AI-Based Doctor Agent for Early Detection of Alzheimer's Applying Hybrid Machine-Learning Techniques
How to cite (COMIEN) :
Alzheimer's disease (AD) is a progressive neurodegenerative disorder characterized by memory loss, cognitive decline, and irreversible brain death, rendering it a global health concern. Even though early detection using machine learning (ML) has gotten better, there are still problems with expensive neuroimaging, limited multimodal data integration, small homogeneous cohorts, and models that are hard to understand. This study proposes an AI-driven "doctor" agent that forecasts Alzheimer's disease state, progression risk, and personalized therapy recommendations, grounded in standard clinical and cognitive indicators. We used Mutual Information-based feature selection and an 80/20 stratified methodology to test seven hybrid ensemble models on 2,149 Kaggle patients. The second hybrid model had an accuracy of 95.81% and an AUC of 0.9492, which was better than both classical and deep-learning baselines. All hybrid models did better than individual learners, showing that they could work well with a wide range of clinical data. With alzheime.pkl, it was possible to make real-time predictions with confidence ratings and comments on diagnoses that were like those made by doctors. Real-time diagnostic results that can be understood, along with confidence ratings and automated "doctor's notes," help research move from the lab to the clinic. Multimodal fusion, external validation, calibrated probability calculation, and explainable decision outputs make Alzheimer's screening fairer, faster, and clearer. The study shortens the time it takes to make a diagnosis and the amount of work that doctors must do, which is good for healthcare systems. Longitudinal, multi-omics, and wearable sensor data will be combined with privacy-preserving federated learning and explainable-AI dashboards to help doctors understand things better and use them in the real world.
"Dementia," World Health Organization. Accessed: Nov. 17, 2025. [Online]. Available: https://www.who.int/news-room/fact-sheets/detail/dementia.
"Dementia," Wikipedia, Jul. 11, 2025. Accessed: Nov. 17, 2025. [Online]. Available: https://en.wikipedia.org/w/index.php?title=Dementia&oldid=1300017519
C. Kavitha, V. Mani, S. R. Srividhya, O. I. Khalaf, and C. A. T. Romero, "Early-Stage Alzheimer’s Disease Prediction Using Machine Learning Models," Front. Public Health, vol. 10, Mar. 2022, doi: 10.3389/fpubh.2022.853294.
"Alzheimer's Disease Dataset," Kaggle. Accessed: Nov. 17, 2025. [Online]. Available: https://www.kaggle.com/datasets/rabieelkharoua/alzheimers-disease-dataset.
M. Sudharsan and G. Thailambal, "Alzheimer’s disease prediction using machine learning techniques and principal component analysis (PCA)," Mater. Today: Proc., vol. 81, pp. 182–190, Jan. 2023, doi:10.1016/j.matpr.2021.03.061.
P. Rani, R. Lamba, R. K. Sachdeva, K. Kumar, and C. Iwendi, "A machine learning model for Alzheimer’s disease prediction," IET Cyber-Phys. Syst.: Theory Appl., vol. 9, no. 2, pp. 125–134, 2024, doi:10.1049/cps2.12090.
S. T. Ahmed and S. M. Kadhem, "Alzheimer’s disease prediction using three machine learning methods," Indones. J. Electr. Eng. Comput. Sci., vol. 27, no. 3, pp. 1689–1697, Sep. 2022, doi:10.11591/ijeecs.v27.i3.pp1689-1697.
B. K. Karaman, E. C. Mormino, M. R. Sabuncu, and Alzheimer’s Disease Neuroimaging Initiative, "Machine learning based multi-modal prediction of future decline toward Alzheimer’s disease: An empirical study," PLoS One, vol. 17, no. 11, Nov. 2022, doi:10.1371/journal.pone.0277322.
W. Khan et al., "An explainable and efficient deep learning framework for EEG-based diagnosis of Alzheimer’s disease and frontotemporal dementia," Front. Med., vol. 12, Jul. 2025, doi:10.3389/fmed.2025.1590201.
F. Gelir, T. Akan, S. Alp et al., "Machine Learning Approaches for Predicting Progression to Alzheimer’s Disease in Patients with Mild Cognitive Impairment," J. Med. Biol. Eng., vol. 45, pp. 63–83, 2025, doi: 10.1007/s40846-024-00918-z.
N. P. D, H. M. V, D. C, S. S, and A. K. S, "Alzheimer’s Disease Prediction Using Machine Learning Methodologies," in Proc. 2022 Int. Conf. Comput. Commun. Informat. (ICCCI), Jan. 2022, pp. 1–6, doi: 10.1109/ICCCI54379.2022.9740942.
A. S. Alatrany, W. Khan, A. Hussain, et al., "An explainable machine learning approach for Alzheimer’s disease classification," Sci. Rep., vol. 14, 2024, Art. no. 2637, doi: 10.1038/s41598-024-51985-w.
A. N. Upadhyaya et al., "Alzheimer Disease Classification Model Using Machine Learning Techniques," Int. J. Basic Appl. Sci., vol. 14, no. 3, pp. 396–405, Jul. 2025, doi: 10.14419/rcsfqr61.
"Alzheimer's Disease Fact Sheet," National Institute on Aging. Accessed: Nov. 17, 2025. [Online]. Available: https://www.nia.nih.gov/health/alzheimers-and-dementia/alzheimers-disease-fact-sheet.
"Medical Tests for Diagnosing Alzheimer’s & Dementia," alz.org. Accessed: Nov. 17, 2025. [Online]. Available: https://www.alz.org/alzheimers-dementia/diagnosis/medical_tests