dc.contributorGiraldo Trujillo, Luis Felipe
dc.contributorCifuentes de la Portilla, Christian Javier
dc.contributorMaldonado Javier Dario
dc.contributorMolina, Maria Alejandra
dc.contributorPrada Lievano, Silvia Alejandra
dc.contributorGarcía Cárdenas, Juan José
dc.contributorArbeláez Escalante, Pablo Andrés
dc.creatorChávez Leyton, Susana Marcela
dc.date.accessioned2023-07-10T19:10:25Z
dc.date.accessioned2023-09-07T01:44:00Z
dc.date.available2023-07-10T19:10:25Z
dc.date.available2023-09-07T01:44:00Z
dc.date.created2023-07-10T19:10:25Z
dc.date.issued2023-06-22
dc.identifierhttp://hdl.handle.net/1992/68272
dc.identifierinstname:Universidad de los Andes
dc.identifierreponame:Repositorio Institucional Séneca
dc.identifierrepourl:https://repositorio.uniandes.edu.co/
dc.identifier.urihttps://repositorioslatinoamericanos.uchile.cl/handle/2250/8728617
dc.description.abstractEl aprendizaje automático ha encontrado una amplia aplicación en el campo de la medicina. Los diferentes métodos están diseñados para complementar la toma de decisiones humanas, proporcionar información y ayudar en tareas médicas. En este caso se explicarán las predicciones sobre dos problemas retadores: predicción de mortalidad en cirugías cardiovasculares y una base de datos para la predicción de los modos de locomoción en escaleras usando sensores IMU.
dc.description.abstractWe address the problem of mortality prediction in cardiovascular surgery patients using preoperative features with the Colombian population in South America. We have a database with a total of 4337 medical histories where 6% corresponds to mortality. An optimized gradient boosting algorithm (XGBoost) was implemented with custom random split datasets and used Monte Carlo cross-validation with them. The results obtained show a F1 of 19% due to the imbalance showing the challenge of the problem and gives the community a first approach to the work in this area.
dc.languageeng
dc.publisherUniversidad de los Andes
dc.publisherMaestría en Ingeniería Electrónica y de Computadores
dc.publisherFacultad de Ingeniería
dc.publisherDepartamento de Ingeniería Eléctrica y Electrónica
dc.relationDANE. Boletín Técnico: Pobreza Multidimensional En Colombia. Bogotá D.C.,Colombia;2019.
dc.relationFerguson TB Jr, Dziuban SW, Edwards FH, etal. The STS. National Database: current changes and challenges for the new millennium. Ann Thorac Surg 2000;69:680-91.
dc.relationNashef SAM, Roques F, Sharples LD, et al. Vallancien G, Emberton M, Harving N, van Moorselaar RJ, Alf-One Study Group. EuroSCORE II dagger. Eur J Cardiothorac Surg. 2012;41:734-745.
dc.relationVictorio C. Carosella and others, The first Latin-American risk stratification system for cardiac surgery: can be used as a graphic pocket-card score, Interactive CardioVascular and Thoracic Surgery, Volume 9, Issue 2, August 2009, Pages 203-208, https://doi.org/10.1510/icvts.2008.199083.
dc.relationMolina, R. S., Molina-Rodríguez, M. A., Rincón, F. M., & Maldonado, J. D. (2022, January). Cardiac Operative Risk in Latin America: A Comparison of Machine Learning Models vs EuroSCORE-II. The Annals of Thoracic Surgery, 113(1), 92-99. https://doi.org/10.1016/j.athoracsur.2021.02.052
dc.relationAutoML: Automatic Machine Learning. H2O. https://docs.h2o.ai/h2o/latest-stable/h2o-docs/automl.html
dc.relationGuogen Shan. Monte Carlo cross-validation for a study with binary outcome and limited sample size. MC Med Inform Decis Mak 2022. DOI: https://doi.org/10.1186/s12911-022-02016-z
dc.relationJ. Camargo, A. Ramanathan, W. Flanagan, and A. Young, "A comprehensive, open-source dataset of lower limb biomechanics in multiple conditions of stairs, ramps, and level-ground ambulation and transitions," Journal of Biomechanics, vol. 119, p. 110320, 2021.
dc.relationV. Losing and M. Hasenjäger, "A Multi-Modal Gait Database of Natural Everyday-walk in an urban environment," Scientific Data, vol. 9, no. 1, 2022. doi:10.1038/s41597-022-01580-3.
dc.relationK. Bhakta, J. Camargo, L. Donovan, K. Herrin, and A. Young, "Machine learning model comparisons of user independent & dependent intent recognition systems for powered prostheses," IEEE Robotics and Automation Letters, vol. 5, no. 4, pp. 5393-5400, 2020. doi:10.1109/lra.2020.3007480
dc.relationKhandelwal, S. & Wickström, N. Evaluation of the performance of accelerometer-based gait event detection algorithms in different real-world scenarios using the MAREA gait database. Gait & Posture 51, 84-90 (2017).
dc.relationGuillén Aguirre, C. (2021). Interpretable ambulation mode classification method for human-wearable robot interaction. Universidad de los Andes.
dc.relation"Movella Dot," Movella.com, https://www.movella.com/products/wearables/movella-dot (accessed Jun. 20, 2023).
dc.relation"Welcome to OpenSim," Welcome to OpenSim - OpenSim Documentation - Global Site, https://simtk-confluence.stanford.edu:8443/display/OpenSim/Welcome+to+OpenSim (accessed Jun. 20, 2023).
dc.relation"XGBoost," XGBoost - H2O 3.40.0.4 documentation, https://docs.h2o.ai/h2o/latest-stable/h2o-docs/data-science/xgboost.html (accessed Jun. 20, 2023).
dc.rightsAttribution-NonCommercial-NoDerivatives 4.0 Internacional
dc.rightshttps://repositorio.uniandes.edu.co/static/pdf/aceptacion_uso_es.pdf
dc.rightsinfo:eu-repo/semantics/openAccess
dc.rightshttp://purl.org/coar/access_right/c_abf2
dc.titleMachine learning in medicine: mortality prediction on cardiac surgeries in Colombia
dc.typeTrabajo de grado - Maestría


Este ítem pertenece a la siguiente institución