Early Detection and Prevention of Pneumonia in ICU Patients: Developing a Decision Support Tool Using Clinical Information System Data and Real-Time Infection Mappin

Entity: Universitat Rovira i Virgili (URV)

Confidenciality: Si

Education area(s): Ciència de Dades Biomèdiques

APS: No

Title in different languages: Detecció precoç i prevenció de la pneumònia en pacients de la UCI: Desenvolupament d'una eina de suport a la decisió mitjançant dades del sistema d'informació clínica i mappin d'infeccions en temps real.

Abstract: This thesis develops a decision support tool for early detection and prevention of Ventilator-Associated Pneumonia (VAP) in ICU patients using patient-specific data like demographics, lab results, drug histories, comorbidities, and ventilation metrics. Variables related to comorbidities, demographics, and patient status 24, 48, and 72 hours before the event were analyzed. The event day was when pneumonia developed; for those without pneumonia, the last ventilation day was the reference. A real-time infection dashboard was integrated to visualize patient trends and aid clinical decisions. Model predictions will remain blind until validated prospectively.

Subject: Pneumònia

Academic year: 2023-2024

Language: en

Work's public defense date: 2024-06-20

Subject areas: Health sciences

Student: Berrueta Llona, Julen

Work's codirector: Gómez Alvarez, Josep

Department: Enginyeria Informàtica i Matemàtiques

Creation date in repository: 2025-03-13

Keywords: ventilator-associated pneumonia (VAP); machine learning (ML); intensive care unit (ICU); dashboard

Title in original language: Early Detection and Prevention of Pneumonia in ICU Patients: Developing a Decision Support Tool Using Clinical Information System Data and Real-Time Infection Mappin

Access Rights: info:eu-repo/semantics/openAccess

Project director: Bodí Saera, Maria Amparo