Early Detection of Dementia: A Machine Learning Based Classification on MRI Images and Insights from ECG Signals

Dementia, including Alzheimer’s Disease, is a progressive brain condition that affects a person’s thinking, memory, and behavior. While it cannot be cured, early diagnosis enables the use of treatment options that may help manage symptoms and support the individual’s independence. Early detection of dementia from ECG signals is possible; more importantly, cognitive decline in humans can be identified through Heart Rate Variability (HRV) features. In the context of mental health-related illnesses, various early diagnosis techniques are employed, such as detecting structural abnormalities in MRI scans or identifying symptoms of dementia through changes in physiological signals like ECG. This study focuses on developing an advanced deep learning model for Alzheimer’s disease (AD) classification using brain scans. The same 3D CNN model was evaluated on a recently gathered clinical MRI dataset, showing accurate predictions after successfully categorizing AD and Cognitively Normal (CN) cases using the ADNI dataset. The results of the experiment have the potential to assist physicians in the early diagnosis of Alzheimer’s disease, leading to timely treatment and improved patient outcomes. Additionally, this work contributes to the advancement of healthcare technology by opening new possibilities for improving deep learning models in medical image analysis.