Aplicabilidad del procesamiento del lenguaje natural y del aprendizaje automático en el desarrollo y validación de modelos predictivos de riesgo de sangrado y recurrencia de enfermedad tromboembólica venosa en pacientes con cáncer activo que reciben terapia anticoagulante
Loading...
Download
Official URL
Full text at PDC
Publication date
2024
Defense date
19/06/2024
Advisors (or tutors)
Editors
Journal Title
Journal ISSN
Volume Title
Publisher
Universidad Complutense de Madrid
Citation
Abstract
La enfermedad tromboembólica venosa (ETV) constituye una complicación frecuente de los pacientes con cáncer. Tras ese primer evento, el riesgo de recurrencia de ETV y de sangrado mayor (SM) está también aumentado en los pacientes oncológicos que reciben terapia anticoagulante. En la práctica clínica sería muy útil poder conocer qué pacientes van a presentar recurrencia de ETV y/o SM para poder implementar una terapia anticoagulante más dirigida y personalizada. Hasta el momento de empezar a escribir esta tesis sólo se había propuesto el modelo Ottawa, un modelo predictivo de recurrencia de ETV específico para pacientes con cáncer que reciben terapia anticoagulante, pero no se emplea en la práctica clínica y ninguna guía clínica recomienda su uso, y ningún modelo predictivo de sangrado. Durante la escritura de ésta se han propuesto dos modelos predictivos de sangrado específicos de pacientes con cáncer, el modelo predictivo CAT-BLEED y el B-CAT, pero no existen publicaciones hasta la fecha de que ninguno se haya validado externamente en otra población...
Venous thromboembolism (VTE) is a common complication in cancer patients. After a first event, cancer patients on anticoagulant therapy also have an increased risk of recurrent VTE and major bleeding (MB). In clinical practice, it would be very useful to know which patients are likely to have a VTE recurrence and/or MB in order to implement targeted and personalized anticoagulant therapy. At the time of writing this thesis, only the Ottawa score has been proposed as a cancer-specific risk assessment model (RAM) for VTE recurrence during anticoagulation. However, it is not used in clinical practice and the main clinical guidelines don’t recommend its use. During the development of this work, the first two cancer-specific RAMs of bleeding have been published (CAT-BLEED and B-CAT); neither of these has been externally validated to date. The development of Artificial Intelligence (AI), with Machine Learning (ML) and Natural Language Processing (NLP), as well as the large amount of Real World Data (RWD) in the healthcare system, may provide new RAMs with better predictive capabilities...
Venous thromboembolism (VTE) is a common complication in cancer patients. After a first event, cancer patients on anticoagulant therapy also have an increased risk of recurrent VTE and major bleeding (MB). In clinical practice, it would be very useful to know which patients are likely to have a VTE recurrence and/or MB in order to implement targeted and personalized anticoagulant therapy. At the time of writing this thesis, only the Ottawa score has been proposed as a cancer-specific risk assessment model (RAM) for VTE recurrence during anticoagulation. However, it is not used in clinical practice and the main clinical guidelines don’t recommend its use. During the development of this work, the first two cancer-specific RAMs of bleeding have been published (CAT-BLEED and B-CAT); neither of these has been externally validated to date. The development of Artificial Intelligence (AI), with Machine Learning (ML) and Natural Language Processing (NLP), as well as the large amount of Real World Data (RWD) in the healthcare system, may provide new RAMs with better predictive capabilities...
Description
Tesis inédita de la Universidad Complutense de Madrid, Facultad de Medicina, leída el 19-06-2024