Novel approach to early prediction of alzheimer's disease progression using integrated deep regulatory genetic neural network and optimized deep belief networks.

UNLABELLED: To enhance the prediction progression of Alzheimer's Disease is very excavating process.

PROBLEM STATEMENT: It is often very difficult to implement in the chronic neurodegenerative disease-related preprocessed gene expression data. Especially, Alzheimer's Disease (AD) prediction is a very crucial process in AD metadata diagnosis. Novelty: To explore this challenging prediction process in brain disease prediction, this research presents a proposed deep learning model, namely the Integrated Deep Regulatory Genetic Neural Network and Optimised Deep Belief Networks (IDRODN). This integration increases the affluence of prediction progression from genomic data. These prediction systems help identify early AD.

METHOD: This research utilizes the IDRODN, which can predict and confine each network's neurons and hidden layers against the benchmark dataset of Alzheimer's gene expression and uncertainty to predict Alzheimer's Disease.

KEY

RESULTS

The comparative analysis on data from the Alzheimer's disease gene expression data Initiative database has achieved an accuracy of 98.3%. In addition, it has achieved a high F1 score of 0.986 for predicting different stages from Gene expression data.

IMPLICATIONS: This shows the most accurate technique for predicting Alzheimer's Disease (AD) using the prognostic IDRODN model.