Premium
Machine learning approaches to predicting amyloid status using data from an online research and recruitment registry: The Brain Health Registry
Author(s) -
Albright Jack,
Ashford Miriam T.,
Jin Chengshi,
Neuhaus John,
Camacho Monica R.,
Rabinovici Gil D.,
TruranSacrey Diana,
Maruff Paul T,
Mackin R. Scott,
Nosheny Rachel L.,
Weiner Mike W.
Publication year - 2021
Publication title -
alzheimer's and dementia
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 6.713
H-Index - 118
eISSN - 1552-5279
pISSN - 1552-5260
DOI - 10.1002/alz.053330
Subject(s) - machine learning , artificial intelligence , test (biology) , psychology , medicine , computer science , biology , paleontology
Background This study investigated the extent to which subjective and objective data from an online registry can be analyzed using machine learning methodologies to provide an accurate prediction of the brain amyloid beta (Aß) status of registry participants. Method We developed and optimized machine learning models using data from up to 664 participants in the Brain Health Registry. Models were assessed on their ability to predict Aß positivity using the results of positron emission tomography as ground truth. Result The optimal feature set included gender, study partner‐assessed Everyday Cognition score, and the Cogstate Brief Battery’s One Card Learning test and resulted in a cross‐validated AUC score of 0.75. Conclusion Inclusion of study partner assessments and certain Cogstate Brief Battery scores increases the ability of machine learning models to predict Aß positivity.