Purpose <p>The brain cells of humans are severely affected by Alzheimer’s Disease (AD). This disease is considered a severe sickness that creates damage to the brain cells of humans and shows symptoms like aging and stress. Therefore, it is necessary to design an accurate diagnosis model. Based on the report of the World Health Organization (WHO), AD is positioned in fifth place based on the death rate of the patients. The presence of neurological and biological features is the most important source to determine AD.</p> Method <p>The deep learning technique uses Magnetic Resonance Imaging (MRI) to detect AD. The diagnostic accuracy is obtained by identifying the volumetric changes in the MRI images. Nevertheless, it is very difficult to extort the features from the MRI images in an automatic way; therefore, the detection of AD is thought of as the most difficult task. In the medical imaging fields, Computer-Aided Diagnosis (CAD) and machine learning have gained demand in recent days. Therefore, this survey investigates different methodologies for identifying AD using MRI images. This work follows the introduction, literature review, and chronological analysis of the AD detection methods. Also, it provides the datasets used for taking the MRI images for achieving the objectives of disease detection. It also explores different landmark detection processes for AD diagnosis. Then, it explains diverse heuristic algorithms for tuning the parameters and various classification processes for detecting AD. Further, it exhibits the information of implementation tools and algorithms utilized for detection.</p> Result and conclusion <p>Diverse evaluation metrics, merits, and demerits are considered to illustrate the efficiency. Finally, the research gaps and challenging factors are given to direct the future development of the AD recognition model.</p>