The following is a summary of the “Predicting time-to-conversion for dementia of Alzheimer’s type using multi-modal deep survival analysis,” published in the January 2023 issue of Neurobiology of Aging by Mirabnahrazam, et al.

It is challenging to predict an individual’s progression trajectory from normal or mildly impaired cognition to Dementia of Alzheimer’s Type (DAT) because DAT is a complex disorder influenced by numerous factors. For preclinical subjects at varying stages of disease development, an accurate estimate of time-to-conversion to DAT may be obtained by carefully examining data from multiple modalities. 

Based on the baseline data of 401 subjects with 63 features from MRI, genetic, and CDC (Cognitive tests, Demographic, and CSF) data from the Alzheimer’s Disease Neuroimaging Initiative (ADNI) database, they were able to predict the time-to-conversion to DAT. The results of our research show that when it comes to predicting the DAT time-to-conversion for subjects with Mild Cognitive Impairment (MCI), CDC data performs better than genetic or MRI data. 

Genetic data, on the other hand, had the greatest predictive power for subjects with Normal Cognition (NC) at the time of the visit. The time-to-event prediction was also enhanced by using both MRI and genetic features rather than just one or the other. When all other features were considered, adding CDC improved performance to that of using CDC alone.

Source: sciencedirect.com/science/article/abs/pii/S0197458022002196