Downloads: 0
Student Project | Computer Science and Engineering | Volume 15 Issue 3, March 2026 | Pages: 751 - 756 | India
Alzheimer's Detection Using Machine Learning
Abstract: Alzheimer's Disease (AD) is a progressive neurodegenerative disease that results in the impairment of cognition and memory loss, hence early diagnosis is crucial in planning effective treatment and care of patients. Conventional diagnostic methods are based on clinical evaluations and manual analysis of MRI images that may be time-wasting, subjective, and vulnerable to human error. In order to overcome such shortcomings, the current work proposes an automated brain MRI-based Alzheimer detector on the basis of a Vision Transformer (ViT) model. The system integrates a preprocessing pipeline with a structured design and resizing, normalization, and data augmentation to control in a manner that it does not affect the integrity of a medical image. ViT model can take advantage of attention-based learning of features and, therefore, it can represent long-range spatial information in MRI structures, which results in better discrimination of four cognitive states: No Impairment, Very Mild, Mild, and Moderate Alzheimer. When evaluated experimentally, the approach proves to have an improved classification and strong generalization when compared with the traditional convolutional architecture. The trained model is incorporated in a Flask-based Web application, which allows real-time inference and producing diagnostic reports to support clinical work. The findings suggest that the suggested method can be of value to the early AD screening and offer a scalable and accessible instrument to the healthcare setting.
Keywords: Alzheimer's Disease, MRI Classification, Vision Transformer, Deep Learning, Computer-Aided Diagnosis, Medical Imaging
How to Cite?: Mithun S, Jeffrin Hannah, "Alzheimer's Detection Using Machine Learning", Volume 15 Issue 3, March 2026, International Journal of Science and Research (IJSR), Pages: 751-756, https://www.ijsr.net/getabstract.php?paperid=SR26302191354, DOI: https://dx.dx.doi.org/10.21275/SR26302191354