Data-driven differential diagnosis of dementia using multiclass Disease State Index

Frontiers in Aging Neuroscience

A. Tolonen, H. Rhodius-Meester, M. Bruun, J. Koikkalainen, F. Barkhof, A. Lemstra, T. Koene, P. Scheltens, C. Teunissen, T. Tong, R. Guerrero, A. Schuh, C. Ledig, M. Baroni, D. Rueckert, H. Soininen, A. Remes, G. Waldemar, S. Hasselbalch, P. Mecocci, W. van der Flier, J. Lötjönen

The PredictND tool was developed as a clinical decision support system (CDSS) that combines information from multiple diagnostic tests (e.g., neuropsychological tests, MRI, cerebrospinal samples) for the differential diagnosis of different types of dementia. The multiclass Disease State Index classifier was used to differentiate between controls and four dementia types: Alzheimer’s disease, vascular dementia, frontotemporal lobar degeneration, and dementia with Lewy bodies. The high level of accuracy of the classifier in differentiation in this study indicates that the PredictND CDSS could be helpful in clinical practice.