Optimización e Interpretabilidad de Modelos Clasificadores para Alzheimer: Un Estudio Basado en el Conjunto de Datos OASIS-2

Heber Zapata; Olanda Prieto-Ordaz; Raymundo Cornejo

doi:10.47756/aihc.y10i1.191

Authors

Heber Zapata Autonomous University of Chihuahua https://orcid.org/0009-0002-7987-9750
Olanda Prieto-Ordaz Autonomous University of Chihuahua https://orcid.org/0000-0002-0023-9535
Raymundo Cornejo Autonomous University of Chihuahua https://orcid.org/0000-0001-7229-1941

DOI:

https://doi.org/10.47756/aihc.y10i1.191

Keywords:

Alzheimer Disease, Medical Applications, Machine Learning, OASIS-2, healthcare applications

Abstract

Early diagnosis of Alzheimer’s disease is crucial to slowing its progression and improving patients’ quality of life. This study evaluates the performance of three machine learning models—logistic regression, support vector machine, and random forest—using the OASIS-2 dataset under two conditions: considering only the first visit of each participant and including all available longitudinal measurements. A standardized preprocessing pipeline, cross-validation, and grid search were applied to optimize the models. Models trained on the extended dataset achieved 96% accuracy, outperforming previously reported results. Explainability techniques, including K-Means clustering and local interpretable model-agnostic explanations, were applied to analyze misclassified instances, revealing that several errors were caused by mislabeled “converted” patients rather than model deficiencies. These findings highlight that classification performance is sensitive to data quality and labeling consistency. The results emphasize the need for rigorous data collection and curation procedures to ensure fairness and clinical applicability of predictive models. Future work should focus on building more representative longitudinal datasets and exploring additional explainability techniques to reduce potential bias and enhance the trustworthiness of early diagnostic systems.

Downloads

Download data is not yet available.

References

Swanberg, M. M., Tractenberg, R. E., Mohs, R., Thal, L. J. and Cummings, J. L. “Executive Dysfunction in Alzheimer Disease”, Arch Neurol, vol. 61, núm. 4, p. 556, abr. 2004, doi: 10.1001/archneur.61.4.556. DOI: https://doi.org/10.1001/archneur.61.4.556

Frias, C. E., Cabrera, E. and Zabalegui, A. “Informal Caregivers’ Roles in Dementia: The Impact on Their Quality of Life”, Life, vol. 10, núm. 11, p. 251, oct. 2020, doi: 10.3390/life10110251. DOI: https://doi.org/10.3390/life10110251

Goren, A., Montgomery, W., Kahle-Wrobleski, K., Nakamura, T. and Ueda, K. “Impact of caring for persons with Alzheimer’s disease or dementia on caregivers’ health outcomes: findings from a community based survey in Japan”, BMC Geriatr, vol. 16, núm. 1, p. 122, dic. 2016, doi: 10.1186/s12877-016-0298-y. DOI: https://doi.org/10.1186/s12877-016-0298-y

Fathi, S., Ahmadi, M. and Dehnad, A. “Early diagnosis of Alzheimer’s disease based on deep learning: A systematic review”, Computers in Biology and Medicine, vol. 146, p. 105634, jul. 2022, doi: 10.1016/j.compbiomed.2022.105634. DOI: https://doi.org/10.1016/j.compbiomed.2022.105634

Tan, W. Y., Hargreaves, C., Chen, C. and Hilal, S. “A Machine Learning Approach for Early Diagnosis of Cognitive Impairment Using Population-Based Data”, JAD, vol. 91, núm. 1, pp. 449–461, ene. 2023, doi: 10.3233/JAD-220776. DOI: https://doi.org/10.3233/JAD-220776

Diogo, V. S., Ferreira, H. A., Prata, D. and for the Alzheimer’s Disease Neuroimaging Initiative, “Early diagnosis of Alzheimer’s disease using machine learning: a multi-diagnostic, generalizable approach”, Alz Res Therapy, vol. 14, núm. 1, p. 107, dic. 2022, doi: 10.1186/s13195-022-01047-y. DOI: https://doi.org/10.1186/s13195-022-01047-y

Ferrara, E. “The Butterfly Effect in artificial intelligence systems: Implications for AI bias and fairness”, Machine Learning with Applications, vol. 15, p. 100525, mar. 2024, doi: 10.1016/j.mlwa.2024.100525. DOI: https://doi.org/10.1016/j.mlwa.2024.100525

Khan A. and Zubair, S. “Longitudinal Magnetic Resonance Imaging as a Potential Correlate in the Diagnosis of Alzheimer Disease: Exploratory Data Analysis”, JMIR Biomed Eng, vol. 5, núm. 1, p. e14389, abr. 2020, doi: 10.2196/14389. DOI: https://doi.org/10.2196/14389

Cockrell, J. R. and Folstein, M. F. “Mini-Mental State Examination (MMSE)”, Psychopharmacol Bull, vol. 24, núm. 4, pp. 689–692, 1988.

