JOURNAL OF ALGEBRAIC STATISTICS Volume 13, No. 3, 2022, p. 4137-4146 https://publishoa.com ISSN: 1309-3452 4137 An Optimized Machine Learning Framework for Detecting Alzheimer's Disease By MRI Dr. T. S. Suganya Dr. K. Geetha Dr. C. Rajan Assistant Professor, Dept of CA Professor, Dept of CSE Professor, Dept of IT SRMIST Excel Engineering college K.S.R College of Technology Ramapuram Campus,India Kumarapalayam, Namakkal, India Namakkal,India tssuganya07@gmail.com geetharajsri@gmail.com rajancsg@gmail.com ABSTRACT Machine learning has extensive application in diverse medical fields.With advancements in medical technologies, access has been given to data for the identification of diseases in theirearly stages. Alzheimer's Disease (AD) is a chronic illnessthat will cause degeneration of the brain cells and ultimately will lead to memoryloss. AD causedcognitive mental problems like forgetfulness and confusion, as well as other symptoms such aspsychologicaland behavioralproblems, are further recommended to undergo test procedures usingneuroimagingtechniques. This work's objective is to utilize the machinelearning algorithms for processing the data acquired via neuroimaging technologies for early-stage AD detection. The framework extracts featuresusingcurvelet transform from MRI brain image. This work will also present the Decision Tree, the Adaptive Boosting (AdaBoost), and the Extreme Gradient Boosting (XGBoost) classifiers. In machine learning, Population-Based Incremental Learning (PBIL) is an optimization algorithm, in spite of being simpler than a conventional genetic algorithm, the PBIL algorithm is able to achieve much better results in several cases.PBIL is used to optimize the AdaBoost and XGBoost classifiers to improve AD classification. The experimental outcomes will demonstrate the proposed approach's superior performance over that of other existing approaches. Keywords: Alzheimer's disease (AD), Machine Learning, Curvelet Transform, Decision Tree, Adaptive Boosting (AdaBoost), Extreme Gradient Boosting (XGBoost), and Population-Based Incremental Learning (PBIL). 1 INTRODUCTION Alois Alzheimer, a German physicist and neuro-pathologist, was the first person to identify as well as to discuss about theAlzheimer'sDisease (AD). As per the World Alzheimer Report, around 50 millionpeopleacross the globe were affected by dementia in the year 2018,andabout two-third of that population were suffering from AD. There will be about 152 million AD patients in the year 2050, and this disease's cost has been forecasted to be USD 2 trillion in the year 2030. At present, the AD treatment is predominantly involved with the usage of either Alzheimer's disease-modifying or delaying drugs instead of drugs which are able to either reverse or permanently stop the disease's progression. Hence, it is essential for early-stage AD prediction so as to make it feasible to delay the effects of the disease [1]. AD is characterized by the loss of neurons as well as synapses within the cerebral cortex,and also specific subcortical regions, that will result in gross atrophy of the affected regions, which is inclusive of degeneration in the temporal lobe andparietal lobe, and also portions of the frontal cortex as well as the cingulate gyrus. Earlier studies have shown the correlation between several impairments in the AD as well as the atrophy in many regions such as the amygdala,the temporal lobe, and the hippocampus. These characteristics are used for delineating the AD patients from the normal patients. The key focus of the researchers is to monitor the change in a patient's health, the disease's clinical progression as well as reaction to the therapy. However, they find it most cumbersome to identify relevant bio-markers which are good representations of the AD as well as the Mild Cognitive Impairment (MCI). The researchers' objective is inclusive of diagnosingearly-stage AD as well as identifyingthe individuals who are at most risk for AD development. Magnetic Resonance Imaging (MRI) is employed by physicians to diagnose AD. The multi-class classification of AD, MCI as well as Normal Control (NC) will employ as biomarkers the individual or combined structural MRIbiomarkerslike the hippocampus's shape as well as texture, corticalmeasurements, and volume measurements [2]. The MRI's key role in the AD analysis is to assess the volume alteration in the characteristic positions so as to provide up to 87% of analytical accuracy of up to 87%. Quite often, the appraisal is carried out on the mesial temporal lobe atrophyand the temporoparietalcortical atrophy. Direct or indirect estimation is done for the mesial temporal lobe atrophy. While the direct estimation is based on measuring the volume loss of hippocampal orparahippocampal tissue, the indirect estimation is dependent ontheparahippocampalfissures' magnification. Normally, analysis of these estimations is done along with the medial temporalatrophy score, that has been proved to be predictive of the progression from MCI to dementia[3].