The aim of this study was to develop a prediction model that integrated various image features and neuropsychological scores to yield a single estimate reflecting the probability of dementia.
A total of 130 subjects belong to Normal control group, AD group, and MCI group, were recruited in this study. For these subjects, the multiple features obtained from different modalities, including structural MRI morphometry (volume / shape), rs-fMRI, and neuropsychological assessment measures (NPA) were used to explore an optimal set of predictors of conversion from MCI to AD. Unlike previous studies using logistic regression analysis, a new method based on learning vector quantization (LVQ) and probabilistic neural network (PNN) is proposed to establish a prediction model.
We test the baseline, 1-year follow-up, and 2-year follow-up scans of 17 AD subjects (M/ F=5/12), 22 normal controls (NC; 13/9), 16 subjects that remain stable MCI (MCI-s; 11/5), and 4 subjects convert to AD within a given timeframe (MCI-c; 2/2). This study found that the proposed quantitative indicator provides well-behaving AD state estimates, corresponding well with the actual diagnosis.
According to the results, all of the test data have the trend that decreased over time. It has the potential to establish an effective decision support and data visualization framework for improving AD diagnostics, allowing clinicians to rapidly analyze large quantities of diverse patient data and as a screening measure and evaluated tool in therapeutic trials.