Publicación:
Modelos predictivos en la clasificación de nacidos vivos y mortinatos: un estudio comparativo entre técnicas de machine learning y regresión logística en función de variables sociodemográficas y clínicas

dc.contributor.authorAraujo Zarate, Pedrospa
dc.contributor.authorMartínez Lobo, Dannyspa
dc.contributor.authorContreras Chávez, Johnspa
dc.date.accessioned2024-10-15T00:00:00Z
dc.date.accessioned2025-05-23T10:00:31Z
dc.date.available2024-10-15T00:00:00Z
dc.date.available2025-05-23T10:00:31Z
dc.date.issued2024-10-15
dc.description.abstractIntroducción: La mortalidad fetal continúa siendo un problema de salud pública en Colombia, que afecta significativamente el bienestar familiar y social. Es fundamental identificar los factores asociados y predecir el riesgo para implementar intervenciones efectivas. Objetivo: Desarrollar un modelo estadístico para clasificar a las gestantes en riesgo de mortalidad fetal en Colombia durante el año 2022. Métodos: Se realizó un estudio de casos y controles utilizando datos de nacidos vivos y defunciones fetales reportados por el Departamento Administrativo Nacional de Estadística. Se aplicaron técnicas de imputación de datos faltantes y balanceo de clases mediante el método SMOTE. Se evaluaron cuatro modelos de clasificación: regresión logística, K-Nearest Neighbors (KNN), árbol de decisión y máquina de soporte vectorial. El rendimiento de los modelos se comparó utilizando métricas de exactitud, sensibilidad, especificidad, puntaje F1 y precisión. Resultados: El conjunto de datos final incluyó 566.806 registros, con 562.828 nacidos vivos y 3.978 muertes fetales. El modelo KNN presentó el mejor rendimiento, con una exactitud de 0,988, sensibilidad de 0,989, especificidad de 0,986 y puntaje F1 de 0,988. Los factores asociados significativamente con la probabilidad de nacer vivo incluyeron el número de hijos, el sexo, el área de residencia, el régimen de afiliación, las semanas de gestación, el peso al nacer, la edad y el nivel educativo de la madre. Conclusión: El modelo KNN demostró ser efectivo en la predicción del riesgo de mortalidad fetal. Los resultados resaltan la importancia de factores socioeconómicos y clínicos en la supervivencia neonatal, sugiriendo la necesidad de intervenciones focalizadas para reducir las muertes fetales en Colombia.spa
dc.description.abstractIntroduction: Fetal mortality continues to be a public health problem in Colombia, significantly affecting family and social well-being. It is essential to identify associated factors and predict risk in order to implement effective interventions. Objective: Develop a statistical model to classify pregnant women at risk of fetal mortality in Colombia during 2022. Methods: A case-control study was conducted using data on live births and fetal deaths reported by the National Administrative Department of Statistics. Data imputation and class balancing techniques were applied using the smote method. Four classification models were evaluated: logistic regression, K-Nearest Neighbors (KNN), decision tree, and support vector machine. Model performance was compared using accuracy, sensitivity, specificity, F1 score, and precision metrics. Results: The final dataset included 566,806 records, with 562,828 live births and 3,978 fetal deaths. The KNN model showed the best performance, with an accuracy of 0.988, sensitivity of 0.989, specificity of 0.986, and F1 score of 0.988. Factors significantly associated with the probability of live birth included the number of children, sex, area of residence, affiliation regime, gestational weeks, birth weight, and mother's age and educational level. Conclusions: The KNN model proved effective in predicting the risk of fetal mortality. The results highlight the importance of socioeconomic and clinical factors in neonatal survival, suggesting the need for targeted interventions to reduce fetal deaths in Colombia.eng
