Comparison of Machine Learning-based Approaches to Predict the Conversion to Alzheimer's Disease from Mild Cognitive Impairment. (15th March 2023)

• Record Type: Journal Article
• Authors: Franciotti, Raffaella; Nardini, Davide; Russo, Mirella; Onofrj, Marco; Sensi, Stefano L.
• Is Part Of: Neuroscience. Volume 514(2023)
• Page Start: 143
• Page End: 152
• Publication Date: 2023-03-15
• Subjects: Alzheimer's disease dementia (AD) prediction -- artificial intelligence -- gradient boosting -- machine learning (ML) algorithms -- mild cognitive impairment (MCI) -- random forest (RF); AdaBoost Adaptive Boosting -- ABETA Ab42 peptide -- AD Alzheimer's Disease Dementia -- ADAS-Cog-11 and 13 Alzheimer's Disease Assessment Scale-Cognitive Subscales-11 items and 13 items -- ADNI Alzheimer's Disease Neuroimaging Initiative -- c-MCI converting MCI -- FAQ Functional Activities Questionnaire -- GB Gradient Boosting -- LM-DEL Logical Memory-Delayed recall -- MCI Mild Cognitive Impairment -- ML Machine Learning -- NPV negative predictive value -- PPV positive predictive value -- PTAU phosphorylated tau protein -- RAVLT-DEL and -IMM Rey Auditory Verbal Learning Test-Delayed and Immediate recall -- p-Tau phosphorylated-Tau -- RF Random Forest -- SHAP SHapley Additive exPlanations -- s-MCI stable MCI -- sMRI structural MRI -- t-Tau total Tau -- XGB eXtreme Gradient Boosting; Neurochemistry -- Periodicals; Neurophysiology -- Periodicals; Neurology -- Periodicals; Neurochimie -- Périodiques; Neurophysiologie -- Périodiques; Neurochemistry; Neurophysiology; Electronic journals; Periodicals; Electronic journals; 612.8
• DOI: 10.1016/j.neuroscience.2023.01.029
• Languages: English
• ISSNs: 0306-4522

Abstract

Highlights: Psychological and AD-related biomarkers reach the highest accuracy in AD prediction. Clinical and biological data are helpful in machine learning algorithms' performances. Multiple machine learning algorithms make more accurate and solid AD predictions. Abstract: In Mild Cognitive Impairment (MCI), identifying a high risk of conversion to Alzheimer's Disease Dementia (AD) is a primary goal for patient management. Machine Learning (ML) algorithms are widely employed to pursue data-driven diagnostic and prognostic goals. An agreement on the stability of these algorithms –when applied to different biomarkers and other conditions– is far from being reached. In this study, we compared the different prognostic performances of three supervised ML algorithms fed with multimodal biomarkers of MCI subjects obtained from the Alzheimer's Disease Neuroimaging Initiative (ADNI) database. Random Forest, Gradient Boosting, and eXtreme Gradient Boosting algorithms predict MCI conversion to AD. They can also be simultaneously employed –with the voting procedure– to improve predictivity. AD prediction accuracy is influenced by the nature of the data (i.e., neuropsychological test scores, cerebrospinal fluid AD-related proteins and APOE ε4, cerebral structural MRI (sMRI) data). In our study, independent of the applied ML algorithms, sMRI data showed the lowest accuracy (0.79) compared to other classes. Multimodal data were helpful in the algorithms' performances by combining clinical and biological measures. Accordingly, using the three ML algorithms, the highest accuracy (0.90) was reached by employing neuropsychological and AD-related biomarkers. Finally, the feature selection procedure indicated that the most critical variables in the respective classes were the ADAS-Cog-13 scale, the medial temporal lobe and hippocampus atrophy, and the ratio between phosphorylated Tau and Aβ42 proteins. In conclusion, our data support the notion that using multiple ML algorithms and multimodal biomarkers helps make more accurate and solid predictions.