<p>Amnestic mild cognitive impairment (aMCI) and Alzheimer’s disease (AD) are closely related. While some aMCI patients convert to AD (conversion), some revert to age-appropriate cognitive functioning (reversion). Early identification of these bidirectional trajectories informs aMCI-AD pathology and aids patient management. Data from 129 aMCI participants in the Alzheimer’s Disease Neuroimaging Initiative, who either converted or reverted, were analyzed. Baseline and longitudinal data relative to exact conversion and reversion times were incorporated to evaluate predictive power using support vector machine analysis. The accuracy of three models, each utilizing different data modalities, was compared. Neurobiological markers associated with these clinical trajectories were examined. The model incorporating brain volumetric changes achieved 92.31% accuracy in classifying conversion vs. reversion. Key neural markers included left inferior lateral ventricle volume for conversion times, left inferior temporal gyrus volume for cognitive status at conversion, and hippocampal and amygdala volumes for memory performance at conversion. The models offer bidirectional, prognostic predictions in aMCI patients. Identified markers provide valuable insights for early intervention. This study informs strategies to reduce AD incidence and optimize resource allocation, contributing to a more comprehensive understanding of disease trajectories and effective management.</p>

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

Bidirectional prognostic predictions of conversion and reversion in amnestic mild cognitive impairment

  • Esther Zhiwei Zheng,
  • Yue Gu,
  • Rachel R. Jin,
  • Tatia M. C. Lee

摘要

Amnestic mild cognitive impairment (aMCI) and Alzheimer’s disease (AD) are closely related. While some aMCI patients convert to AD (conversion), some revert to age-appropriate cognitive functioning (reversion). Early identification of these bidirectional trajectories informs aMCI-AD pathology and aids patient management. Data from 129 aMCI participants in the Alzheimer’s Disease Neuroimaging Initiative, who either converted or reverted, were analyzed. Baseline and longitudinal data relative to exact conversion and reversion times were incorporated to evaluate predictive power using support vector machine analysis. The accuracy of three models, each utilizing different data modalities, was compared. Neurobiological markers associated with these clinical trajectories were examined. The model incorporating brain volumetric changes achieved 92.31% accuracy in classifying conversion vs. reversion. Key neural markers included left inferior lateral ventricle volume for conversion times, left inferior temporal gyrus volume for cognitive status at conversion, and hippocampal and amygdala volumes for memory performance at conversion. The models offer bidirectional, prognostic predictions in aMCI patients. Identified markers provide valuable insights for early intervention. This study informs strategies to reduce AD incidence and optimize resource allocation, contributing to a more comprehensive understanding of disease trajectories and effective management.