Diagnosis of Alzheimer's disease and behavioural variant frontotemporal dementia with machine learning‐aided neuropsychological assessment using feature engineering and genetic algorithms

Abstract

Background Neuropsychological assessment is considered a valid tool in the diagnosis of neurodegenerative disorders. However, there is an important overlap in cognitive profiles between Alzheimer's disease (AD) and behavioural variant frontotemporal dementia (bvFTD), and the usefulness in diagnosis is uncertain. We aimed to develop machine learning‐based models for the diagnosis using cognitive tests. Methods Three hundred and twenty‐nine participants (170 AD, 72 bvFTD, 87 healthy control [HC]) were enrolled. Evolutionary algorithms, inspired by the process of natural selection, were applied for both mono‐objective and multi‐objective classification and feature selection. Classical algorithms (NativeBayes, Support Vector Machines, among others) were also used, and a meta‐model strategy. Results Accuracies for the diagnosis of AD, bvFTD and the differential diagnosis between them were higher than 84%. Algorithms were able to significantly reduce the number of tests and scores needed. Free and Cued Selective Reminding Test, verbal fluency and Addenbrooke's Cognitive Examination were amongst the most meaningful tests. Conclusions Our study found high levels of accuracy for diagnosis using exclusively neuropsychological tests, which supports the usefulness of cognitive assessment in diagnosis. Machine learning may have a role in improving the interpretation and test selection.

Neural basis of visuospatial tests in behavioral variant frontotemporal dementia

“…In addition, we found that visuospatial task performance worsens in both AD and bvFTD when the dementia stage progresses, according to previous studies (Ranasinghe et al, 2016). Although we did not find statistically significant differences between bvFTD and AD groups in the visuospatial tests used in our study, the analysis of visuospatial performance in the context of the other tests and clinical staging could be helpful for differential diagnosis and monitoring (Ranasinghe et al, 2016;García-Gutiérrez et al, 2021). Impairment in visuospatial tests was greater in AD than bvFTD in the mild dementia stage but not in mild cognitive impairment stage, which suggests a different progression rate.…”

GA-MADRID: design and validation of a machine learning tool for the diagnosis of Alzheimer’s disease and frontotemporal dementia using genetic algorithms

“…Although we agree that this is a very limiting threshold, the purpose of this work is to present a parameterizable computing framework, in which this threshold, as many other parameters, can be selected by the expert user in order to meet its clinical goals. The clinical value of the results obtained by the use of our proposed framework is out of the scope of this publication, but is has been already proven in [27].…”

“…The aim of this work is the development of the computational framework, which is widely customisable and scalable. In this publication, we do not target the accuracy of the clinical assessment provided by the tool and presented in publications like [27], but we discuss around a case of study to show the functionalities of the AI-based tool. This computational framework has been evaluated using a dataset, which includes cognitive and PET data from 329 patients (171 AD, 72 bvFTD and 87 Healthy controls.…”

An Intelligent Learning System for Unbiased Prediction of Dementia Based on Autoencoder and Adaboost Ensemble Learning

“…F.G. Gutierrez et al had designed an automated diagnostic system for the detection of AD and frontotemporal dementia (FTD) by using feature engineering and genetic algorithms. Their proposed system had obtained an accuracy of 84% [ 32 ]. G. Mirzaei & H. Adeli analyzed state-of-the-art ML techniques for the detection and classification of AD [ 33 ].…”