dc.format.mimetypeapplication/pdfspa
dc.identifier.doi10.32997/rcb-2024-4940
dc.identifier.eissn2389-7252
dc.identifier.issn2215-7840
dc.identifier.urihttps://hdl.handle.net/11227/19530
dc.identifier.urlhttps://doi.org/10.32997/rcb-2024-4940
dc.language.isospaspa
dc.publisherUniversidad de Cartagenaspa
dc.relation.bitstreamhttps://revistas.unicartagena.edu.co/index.php/cbiomedicas/article/download/4940/4025
dc.relation.citationendpage189
dc.relation.citationissue4spa
dc.relation.citationstartpage175
dc.relation.citationvolume13spa
dc.relation.ispartofjournalRevista Ciencias Biomédicasspa
dc.relation.referencesGoldenberg RL, McClure EM, Belizán JM. Commentary: Reducing the world’s stillbirths. Vol. 9, BMC Pregnancy and Childbirth. 2009.spa
dc.relation.referencesGarcía MV, Cuberos MA, Vera MÁ, Vivas García M, Cuberos MA, Vera MÁ. Mortalidad fetal en madres adolescentes y adultas, un problema de salud pública TT - Fetal mortality in adolescent and adult mothers: a public health problem. Salud(i)Ciencia. 2019;23 (5).spa
dc.relation.referencesLawn JE, Blencowe H, Pattinson R, Cousens S, Kumar R, Ibiebele I, et al. Stillbirths: Where? When? Why? How to make the data count? Vol. 377, The Lancet. 2011.spa
dc.relation.referencesDepartamento Administrativo Nacional de Estadisticas. Boletín técnico Estadísticas Vitales (EEVV) [Internet]. Bogotá D.C.; 2024 Mar [cited 2024 Sep 20]. Available from: https://www.dane.gov.co/files/operaciones/EEVV/bol-EEVV-Defunciones-IVtrim2023.pdfspa
dc.relation.referencesWinkler AM, Ridgway GR, Webster MA, Smith SM, Nichols TE. Permutation inference for the general linear model. Neuroimage. 2014 May 15;92:381–97.spa
dc.relation.referencesCasella G, Fienberg S, Olkin I. Springer Texts in Statistics [Internet]. Available from: www.springer.com/series/417spa
dc.relation.referencesChartier S, Faulkner A. General Linear Models: An Integrated Approach to Statistics. Tutor Quant Methods Psychol. 2008 Sep 1;4(2):65–78.spa
dc.relation.referencesHelwig NE. Robust nonparametric tests of general linear model coefficients: A comparison of permutation methods and test statistics. Neuroimage. 2019 Nov 1;201.spa
dc.relation.referencesAizerman MA, Braverman EM, Rozonoer LI. Theoretical Foundatiosn of the Potential Function Method in Pattern Recognition Learning. Telemekhanika. 1964;25(6):971–936.spa
dc.relation.referencesHuang S, Nianguang CAI, Penzuti Pacheco P, Narandes S, Wang Y, Wayne XU. Applications of support vector machine (SVM) learning in cancer genomics. Vol. 15, Cancer Genomics and Proteomics. International Institute of Anticancer Research; 2018. p. 41–51.spa
dc.relation.referencesNoble WS. What is a support vector machine? [Internet]. Vol. 24, NATURE BIOTECHNOLOGY. 2006. Available from: http://www.nature.com/naturebiotechnologyspa
dc.relation.referencesHu LY, Huang MW, Ke SW, Tsai CF. The distance function effect on k-nearest neighbor classification for medical datasets. Springerplus. 2016;5(1).spa
dc.relation.referencesEhsani R, Drabløs F. Robust Distance Measures for kNN Classification of Cancer Data. Cancer Inform. 2020;19.spa
