Prediction of 7‐year's conversion from subjective cognitive decline to mild cognitive impairment

• Published in: Human brain mapping Vol. 42; no. 1; pp. 192 - 203
• Main Authors: Yue, Ling, Hu, Dan, Zhang, Han, Wen, Junhao, Wu, Ye, Li, Wei, Sun, Lin, Li, Xia, Wang, Jinghua, Li, Guanjun, Wang, Tao, Shen, Dinggang, Xiao, Shifu
• Format: Journal Article
• Language: English
• Published: Hoboken, USA John Wiley & Sons, Inc 01.01.2021; Wiley
• Subjects: Aged; Alzheimer's disease; Amnesia - diagnosis; Amnesia - pathology; Amnesia - physiopathology; Amygdala; Amygdala - diagnostic imaging; Amygdala - pathology; Brain - diagnostic imaging; Brain - pathology; Cognitive ability; Cognitive Dysfunction - diagnosis; Cognitive Dysfunction - pathology; Cognitive Dysfunction - physiopathology; Conversion; Diagnostic Self Evaluation; Disease Progression; Educational Status; Female; Follow-Up Studies; Human health and pathology; Humans; Impairment; Learning algorithms; Life Sciences; Machine Learning; Magnetic Resonance Imaging; Male; Medical imaging; Middle Aged; MRI; Neurodegenerative diseases; Neuroimaging; Neuropsychological Tests; prediction; Prognosis; Psychiatrics and mental health; Retrospective Studies; Self Report; subjective cognitive decline; Substantia alba; Superior temporal sulcus; White Matter - diagnostic imaging; White Matter - pathology; prediction; subjective cognitive decline; machine learning; MRI; Subjective cognitive decline; Machine learning; Prediction
• ISSN: 1065-9471; 1097-0193; 1097-0193
• DOI: 10.1002/hbm.25216

ABSTRACT Subjective cognitive decline (SCD) is a high‐risk yet less understood status before developing Alzheimer's disease (AD). This work included 76 SCD individuals with two (baseline and 7 years later) neuropsychological evaluations and a baseline T1‐weighted structural MRI. A machine learning‐based model was trained based on 198 baseline neuroimaging (morphometric) features and a battery of 25 clinical measurements to discriminate 24 progressive SCDs who converted to mild cognitive impairment (MCI) at follow‐up from 52 stable SCDs. The SCD progression was satisfactorily predicted with the combined features. A history of stroke, a low education level, a low baseline MoCA score, a shrunk left amygdala, and enlarged white matter at the banks of the right superior temporal sulcus were found to favor the progression. This is to date the largest retrospective study of SCD‐to‐MCI conversion with the longest follow‐up, suggesting predictable far‐future cognitive decline for the risky populations with baseline measures only. These findings provide valuable knowledge to the future neuropathological studies of AD in its prodromal phase. In this article, we used a community‐based cohort of subjective cognitive decline (SCD) elderly subjects with 7‐year follow‐up and retrospectively identified individual SCD who had converted to mild cognitive impairment (MCI) during the follow‐up. After identifying progressive SCDs from stable SCDs, we constructed a machine learning‐based prediction model that successfully identified most progressive SCD individuals based on 198 morphometric features derived from structural MRI and 25 comprehensive clinical features. The striking finding is that as few as five neuroimaging and clinical features derived from baseline were able to predict cognitive impairment occurred 7 years later.