eng Oriental Scientific Publishing Company Biomedical and Pharmacology Journal 0974-6242 2025-10-20 18 October Spl Edition 68400 article Tulip Das 1 Chinmaya Kumar Nayak 1 Parthasarathi Pattnayak 2 Faculty of Engineering and Technology, SRI SRI University, Cuttack, India School of Computer Applications, KIIT Deemed to be University, Bhubaneswar, India Alzheimer's disease (AD), also called dementia, is a neurological illness that progressively impairs memory and cognitive functioning. It hinders critical mental processes like reasoning, memory, and judgement. While precise diagnosis is often achieved through the use of standard imaging techniques such as CT, MRI, and PET, deep learning (DL) frameworks provide a quicker, less technologically demanding, and less human intervention method. These models, which have been validated by clinically collected medical data, also become more accurate with time and can help predict AD. This paper presents a model built on AI that consists of two main stages: a voting classifier that uses transfer learning (TL) and machine learning (ML) based on permutations. In the first stage of the model, DenseNet-121 and DenseNet-201 are used for feature extraction. In the second stage, three distinct machine learning classifiers are employed for classification: Naïve Bayes, Support Vector Machine (SVM) and XG Boost. To test the model, 6,400 photos from Kaggle's training dataset were used. The results show that this suggested model works better than current cutting-edge methods. Based on these results, the model may be used to create clinically useful techniques for AD diagnosis using MRI images. https://biomedpharmajournal.org/vol18octoberspledition/enhanced-hybrid-ai-based-framework-for-improving-the-detection-and-diagnosis-of-alzheimers-disease/ Alzheimer’s Disease DenseNet-121 DenseNet-201 Gaussian Naïve Base SVM XG Boost