Morris, J. C. et al., “Clinical Dementia Rating training and reliability in multicenter studies: The Alzheimer’s Disease Cooperative Study experience”, Neurology, vol. 48, núm. 6, pp. 1508–1510, jun. 1997, doi: 10.1212/WNL.48.6.1508. DOI: https://doi.org/10.1212/WNL.48.6.1508

Marcus, D. S., Fotenos, A. F., Csernansky, J. G., Morris, J. C., and Buckner, R. L. “Open Access Series of Imaging Studies: Longitudinal MRI Data in Nondemented and Demented Older Adults”, Journal of Cognitive Neuroscience, vol. 22, núm. 12, pp. 2677–2684, dic. 2010, doi: 10.1162/jocn.2009.21407. DOI: https://doi.org/10.1162/jocn.2009.21407

Choraś, M., Pawlicki, M., Puchalski, D. and Kozik, R. “Machine Learning – The Results Are Not the only Thing that Matters! What About Security, Explainability and Fairness?”, en Computational Science – ICCS 2020, vol. 12140, V. V. Krzhizhanovskaya, G. Závodszky, M. H. Lees, J. J. Dongarra, P. M. A. Sloot, S. Brissos, y J. Teixeira, Eds., en Lecture Notes in Computer Science, vol. 12140., Cham: Springer International Publishing, 2020, pp. 615–628. doi: 10.1007/978-3-030-50423-6_46. DOI: https://doi.org/10.1007/978-3-030-50423-6_46

Cutillo, C. M. et al., “Machine intelligence in healthcare—perspectives on trustworthiness, explainability, usability, and transparency”, npj Digit. Med., vol. 3, núm. 1, p. 47, mar. 2020, doi: 10.1038/s41746-020-0254-2. DOI: https://doi.org/10.1038/s41746-020-0254-2

Dosilovic, F. K., Brcic, M. and Hlupic, N. “Explainable artificial intelligence: A survey”, en 2018 41st International Convention on Information and Communication Technology, Electronics and Microelectronics (MIPRO), Opatija: IEEE, may 2018, pp. 0210–0215. doi: 10.23919/MIPRO.2018.8400040. DOI: https://doi.org/10.23919/MIPRO.2018.8400040

Barocas, S. and Selbst, A.D. “Big Data's Disparate Impact”, 2016, doi: 10.15779/Z38BG31. DOI: https://doi.org/10.2139/ssrn.2477899

Battineni, G., Chintalapudi, N. and Amenta, F. “Machine learning in medicine: Performance calculation of dementia prediction by support vector machines (SVM)”, Informatics in Medicine Unlocked, vol. 16, p. 100200, 2019, doi: 10.1016/j.imu.2019.100200. DOI: https://doi.org/10.1016/j.imu.2019.100200

Sørensen, L. and Nielsen, M. “Ensemble support vector machine classification of dementia using structural MRI and mini-mental state examination”, Journal of Neuroscience Methods, vol. 302, pp. 66–74, may 2018, doi: 10.1016/j.jneumeth.2018.01.003. DOI: https://doi.org/10.1016/j.jneumeth.2018.01.003

Bari Antor, M. et al., “A Comparative Analysis of Machine Learning Algorithms to Predict Alzheimer’s Disease”, Journal of Healthcare Engineering, vol. 2021, pp. 1–12, jul. 2021, doi: 10.1155/2021/9917919. DOI: https://doi.org/10.1155/2021/9917919

Khan A. and Zubair, S. “An Improved Multi-Modal based Machine Learning Approach for the Prognosis of Alzheimer’s disease”, Journal of King Saud University - Computer and Information Sciences, vol. 34, núm. 6, pp. 2688–2706, jun. 2022, doi: 10.1016/j.jksuci.2020.04.004. DOI: https://doi.org/10.1016/j.jksuci.2020.04.004

Morales‐Forero, A., Rueda Jaime, L., Gil‐Quiñones, S.R., Barrera Montañez, M.Y., Bassetto, S. and Coatanea, E. “An insight into racial bias in dermoscopy repositories: A HAM10000 data set analysis”, JEADV Clinical Practice, vol. 3, núm. 3, pp. 836–843, jul. 2024, doi: 10.1002/jvc2.477. DOI: https://doi.org/10.1002/jvc2.477

Obermeyer, Z., Powers, B., Vogeli, C. and Mullainathan, S. “Dissecting racial bias in an algorithm used to manage the health of populations”, Science, vol. 366, núm. 6464, pp. 447–453, oct. 2019, doi: 10.1126/science.aax2342. DOI: https://doi.org/10.1126/science.aax2342

Vellido, A. “The importance of interpretability and visualization in machine learning for applications in medicine and health care”, Neural Comput & Applic, vol. 32, núm. 24, pp. 18069–18083, dic. 2020, doi: 10.1007/s00521-019-04051-w. DOI: https://doi.org/10.1007/s00521-019-04051-w

Hassan, S. U., Abdulkadir, S. J., Zahid, M. S. M. and Al-Selwi, S. M. “Local interpretable model-agnostic explanation approach for medical imaging analysis: A systematic literature review”, Computers in Biology and Medicine, vol. 185, p. 109569, feb. 2025, doi: 10.1016/j.compbiomed.2024.109569. DOI: https://doi.org/10.1016/j.compbiomed.2024.109569

Shaikh, A. S., Samant, R. M., Patil, K. S., Patil, N. R. and Mirkale, A. R. “Review on Explainable AI by using LIME and SHAP Models for Healthcare Domain”, IJCA, vol. 185, núm. 45, pp. 18–23, nov. 2023, doi: 10.5120/ijca2023923263. DOI: https://doi.org/10.5120/ijca2023923263

Magesh, P. R., Myloth, R. D. and Tom, R. J. “An Explainable Machine Learning Model for Early Detection of Parkinson’s Disease using LIME on DaTSCAN Imagery”, Computers in Biology and Medicine, vol. 126, p. 104041, nov. 2020, doi: 10.1016/j.compbiomed.2020.104041. DOI: https://doi.org/10.1016/j.compbiomed.2020.104041

Falvo F. R. and Cannataro, M. “Explainability techniques for Artificial Intelligence models in medical diagnostic”, en 2024 IEEE International Conference on Bioinformatics and Biomedicine (BIBM), Lisbon, Portugal: IEEE, dic. 2024, pp. 6907–6913. doi: 10.1109/BIBM62325.2024.10821826. DOI: https://doi.org/10.1109/BIBM62325.2024.10821826

Guzmán Ponce, A., López-Bautista, J. and Fernandez-Beltran, R. “Interpretando Modelos de IA en Cáncer de Mama con SHAP y LIME”, IRCI, vol. 2, núm. 2, p. 15, jul. 2024, doi: 10.36677/ideaseningenieria.v2i2.23952. DOI: https://doi.org/10.36677/ideaseningenieria.v2i2.23952

Varghese, A., Sherimon, V., Raja, X. C., Ephrem, B. G. and Gouda, P. “Neural Imaging for Alzheimer’s Prediction using AI: Exploring CNNs and LIME Explanations”, Alzheimer’s & Dementia, vol. 20, núm. S4, p. e088802, dic. 2024, doi: 10.1002/alz.088802. DOI: https://doi.org/10.1002/alz.088802

Kamal, Md. S., Northcote, A., Chowdhury, L., Dey, N., Crespo, R.G. and Herrera-Viedma, E. “Alzheimer’s Patient Analysis Using Image and Gene Expression Data and Explainable-AI to Present Associated Genes”, IEEE Trans. Instrum. Meas., vol. 70, pp. 1–7, 2021, doi: 10.1109/TIM.2021.3107056. DOI: https://doi.org/10.1109/TIM.2021.3107056

Soladoye, A.A., Aderinto, N., Osho, D. and Olawade, D.B. Explainable machine learning models for early Alzheimer’s disease detection using multimodal clinical data”, International Journal of Medical Informatics, vol. 204, p. 106093, ago. 2025, doi: 10.1016/j.ijmedinf.2025.106093. DOI: https://doi.org/10.1016/j.ijmedinf.2025.106093

Loveleen, G., Mohan, B., Shikhar, B. S., Nz, J., Shorfuzzaman, M. and Masud, M. “Explanation-Driven HCI Model to Examine the Mini-Mental State for Alzheimer’s Disease”, ACM Trans. Multimedia Comput. Commun. Appl., vol. 20, núm. 2, pp. 1–16, feb. 2024, doi: 10.1145/3527174. DOI: https://doi.org/10.1145/3527174

Eckhardt, C.M. et al., “Unsupervised machine learning methods and emerging applications in healthcare”, Knee surg. sports traumatol. arthrosc., vol. 31, núm. 2, pp. 376–381, feb. 2023, doi: 10.1007/s00167-022-07233-7. DOI: https://doi.org/10.1007/s00167-022-07233-7

Ripan, R. C., Sarker, I. H., Hasan Furhad, Md., Musfique Anwar, M. and Hoque, M. M. “An Effective Heart Disease Prediction Model Based on Machine Learning Techniques”, en Hybrid Intelligent Systems, A. Abraham, T. Hanne, O. Castillo, N. Gandhi, T. Nogueira Rios, y T.-P. Hong, Eds., en Advances in Intelligent Systems and Computing, vol. 1375. Cham: Springer International Publishing, 2021, pp. 280–288. doi: https://doi.org/10.1007/978-3-030-73050-5_28. DOI: https://doi.org/10.1007/978-3-030-73050-5_28

Nedyalkova, M., Madurga, S. and Simeonov, V. “Combinatorial K-Means Clustering as a Machine Learning Tool Applied to Diabetes Mellitus Type 2”, IJERPH, vol. 18, núm. 4, p. 1919, feb. 2021, doi: 10.3390/ijerph18041919. DOI: https://doi.org/10.3390/ijerph18041919