ANÁLISIS DEL DESEMPEÑO DE UN PERCEPTRÓN MULTICAPA PARA EL DIAGNÓSTICO DE ALZHEIMER CONSIDERANDO LA COMBINACIÓN DE TÉCNICAS DE SELECCIÓN DE VARIABLES Y MÉTODOS DE ESCALADO (PERFORMANCE ANALYSIS OF A MULTILAYER PERCEPTRON FOR ALZHEIMER’S DIAGNOSIS CONSIDERING THE COMBINATION OF FEATURE SELECTION TECHNIQUES AND SCALING METHODS)

Alejandra Guadalupe Bravo García, Maricela Quintana López, Victor Manuel Landassuri Moreno, Saul Lazcano Salas, Asdrúbal López Chau

Resumen


Resumen
El diagnóstico del Alzheimer representa un reto clínico, debido a su carácter multifactorial y la necesidad de identificar patrones sutiles en datos clínicos. Este artículo evalúa el desempeño de un Perceptrón Multicapa (MLP) para el diagnóstico de Alzheimer, comparando cuatro técnicas de selección de características: Chi-cuadrado (Chi²), Información Mutua (MI), Bosque Aleatorio (RF) y Regresión Logística L1; combinadas con cuatro métodos de escalado: Min-Max, StandardScaler, RobustScaler y normalización en dos pasos, sobre datos clínicos. Los resultados muestran que Chi² con RobustScaler alcanzó el mejor desempeño global (exactitud = 0.9069; AUC = 0.9361), y que Chi² y RF fueron los métodos de selección más estables entre métricas. La comparación evidencia que la elección del escalado influye de manera sustantiva en el rendimiento del clasificador e interactúa con la selección de características. Se concluye que flujos de preprocesamiento bien diseñados, integrando selección y escalado, potencian los modelos MLP para la detección de Alzheimer.
Palabras Clave: Escalado de datos, Perceptrón Multicapa (MLP), selección de variables.

Abstract
The diagnosis of Alzheimer’s disease represents a clinical challenge due to its multifactorial nature and the need to identify subtle patterns in clinical data. This study evaluates the performance of a Multilayer Perceptron (MLP) for Alzheimer’s diagnosis, comparing four feature selection techniques: Chi-Squared (Chi²), Mutual Information (MI), Random Forest (RF), and L1 Logistic Regression; combined with four scaling methods: Min-Max, StandardScaler, RobustScaler, and Two-Step Normalization, applied to clinical data. The results show that the Chi² with RobustScaler combination achieved the best overall performance (accuracy = 0.9069; AUC = 0.9361), and that Chi² and RF were the most stable feature selection methods across metrics. The comparison highlights that the choice of scaling significantly affects classifier performance and interacts with feature selection. It is concluded that well-designed preprocessing pipelines integrating selection and scaling enhance the effectiveness of MLP models for Alzheimer’s detection.
Keywords: Data scaling, feature selection, Multilayer Perceptron (MLP).

Texto completo:

308-325 PDF

Referencias


Aguayo, G. A., Zhang, L., Vaillant, M., Ngari, M., Perquin, M., Moran, V., Huiart, L., Krüger, R., Azuaje, F., Ferdynus, C., Fagherazzi, G., (2023). Machine learning for predicting neurodegenerative diseases in the general older population: A cohort study. BMC Medical Research Methodology, Vol. 23, No. 1, Article 1. https://doi.org/10.1186/s12874-023-01837-4.

Cabanillas, M., Zapata, J., (2025). Evaluation of machine learning models for the prediction of Alzheimer’s: In search of the best performance. Brain, Behavior, & Immunity Health, Vol. 44, 100957. https://doi.org/10.1016/j.bbih.2025.100957.

Davuluri, R., (2020). A Survey of Different Machine Learning Models for Alzheimer Disease Prediction. International Journal of Emerging Trends in Engineering Research, Vol. 8, No. 7, Article 7. https://doi.org/10.30534/ijeter/2020/73872020.

El-Sappagh, S., Alonso, J. M., Islam, S. M. R., Sultan, A. M., Kwak, K. S., (2021). A multilayer multimodal detection and prediction model based on explainable artificial intelligence for Alzheimer’s disease. Scientific Reports, Vol. 11, No. 1, 2660. https://doi.org/10.1038/s41598-021-82098-3.

Gowri, G., Lun, X. K., Klein, A. M., Yin, P. Approximating mutual information of high-dimensional variables using learned representations, 2024.

Pudjihartono, N., Fadason, T., Kempa, A. W., O’Sullivan, J. M., (2022). A Review of Feature Selection Methods for Machine Learning-Based Disease Risk Prediction. Frontiers in Bioinformatics, Vol. 2. https://doi.org/10.3389/fbinf.2022.927312.

Rahman, M. M., Usman, O. L., Muniyandi, R. C., Sahran, S., Mohamed, S., Razak, R. A., (2020). A Review of Machine Learning Methods of Feature Selection and Classification for Autism Spectrum Disorder. Brain Sciences, Vol. 10, No. 12, Article 12. https://doi.org/10.3390/brainsci10120949.

Sharma, V., (2022). A Study on Data Scaling Methods for Machine Learning. International Journal for Global Academic & Scientific Research, Vol. 1, No. 1. https://doi.org/10.55938/ijgasr.v1i1.4.

Silpa, N., Swain, S. K., Rao V. V. R., (2024). Revolutionizing feature engineering for robust ensemble machine learning by hybridizing MRMR insight and CHI2 independence. Proceedings on Engineering Sciences, Vol. 6, No. 3, pp. 1337–1348. https://doi.org/10.24874/PES.SI.24.03.017.

Speiser, J. L., (2021). A random forest method with feature selection for developing medical prediction models with clustered and longitudinal data. Journal of Biomedical Informatics, Vol. 117, 103763. https://doi.org/10.1016/j.jbi.2021.103763.

Sree, K. D., Bindu, C. S., (2018). Data analytics: Why data normalization. International Journal of Engineering and Technology (UAE), Vol. 7, No. 4.6 (Special Issue 6). https://doi.org/10.14419/ijet.v7i4.6.20464.

Venkatesh, B., Anuradha, J., (2019). A Review of Feature Selection and Its Methods. Cybernetics and Information Technologies, Vol. 19, No. 1, pp. 3–26. https://doi.org/10.2478/cait-2019-0001.

Zhou, H., Wang, X., Zhang, Y., (2024). Feature selection based on weighted conditional mutual information. Applied Computing and Informatics, Vol. 20, No. 1–2, Article 1–2, https://doi.org/10.1016/j.aci.2019.12.003.






URL de la licencia: https://creativecommons.org/licenses/by/3.0/deed.es

Barra de separación

Licencia Creative Commons    Pistas Educativas está bajo la Licencia Creative Commons Atribución 3.0 No portada.    

TECNOLÓGICO NACIONAL DE MÉXICO / INSTITUTO TECNOLÓGICO DE CELAYA

Antonio García Cubas Pte #600 esq. Av. Tecnológico, Celaya, Gto. México

Tel. 461 61 17575 Ext 5450 y 5146

pistaseducativas@itcelaya.edu.mx

http://pistaseducativas.celaya.tecnm.mx/index.php/pistas