Machine learning framework for early MRI-based Alzheimer's conversion prediction in MCI subjects

Area of Science:

• Neuroimaging
• Machine Learning
• Neurology

Background:

• Mild Cognitive Impairment (MCI) represents a critical transitional phase between normal aging and Alzheimer's Disease (AD).
• Early identification of MCI patients at high risk of progressing to AD is crucial for timely intervention and effective treatment strategies.
• Current diagnostic methods may not reliably predict MCI-to-AD conversion in the early stages.

Purpose of the Study:

• To develop and validate a novel magnetic resonance imaging (MRI)-based method for predicting the conversion of MCI to AD.
• To integrate MRI biomarkers with clinical and demographic data to enhance prediction accuracy.
• To enable earlier diagnosis of Alzheimer's Disease by identifying at-risk individuals in the MCI stage.

Main Methods:

• Developed a novel MRI biomarker for MCI-to-AD conversion using semi-supervised learning (low density separation).
• Employed regularized logistic regression for feature selection on MRI data from AD and control subjects, excluding MCI data.
• Removed age-related effects from MRI data prior to classifier training to mitigate confounding factors.
• Constructed an aggregate biomarker by combining the MRI biomarker with age and cognitive measures using a random forest classifier.

Main Results:

• The standalone MRI biomarker achieved an Area Under the Curve (AUC) of 0.7661 in discriminating progressive MCI (pMCI) from stable MCI (sMCI) using 10-fold cross-validation.
• The aggregate biomarker, incorporating MRI data, cognitive measures, and age, achieved a significantly higher AUC of 0.9020 in predicting pMCI from sMCI.
• Demonstrated the added value of the novel characteristics and the superiority of combining MRI with cognitive data for prediction.

Conclusions:

• The proposed MRI-based approach shows significant potential for the early diagnosis of Alzheimer's Disease.
• Magnetic Resonance Imaging plays a vital role in predicting MCI-to-AD conversion.
• Combining MRI data with cognitive test results substantially improves the accuracy of predicting MCI-to-AD conversion, highlighting its clinical utility.