| Summary: | Alzheimer’s disease is a type of dementia that is well known and responsible for affecting the lives of the elderly. It is defined by the gradual loss of structure and function of neurons in the brain leading to memory, thinking and other activities. This is the most crucial step since a patient’s quality of life and the disease’s progression may both be improved with an early diagnosis. Nevertheless, existing diagnostic tests rarely diagnose the disease in its preliminary stage, and this has a significant impact on the course of the illness. The more conventional assessment techniques that involve neuroimaging and cognitive ability standardized tests are usually unable to pick up early stage alterations. To address these limitations, we have developed a new Hybrid AI Model, which combines both the conventional machine learning techniques, namely SVM, Naive Bayes, Cat boost, and XGBoost and Stacked DL model. This combination uses the advantages of the proposed models to enhance the diagnostic sensitivity based on the early AD biomarkers. The MRI data was obtained from Kaggle and the proposed Stacked DL Model achieved an accuracy93%, an f1-score94, and a specificity99%. The Voting classifier (ML models) outperformed the other models with an accuracy94.22%, an f1-score94%, and a specificity99%. proving the proposed model superior to the prior state of the art. The implications for clinical care contained in this model are vast. SPECT imaging with PIB is a very accurate means of identifying very early signs of AD that needs to be treated after prevent further deterioration, lessening the patient’s discomfort and saving money for the healthcare industry in the long run. Because the failures of this approach have been widely identified in early stage detection, it can, therefore, be greatly beneficial to lower the social and economic implications of AD. The Hybrid AI Model therefore offers a potential solution to the problem of developing better, more efficient approaches to diagnosing Alzheimer’s – an issue that could in turn dramatically transform current clinicians’ ability to identify this terrible disease.
|
|---|