Enhancing Chronic Disease Prediction in IoMT-Enabled Healthcare 5.0 Using Deep Machine Learning: Alzheimer’s Disease as a Case Study

The advancement of Internet of Medical Things (IoMT) in healthcare 5.0 has opened new avenues for proactive and precise chronic disease prediction. This study proposes an integrated framework that leverages image processing and deep learning techniques, specifically DenseNet-201, to classify and stage multiple chronic diseases, including pancreatic diseases, brain tumors, and Alzheimer’s disease, using medical imaging datasets. Ultrasound images of the pancreas were analysed to identify Pancreatic Cancer, Pancreatitis, and Pancreatic Neuroendocrine Tumor (PNET), while MRI scans were employed for brain tumor detection and classification into Glioma (Stage II), Meningioma, Pituitary tumor, and No Tumor categories. Alzheimer’s MRI datasets were also utilized to categorize Early-Onset, Late-Onset, and Familial Alzheimer’s cases. The proposed approach incorporates automated preprocessing, feature extraction, and classification pipelines, enabling accurate differentiation among disease types and stages. In brain tumor detection, if a tumor is identified, the system further predicts its stage (Stage 1, Stage 2, or Stage 3), providing critical insights for clinical decision-making. The framework was validated using publicly available datasets, demonstrating high classification accuracy, reliability, and scalability. Overall, this study highlights the potential of IoMT-enabled deep learning frameworks in enhancing early diagnosis, treatment planning, and personalized care for chronic disease management in modern healthcare environments.

Keywords: Chronic Disease Prediction, Deep Learning, Densenet-201, U-Net Segmentation, Dementia Classification.