Dementia refers to a group of conditions and diseases that affect cognitive functions such as memory, reasoning, personality, mood, behaviour, and the ability to think clearly. This cognitive decline interferes with everyday tasks and routines. Studies suggest that nearly 60% of individuals aged 60 and above are affected by dementia. Deep learning techniques have shown success in numerous medical imaging applications, including the identification of dementia. These methods are capable of learning and extracting significant features from large volumes of data, making them well-suited for analyzing medical images. In this work, we present a comparative analysis aimed at accurately identifying dementia using MRI scans. For this purpose, we utilize a Kaggle dataset, which, despite having a relatively small size of around 6400 images, includes a large number of samples. The focus of this paper is on multi-classification problems involving various deep learning models.