Improvement of bowel content segmentation in abdominal CT images using the MONAI project

Medical imaging plays a pivotal role in the diagnosis, treatment planning, and monitoring of various diseases. In particular, abdominal computed tomography (CT) imaging provides invaluable insight into gastrointestinal conditions. Despite its importance, segmenting bowel contents from abdominal CT s...

Descripción completa

Detalles Bibliográficos
Autor: Ramirez Mijarra, Ivan
Tipo de recurso: tesis de maestría
Fecha de publicación:2025
País:España
Institución:Universitat Politècnica de Catalunya (UPC)
Repositorio:UPCommons. Portal del coneixement obert de la UPC
Idioma:inglés
OAI Identifier:oai:upcommons.upc.edu:2117/449182
Acceso en línea:https://hdl.handle.net/2117/449182
Access Level:acceso abierto
Palabra clave:Imaging systems in medicine
Neural networks (Computer science)
Abdomen--Ultrasonic imaging
MONAI
nnUNET
Imatges mèdiques
CT scans
AI
Reds neuronals
Medical image
Neural network
Imatgeria mèdica
Xarxes neuronals (Informàtica)
Ecografia abdominal
Àrees temàtiques de la UPC::Informàtica::Intel·ligència artificial::Aprenentatge automàtic
Descripción
Sumario:Medical imaging plays a pivotal role in the diagnosis, treatment planning, and monitoring of various diseases. In particular, abdominal computed tomography (CT) imaging provides invaluable insight into gastrointestinal conditions. Despite its importance, segmenting bowel contents from abdominal CT scans remains a challenging task due to the anatomical variability and complexity of the human abdomen. This thesis investigates the application of the MONAI (Medical Open Network for AI) frame- work to enhance bowel content segmentation. Accurate segmentation is critical for clinical interpretation and decision-making, and MONAI offers a robust, research-oriented foundation for deep learning workflows in medical imaging. An additional goal of this project is to provide thorough documentation of the methods and processes involved, ensuring reproducibility and facilitating future developments as the project continues to evolve.