dc.relation.referencesHu J, Szymczak S. A review on longitudinal data analysis with random forest. Vol. 24, Briefings in Bioinformatics. 2023.spa
dc.relation.referencesBreiman L. Random Forests. Vol. 45. 2001.spa
dc.relation.referencesCutler DR, Edwards TC, Beard KH, Cutler A, Hess KT, Gibson J, et al. Random forests for classification in ecology. Ecology. 2007;88 (11).spa
dc.relation.referencesGraham SE, Clarke SL, Wu KHH, Kanoni S, Zajac GJM, Ramdas S, et al. The power of genetic diversity in genome-wide association studies of lipids. Nature. 2021;600 (7890).spa
dc.relation.referencesPowers DMW. Evaluation: from precision, recall and F-measure to ROC, informedness, markedness and correlation. International Journal of Machine Learning Technology. 2011;2 (1):37–63.spa
dc.relation.referencesFawcett T. An introduction to ROC analysis. Pattern Recognit Lett. 2006 Jun;27 (8):861–74.spa
dc.relation.referencesHe H, Garcia EA. Learning from Imbalanced Data. IEEE Trans Knowl Data Eng. 2009;21 (9):1263–84.spa
dc.relation.referencesSokolova M, Lapalme G. A systematic analysis of performance measures for classification tasks. Inf Process Manag. 2009 Jul;45 (4):427–37.spa
dc.relation.referencesVan Rijsbergen CJ. Information retrievial. 1979.spa
dc.relation.referencesManning C, Raghavan P, Schütze H. An introduction to information retrieval. Cambridge; 2009.spa
dc.relation.referencesJapkowicz N, Stephen S. The class imbalance problem: A systematic study. Intell Data Anal. 2002 Oct;6(5):429–49.spa
dc.relation.referencesDepartamento Administrativo Nacional de Estadistica DANE. Boletín Técnico Estadísticas Vitales (EEVV) Nacidos Vivos. Bogotá, DC; 2023 Mar.spa
dc.relation.referencesDepartamento Administrativo Nacional de Estadistica. Boletín técnico Estadísticas Vitales (EEVV) Defunciones. Bogotá, DC; 2023 Mar.spa
dc.relation.referencesMcClure EM, Saleem S, Goudar SS, Garces A, Whitworth R, Esamai F, et al. Stillbirth 2010–2018: a prospective, population-based, multi-country study from the Global Network. Reprod Health. 2020 Nov 30;17 (S2):146.spa
dc.relation.referencesMgaya AH, Massawe SN, Kidanto HL, Mgaya HN. Grand multiparity: is it still a risk in pregnancy? BMC Pregnancy Childbirth. 2013 Dec 23;13(1):241.spa
dc.relation.referencesGalvis-Aponte L, Rico J. Desigualdades regionales en la salud en Colombia. 2023 Jun.spa
dc.relation.referencesMinisterio S y PS. Informe Nacional de Calidad de la Atención en Salud 2015 [Internet]. Bogotá, DC; 2015 Dec [cited 2024 Nov 10]. Available from: https://www.minsalud.gov.co/sites/rid/Lists/BibliotecaDigital/RIDE/DE/DIJ/informe-nal-calidad-atencion-salud-2015.pdfspa
dc.relation.referencesCáceres-Manrique F de M, Molina-Marín G. Inequidades sociales en atención materna. Rev Colomb Obstet Ginecol. 2010 Sep 30;61 (3):231–8.spa
dc.relation.referencesMerlano Porras C, Gorbanev I. Health system in Colombia: A systematic review of literature. Revista Gerencia y Políticas de Salud. 2013 Jun 1;12:74–86.spa
dc.relation.referencesKramer MS. Determinants of low birth weight: methodological assessment and meta-analysis. Bull World Health Organ. 1987;65 (5):663–737.spa
dc.relation.referencesBlencowe H, Cousens S, Chou D, Oestergaard M, Say L, Moller AB, et al. Born Too Soon: The global epidemiology of 15 million preterm births. Reprod Health. 2013 Nov 15;10 (S1):S2.spa
