Högskolan i Skövde

Dementia is a neurological condition that primarily affects older adults and there is stillno cure or therapy available to cure it. The symptoms of dementia can appear as early as 10 yearsbefore the beginning of actual diagnosed dementia. Hence, machine learning (ML) researchershave presented several methods for early detection of dementia based on symptoms. However,these techniques suffer from two major flaws. The first issue is the bias of ML models caused byimbalanced classes in the dataset. Past research did not address this issue well and did not takepreventative precautions. Different ML models were developed to illustrate this bias. To alleviate theproblem of bias, we deployed a synthetic minority oversampling technique (SMOTE) to balance thetraining process of the proposed ML model. The second issue is the poor classification accuracy ofML models, which leads to a limited clinical significance. To improve dementia prediction accuracy,we proposed an intelligent learning system that is a hybrid of an autoencoder and adaptive boostmodel. The autoencoder is used to extract relevant features from the feature space and the Adaboostmodel is deployed for the classification of dementia by using an extracted subset of features. Thehyperparameters of the Adaboost model are fine-tuned using a grid search algorithm. Experimentalfindings reveal that the suggested learning system outperforms eleven similar systems which wereproposed in the literature. Furthermore, it was also observed that the proposed learning systemimproves the strength of the conventional Adaboost model by 9.8% and reduces its time complexity.Lastly, the proposed learning system achieved classification accuracy of 90.23%, sensitivity of 98.00%and specificity of 96.65%.