Please use this identifier to cite or link to this item: http://hdl.handle.net/10609/146168
Title: Modelo de predicción de mortalidad en la Insuficiencia Respiratoria Aguda: análisis de Registros Electrónicos de Salud
Author: González Constán, Eduardo
Tutor: Perez-Alvarez, Nuria  
Others: Ventura, Carles  
Abstract: In recent years, the use of electronic health data (EHR) has become increasingly widespread and has been massively collected in EHR systems. Although these data initially had a purely administrative function, their objectives have been extended to include both medical and research use. An example of an EHR is the MIMIC III dataset, which collects health data from more than 55,000 intensive care inpatients. This study has focused on obtaining a 30-day mortality prediction model for patients admitted with acute respiratory failure using machine learning (ML) techniques and comparing its result with 2 commonly used ICU scores: OASIS and SAPS II. Both ML algorithms and deep neural networks were used to find the best classification model. The best model obtained in the test group corresponds to logistic regression, achieving an AUC of 0.71. This result is slightly higher than that obtained with the SAPS II and OASIS scores (AUC 0.69 and 0.63, respectively). We can conclude, therefore, that the use of ML techniques can improve the usual mortality prediction scores. However, there is still limited performance when applying these techniques to specific cohorts of patients, so more studies of this type are needed in which ML is applied to predict mortality in patients who share similar pathological processes, rather than applying these techniques in a general way.
Keywords: biostatistics
bioinformatics
electronic medical records
Document type: info:eu-repo/semantics/masterThesis
Issue Date: 15-Jun-2022
Publication license: http://creativecommons.org/licenses/by-nc-nd/3.0/es/  
Appears in Collections:Trabajos finales de carrera, trabajos de investigación, etc.

Files in This Item:
File Description SizeFormat 
egonzalezcoTFM0622memoria.pdfMemoria del TFM1,36 MBAdobe PDFThumbnail
View/Open