AI based Early Prediction of Alzheimer’s Disease using Neuropsychological Tests

Abstract

Within the field of biomedical engineering, a novel imputation model employing machine learning techniques is proposed as a unique approach for the early diagnosis of Alzheimer's disease (AD). Early identification is essential to postpone the progression of Alzheimer's disease (AD) and lessen its burden on patients. AD is characterized by progressive cognitive loss. The planned study is carried out in two stages. Phase one addresses a major obstacle in early AD identification by introducing a state-of-the-art technique for imputing missing values in clinical datasets. Data integrity is maintained using the imputation process, which guarantees the preservation of statistical properties within each feature. Phase two involves restructuring the clinical data and applying the proposed imputation model to impute the missing values. The training of a classifier model that is intended to function with unique labels based on patient prognosis comes next. With an accuracy rate of 92%, the imputation model and classifier integration show a notable improvement in early AD identification. The results highlight the usefulness of neuropsychological evaluations as reliable markers for the early detection of Alzheimer's disease (AD), made possible by cutting-edge machine learning techniques. This research contributes to the field of artificial intelligence by presenting a robust imputation framework and its practical application in enhancing early diagnostic capabilities in biomedical engineering.