dc.relation.referencesLiu L, Oza S, Hogan D, Chu Y, Perin J, Zhu J, et al. Global, regional, and national causes of under-5 mortality in 2000–15: an updated systematic analysis with implications for the Sustainable Development Goals. The Lancet. 2016 Dec;388 (10063):3027–35.spa
dc.relation.referencesGanchimeg T, Ota E, Morisaki N, Laopaiboon M, Lumbiganon P, Zhang J, et al. Pregnancy and childbirth outcomes among adolescent mothers: a World Health Organization multicountry study. BJOG. 2014 Mar 18;121 (s1):40–8.spa
dc.relation.referencesJolly M, Sebire N, Harris J, Robinson S, Regan L. The risks associated with pregnancy in women aged 35 years or older. Human Reproduction. 2000 Nov 1;15 (11):2433–7.spa
dc.relation.referencesD’Ascoli PT, Alexander GR, Petersen DJ, Kogan MD. Parental factors influencing patterns of prenatal care utilization. J Perinatol. 1997;17(4):283–7.spa
dc.relation.referencesVictora CG, Aquino EM, do Carmo Leal M, Monteiro CA, Barros FC, Szwarcwald CL. Maternal and child health in Brazil: progress and challenges. The Lancet. 2011 May;377 (9780):1863–76.spa
dc.relation.referencesGakidou E, Cowling K, Lozano R, Murray CJ. Increased educational attainment and its effect on child mortality in 175 countries between 1970 and 2009: a systematic analysis. The Lancet. 2010 Sep;376 (9745):959–74.spa
dc.relation.referencesFuchs R, Pamuk E, Lutz W. Education or wealth: Which matters more for reducing child mortality in developing countries? Vienna Yearb Popul Res. 2010 Nov;8:175–99.spa
dc.relation.referencesGüneş PM. The role of maternal education in child health: Evidence from a compulsory schooling law. Econ Educ Rev. 2015 Aug;47:1–16.spa
dc.relation.referencesCover T, Hart P. Nearest neighbor pattern classification. IEEE Trans Inf Theory. 1967 Jan;13 (1):21–7.spa
dc.rightsPedro Araujo Zarate, Danny Martínez, John Jairo Contreras Chávez - 2024spa
dc.rights.accessrightsinfo:eu-repo/semantics/openAccessspa
dc.rights.coarhttp://purl.org/coar/access_right/c_abf2spa
dc.rights.creativecommonsEsta obra está bajo una licencia internacional Creative Commons Atribución-NoComercial-SinDerivadas 4.0.spa
dc.rights.urihttps://creativecommons.org/licenses/by-nc-nd/4.0spa
dc.sourcehttps://revistas.unicartagena.edu.co/index.php/cbiomedicas/article/view/4940spa
dc.subjectMortalidad fetalspa
dc.subjectModelos estadísticosspa
dc.subjectSalud públicaspa
dc.subjectMachine learningspa
dc.subjectFetal Deatheng
dc.subjectStatistical Modelseng
dc.subjectPublic Healtheng
dc.subjectMachine Learningeng
dc.titleModelos predictivos en la clasificación de nacidos vivos y mortinatos: un estudio comparativo entre técnicas de machine learning y regresión logística en función de variables sociodemográficas y clínicasspa
dc.title.translatedPredictive models in the classification of live births and stillbirths: a comparative study between machine learning and logistic regression techniques as a function of sociodemographic and clinical variableseng
dc.typeArtículo de revistaspa
dc.type.coarhttp://purl.org/coar/resource_type/c_6501spa
dc.type.coarhttp://purl.org/coar/resource_type/c_2df8fbb1spa
dc.type.coarversionhttp://purl.org/coar/version/c_970fb48d4fbd8a85spa
dc.type.contentTextspa
dc.type.driverinfo:eu-repo/semantics/articlespa
dc.type.localJournal articleeng
dc.type.redcolhttp://purl.org/redcol/resource_type/ARTspa
dc.type.versioninfo:eu-repo/semantics/publishedVersionspa
dspace.entity.typePublicationspa

Archivos

Datos de Contacto

Imagen Escudo Universidad de Cartagena

 

 

 

Línea de Atención

Línea Anticorrupción

Síguenos en: