An efficient deep learning algorithm for early stage Alzheimer’s disease identification using CNN compared with AdaBoost classifier

The purpose of this research was to evaluate two methods for improving the accuracy rate of AD prediction in MRI images: AdaBoost (AB) and a novel convolutional neural network (CNN) technique. Materials and Methods: After being extracted from the ADNI database, the experimental data goes through a series of image preparation and feature extraction procedures. There were a total of 26 people included in the sample for the AD diagnosis; 13 from Group 1 and 13 from Group 2. Alpha and beta were used in the computation, along with a G-power of 0.8, values of 0.05 and 0.2, and a confidence range of 95%. In order to forecast the presence of Alzheimer’s disease in magnetic resonance imaging (MRI) scans, a combination of the innovative AdaBoost (AB) model and a Convolutional Neural Network (CNN) classifier using a large sample size (N=13) was employed. Results: The novel convolutional Neural Network (CNN) classifier outperforms the AdaBoost (AB) model, which achieved an accuracy rate of 82.96%, by a wide margin. Discussion and conclusion:-It was decided that p = 0.001 (p<0.05) was the level of significance for the inquiry. This provides more evidence that the test groups are distinct from one another. Finally, for MRI-based Alzheimer disease prediction, the novel convolutional Neural Network (CNN) classifier outperforms the AdaBoost (AB) with a higher accuracy of 93.78%.