A DTI-Radiomics and Clinical Integration Model for Predicting MCI-to-AD Progression Using Corpus Callosum Features.

Area of Science:

• Neuroimaging
• Radiomics
• Alzheimer's Disease Research

Background:

• Alzheimer's disease (AD) poses a significant diagnostic challenge, particularly in early stages.
• Distinguishing between stable mild cognitive impairment (MCI) and progression to AD is crucial for timely intervention.

Purpose of the Study:

• To evaluate the efficacy of diffusion tensor imaging (DTI)-based radiomics for early AD diagnosis.
• To predict the conversion of MCI to AD using DTI radiomics signatures.

Main Methods:

• Radiomic features were extracted from DTI images of 186 MCI patients from the ADNI database.
• A Least Absolute Shrinkage and Selection Operator (LASSO) logistic regression model was used to build radiomic signatures.
• Model performance was assessed using receiver operating characteristic (ROC) and decision curve analysis (DCA).

Main Results:

• The combined radiomics and clinical model achieved an Area Under the Curve (AUC) of 0.936 in predicting MCI to AD conversion.
• Individual DTI radiomic signatures showed AUCs ranging from 0.824 to 0.862.
• The combined model demonstrated superior predictive performance compared to clinical or radiomics models alone.

Conclusions:

• DTI-based radiomics, integrated with clinical data, offers a robust method for forecasting MCI to AD progression.
• This approach holds promise for early identification and intervention in individuals at risk of developing Alzheimer's disease.