Construction of Diagnosis System for Early AD Based on Multi-Modality MRI Technology

Mild Cognitive Impairment Clinical Trial

Clinical Trial Details — Status: Completed

Administrative data

• NCT number: NCT02353845
• Other study ID #: hanying2
• Status: Completed
• Phase: N/A
• First received: October 2, 2014
• Last updated: August 21, 2016
• Start date: November 2013
• Est. completion date: August 2016

Study information

• Verified date: August 2016
• Source: Xuanwu Hospital, Beijing
• Contact: n/a
• Is FDA regulated: No
• Health authority: China: Ethics CommitteeChina: National Natural Science Foundation
• Study type: Observational

Clinical Trial Summary

One purpose of this study is to construct the diagnosis system for early Alzheimer's disease(AD), which is also called amnestic mild cognitive impairment (aMCI), and then further construct the predictable classifier from aMCI to AD based on Multi-Modality MRI characteristics of aMCI patients.

Description:

The cognition of aMCI is between normal aging and dementia, which is thought the transitional stage of dementia. Patients with aMCI have heavy risk to convert to AD, so in this study, the investigators focus on the construction of the diagnosis system for early AD based on multi-modality MRI characteristics of aMCI patients. Every patient underwent β-Amyloid PET, fluorodeoxyglucose-PET(FDG-PET), structural MRI, diffusion tensor imaging and functional MRI. Then investigators further study the patients who convert to AD and explore their MRI and metabolism characteristics on baseline, in order to construct the predictable classifier from aMCI to AD. The investigators want to achieve the early diagnosis of AD and help clinicians interfere with the progress of this disease.

Recruitment information / eligibility

• Status: Completed
• Enrollment: 297
• Est. completion date: August 2016
• Est. primary completion date: October 2015
• Accepts healthy volunteers: Accepts Healthy Volunteers
• Gender: Both
• Age group: N/A and older

Eligibility

Inclusion Criteria:

• Memory loss complaint and confirmed by an informant

• Cognitive impairment in single or multiple domains, adjusted for age and education

• Normal or near-normal performance on general cognitive function and no or minimum impairment of daily life activities

• A Clinical Dementia Rating (CDR) score is 0.5 and consistent with the boundary of neuropsychological scale

• Failure to meet the criteria for dementia

• Must be able to accept examination of MRI, sight and hearing allow to complete test

• Right handedness

Exclusion Criteria:

• Other diseases that cause cognitive impairment, such as thyroid disease, stroke and so on

• People who have severe visual and hearing impairment

Study Design

Observational Model: Case Control, Time Perspective: Prospective

Related Conditions & MeSH terms

• Alzheimer Disease
• Alzheimer's Disease
• Cognition Disorders
• Mild Cognitive Impairment

Locations

Country	Name	City	State
China	Department of Neurolgy,Xuanwu Hospital of Capital Medical University	Beijing	Beijing

Sponsors (1)

Lead Sponsor	Collaborator
XuanwuH 2

Country where clinical trial is conducted

China, 

Outcome

Type	Measure	Description	Time frame	Safety issue
Primary	number of participants correctly classified by the support vector machine (SVM) classifier for the aMCI diagnosis	two-hundred aMCI subjects and 100 normal controls recruited will undergo structure,resting-state functional magnetic resonance imaging and diffusion tensor imaging. An SVM classifier for diagnosis will be trained based on these neuroimaging data.Then leave-one-out cross validation will be used to estimate the performance of the classifier including accuracy,sensitivity,specificity.The classification accuracy will be measured by the proportion of observations that are correctly classified into the aMCI or control groups.The sensitivity is defined as TP/(TP+FN), and specificity is defined as TN/(TN+FP). The TP (true positive) is the number of aMCI images correctly classified,whereas the TN (true negative) is the number of control images correctly classified. The FP (false positive) is the number of control images classified as the aMCI, whereas the FN (false negative) is the number of aMCI images classified as controls.	3 years	Yes
Primary	number of participants correctly predicted by the SVM classifier for predicting conversion from aMCI to AD	During 2-year follow-up, the group of aMCI will be divided into progressive aMCI (aMCIp) and stable aMCI(aMCIs).According to the baseline neuroimaging data, an SVM classifier for predicting conversion from aMCI to AD will be trained. Then leave-one-out cross validation will be used to validate the performance of the classifier including accuracy,sensitivity,specificity.The classification accuracy will be measured by the proportion of aMCI that are correctly classified into the aMCIp or aMCIs groups.The sensitivity is defined as TP/(TP+FN), and specificity is defined as TN/(TN+FP). The TP (true positive) is the number of aMCIp images correctly classified,whereas the TN (true negative) is the number of aMCIs correctly classified. The FP (false positive) is the number of aMCIs classified as aMCIp, whereas the FN (false negative) is the number of aMCIp classified as aMCIs.	3 years	Yes
Secondary	regional cerebral metabolism (CMgl) measured by FDG-PET	different glucose consumption rate in some regions between aMCI and normal controls, and also between aMCIp and aMCIs	3 years	Yes
Secondary	changed regional cerebral blood flow measured by FDG-PET		3 years	Yes