montreal cognitive assessment, MoCA, and clinical dementia rating scale, CDR), neuroimaging data including medial temporal atrophy (MTA) score, Fazekas score, MRI, and PET(optional), as well as the biological sample, such as CSF and peripheral blood.
3. To quantify levels of Aβ42, Aβ40, t-tau, p-tau, inflammatory factors, etc. in CSF and blood.
4. Detect all participants ApoE genotype.
5. Independently establish the best cut-off value of Chinese people in our laboratory.
6. Analysis the relationships between core biomarkers in CSF/blood and ApoE genotype and neuroimaging data.
7. Establish combined diagnostic model.

Inclusion Criteria:

• Aged 55-75. Written informed consent obtained from participant or legal guardian prior to any study-related procedures. The diagnosis of AD is made using the National Institute on Aging and the Alzheimer's Association (NIA-AA) criteria. The diagnosis of aMCI is assigned according to 2004 Petersen criteria. As for non-AD dementia, the McKeith criteria are used for DLB, the revised diagnostic criteria proposed by the International bvFTD Criteria Consortium for behavioral variant FTD, the Gorno-Tempini criteria for the semantic variant FTD or non-fluent aphasia, the Movement Disorder Society Task Force criteria for PDD, the vascular behavioral and cognitive disorders (Vas-Cog) criteria for VaD, the Armstrong's criteria for CBD, the CDC's diagnostic criteria for CJD, etc. In addition, normal cognition is supported by MMSE, CDR and other cognitive function scales.

Exclusion Criteria:

• Other medical or psychiatric illness. No one can serve as an informant. Refused to complete a cognitive test and provide biospecimen.

AD patients (n=800); aMCI patients (n=800);non-AD dementia patients (n=800). cognitively normal controls (n=800);