Symptoms of Alzheimer's disease are dullness, memory loss, and an inability to socialize. These symptoms of Alzheimer's disease are caused by the shrinking of the brain, which leads to the death of brain cells and prevents the patient from accessing his or her memories and making normal decisions. Alzheimer's disease affects about 10 percent of people over the age of 65 in the United States. But even with such a large number of patients, scientists still don't know much about Alzheimer's disease. Currently, scientists believe that the dominant causes of Alzheimer's are age and family history. The mainstay of diagnosing Alzheimer's disease is a combination of patient history, neuropsychology, and MRIs, as well as the use of medications to slow down the disease (Alzheimer's disease is not curable at the current state of medical care). However, as the United States and a growing number of countries around the world experience declining fertility rates and aging societies, the number of patients with Alzheimer's disease is beginning to increase. As a result, society's medical resources are becoming overwhelmed by the diagnosis of Alzheimer's disease. Society needs a more efficient way to diagnose Alzheimer's disease. Therefore, some scholars began to try to use AI deep learning to carry out the prediction of Alzheimer's disease. After validation, the network model trained using AI deep learning can now achieve a diagnosis of Alzheimer's disease accuracy better than that of ordinary neurologists. Thus, it can be seen that the direction of AI deep learning is feasible for the future to solve the problem of overloading the social healthcare system with Alzheimer's diagnosis. This paper is based on the FCN model in Development and Validation of an Interpretable Deep Learning Framework for Alzheimer's Disease Classification. We also used the same ANDI dataset as that paper. The model is an FCN and it randomly selects 3000 voxels of size 47x47x47 from each MRI scan for each training session #go back to the FCN section of the paper#. This FCN model is divided into six convolutional blocks. The first four convolutional blocks are 3D Convolutional Layer #explain more about these four convolutional blocks#. The remaining two convolutional blocks act as speedups for model efficiency #explain more about these two convolutional blocks convolutional blocks (speedups for model efficiency)# . After training, the final model that comes out can successfully predict the MRI images. After completing the reconstruction of this model and going through 3000 epochs of training, the model is slightly less accurate #To be completed#. But the model already has the ability to make predictions on MRI images.

Mengyu Ge. Prediction and Visualization of Alzheimer's Disease Using Deep Learning Models. Academic Journal of Medicine & Health Sciences (2023) Vol. 4, Issue 10: 57-60. https://doi.org/10.25236/AJMHS.2023.041009.

[1] Acquarelli J; van Laarhoven T; Postma GJ; Jansen JJ;Rijpma A; van Asten S; Heerschap A;Buydens LMC; Marchiori E; “Convolutional Neural Networks to Predict Brain Tumor Grades and Alzheimer’s Disease with Mr Spectroscopic Imaging Data. ” PloS One, U. S. National Library of Medicine, pubmed. ncbi.nlm.nih.gov/36001537/.Accessed 12 Sept. 2023. 

[2] Swapna. “Convolutional Neural Network: Deep Learning.” Developers Breach, 25 Jan. 2022, developersbreach.com/convolution-neural-network-deep-learning/.  

[3] Balachandran, Sandeep. “Machine Learning - Max Pooling with Color Images.” DEV Community, DEV Community, 30 Mar. 2020, dev.to/sandeepbalachandran/machine-learning-max-pooling-with-color-images-2i21. 

[4] “A Comprehensive Guide to Convolutional Neural Networks - the Eli5 Way.” Saturn Cloud Blog, 17 June 2023, saturncloud.io/blog/a-comprehensive-guide-to-convolutional-neural-networks-the-eli5-way/. 

[5] Gauriau R; Bizzo BC; Kitamura FC; Landi Junior O; Ferraciolli SF; Macruz FBC; Sanchez TA; Garcia MRT; Vedolin LM; Domingues RC; Gasparetto EL; Andriole KP; “A Deep Learning-Based Model for Detecting Abnormalities on Brain MR Images for Triaging: Preliminary Results from a Multisite Experience. ” Radiology. Artificial Intelligence, U. S. National Library of Medicine, pubmed. ncbi. nlm. nih. gov/34350408/. Accessed 12 Sept. 2023. 

[6] Gradient-Based Learning Applied to Document Recognition, ieeexplore.ieee.org/document/726791. Accessed 12 Sept. 2023. 

[7] Krishnapriya, Srigiri, and Yepuganti Karuna. “Pre-Trained Deep Learning Models for Brain MRI Image Classification.” Frontiers, Frontiers, 6 Mar. 2023, www. frontiersin.org/articles/10. 3389/fnhum. 2023. 1150120/full. 

[8] Long, Jonathan, et al. “Fully Convolutional Networks for Semantic Segmentation.” arXiv.Org,8 Mar.2015,arxiv.org/abs/1411.4038.