Authors:
Malini Premakumari William, S. Silvia Priscila, S. Briskline Kiruba, T. R. Nisha Dayana, P. Velavan

Addresses:
Department of Computer Systems Engineer, Mirage Software Inc., DBA Bourntec Solutions, Schaumburg, Illinois, United States of America. Department of Computer Science, Bharath Institute of Higher Education and Research, Chennai, Tamil Nadu, India. Department of Computer Science, Jaya College of Arts and Science, Chennai, Tamil Nadu, India. Department of Computer Science, Vels Institute of Science Technology and Advance Studies, Chennai, Tamil Nadu, India. Department of Computer Science and Business Systems, Dhaanish Ahmed College of Engineering, Chennai, Tamil Nadu, India.              

Abstract:

Alzheimer’s disease has been a serious public health issue that needs prompt and precise diagnosis in the management and treatment phase. The following paper presents a machine learning system mainly engineered to classify AD via neuroimaging and clinical features comprising air time1, disp index1, gmrt in air1, max x extension1, and max y extension1 to detect subtle patterns possibly indicative of Alzheimer’s pathology. Systematic data preprocessing and visualization techniques, such as correlation heatmaps and 3D scatter plots, reflect complicated interdependencies in AD progression. The classifiers utilized include Cat Boost and Light GBM for AD and healthy controls, and AdaBoost for mild cognitive impairment. Model performance is measured by accuracy, precision, recall, F1 score, classification reports, and confusion matrices, helping to discover model strengths and weaknesses. A learning curve showed that the models are general and flexible enough for practical circumstances. The work shows that machine learning can integrate multi-modal data modalities into AD diagnosis. Thus, it improves diagnostic accuracy and enables personalized treatment strategies, improving patient outcomes and supporting cutting-edge neurodegenerative disease clinical decisions.

Keywords: Alzheimer’s Disease (AD); Precise Diagnosis; Robust Machine Learning System; Correlation Heatmaps; Unravel Complex Data; Underlying Mechanisms; Distinguishing Nuanced; Adaboost Classifiers.

Received on: 27/04/2024, Revised on: 05/07/2024, Accepted on: 29/08/2024, Published on: 01/09/2024

DOI: 10.69888/FTSHSL.2024.000253

FMDB Transactions on Sustainable Health Science Letters, 2024 Vol. 2 No. 3, Pages: 175-187

• Views : 119
• Downloads : 11

Download PDF