AUTHOR=Zhu Xia Wei , Liu Si Bo , Ji Chen Hua , Liu Jin Jie , Huang Chao 

TITLE=Machine learning-based prediction of mild cognitive impairment among individuals with normal cognitive function

JOURNAL=Frontiers in Neurology

VOLUME=Volume 15 - 2024

YEAR=2024

URL=https://www.frontiersin.org/journals/neurology/articles/10.3389/fneur.2024.1352423

DOI=10.3389/fneur.2024.1352423

ISSN=1664-2295

ABSTRACT=Prediction in individuals with normal cognitive function for a future risk of mild cognitive impairment (MCI) has not been well explored, as previous studies mainly focused on patients with MCI or dementia. The aim of the study was to provide basis for preventing MCI in normal populations. The data came from a longitudinal retrospective study involving individuals with brain magnetic resonance imaging scans and clinical visits with interval of more than 3 years. Multiple machine-learning technologies, including random forest, support vector machine, logistic regression, eXtreme Gradient Boosting, and naïve Bayes, were used to establish a prediction model of a future risk of MCI through a combination of clinical and image variables. Recall and precision were used as evaluation indicators. Among these machine learning models; eXtreme Gradient Boosting was the best classification model. The classification accuracy of clinical variables was 65.90%, of image variables was 79.54%, of a combination of clinical and image variables was 94.32%. The best result of the combination was an accuracy of 94.32%, a precision of 96.21%, and a recall of 93.08%. Extreme Gradient Boosting model with a combination of clinical and image variables had a potential prospect for the risk prediction of MCI among individuals with normal cognitive function. From clinical perspective, the degree of white matter hyperintensity, especially in the frontal lobe, and the control of systolic blood pressure were the most important risk factor for the development of MCI in normal populations.