Abstract
Aims
Predicting the diagnosis of an older adult solely based on their financial capacity performance or other neuropsychological test performance is still an open question. The aim of this study is to highlight which tests are of importance in diagnostic protocols by using recent advancements in machine learning.
Methods
For this reason, a neuropsychological battery was administered in 543 older Greek patients already diagnosed with different types of neurocognitive disorders along with a test specifically measuring financial capacity, that is, Legal Capacity for Property Law Transactions Assessment Scale (LCPLTAS). The battery was analysed using a random forest algorithm. The objective was to predict whether an older person suffers from dementia. The algorithm’s performance was tested through cross-validation.
Results
Machine learning was applied for the first time in data analysis regarding financial capacity and three factors-tests were revealed as the best predictors: two subscales from the LCPLTAS measuring ‘financial decision making’ and ‘cash transactions’, and the widely used MMSE which measures general cognition. The algorithm demonstrated good performance as measured by the F1-score, which is a measure of the harmonic mean of precision and recall. This evaluation metric in binary and multi-class classification integrates precision and recall into a single metric to gain a better understanding of model performance.
Conclusions
These findings reveal the importance of focusing on these scales and tests in neuropsychological assessment protocols. Future research may clarify in other cultural settings if the same variables are of importance.
If you do not see content above, kindly GO TO SOURCE.
Not all publishers encode content in a way that enables republishing at Neuro.vip.
This post is Copyright: Vaitsa Giannouli,
Stylianos Kampakis | December 19, 2024
Wiley: Journal of Neuropsychology: Table of Contents