Reconstrucción de imágenes pulmonares a partir de mediciones de tomografía por impedancia eléctrica torácica.
Loading...
Date
2024
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Universidad de Concepción
Abstract
Las enfermedades respiratorias suponen una carga importante para los sistemas de salud mundiales y afectan a más de mil millones de personas con enfermedades crónicas como el asma y la EPOC (Enfermedades pulmonares obstructivas crónicas). Esto subraya la urgencia de la lucha contra las enfermedades pulmonares.
La tomografía por impedancia eléctrica es una técnica médica establecida que utiliza principios fí- sicos para obtener imágenes de la impedancia de los tejidos. Consiste en colocar electrodos en zonas específicas del cuerpo y utilizar pequeñas corrientes eléctricas para crear imágenes de la zona de interés. Esta tecnología se caracteriza por su asequibilidad y facilidad de implementación, y puede utilizarse para una variedad de aplicaciones médicas y clínicas. Su objetivo principal es monitorizar la función pulmonar en tiempo real y proporcionar información continua.
El presente informe describe el desarrollo y resultados obtenidos de procesos de validación, reconstrucción y análisis de imágenes pulmonares a partir de mediciones de tomografía por Impedancia Eléctrica torácica, parte de un proyecto más amplio de fabricación de un dispositivo para apoyo diagnóstico. Se detalla el desarrollo del sistema, destacando la validación de imágenes a partir de un fantoma de resistencias, conversión y manejo de datos. Con estos datos se reconstruyen imágenes a partir del software EIDORS, el cual a través de manejo de parámetros como solver, hyperparameter y algoritmos de optimización, reconstruye imágenes en un modelo pulmonar utilizando FEM (Método de Elementos Finitos). Por otra parte se muestra la implementación y diseño de una aplicación para la reconstrucción y análisis de imágenes pulmonares realizada en MATLAB App Designer, con bajos tiempos de procesamiento.
Los resultados indican que las imágenes son las esperadas, correspondientes a las mediciones de impedancia. Es posible observar cómo las imágenes varían a lo largo del tiempo, representando la respi- ración de 20 voluntarios. El análisis de la variación de un píxel en el tiempo muestra cambios de impedancia, correlacionados con el flujo de aire correspondiente a la respiración pulmonar en un 87.31 % para uno de los voluntarios. Este proceso de reconstrucción del total de datos fue realizado en 90.9 segundos en promedio, siendo 3 veces menor que el tiempo de adquisición de los datos en los voluntarios.
En conclusión, los resultados de esta memoria muestran la efectividad del sistema EIT implementado, la posibilidad de contar con una aplicación para la visualización y análisis, y una velocidad de reconstrucción adecuada, que permitiría su uso en un dispositivo Point of Care.
Respiratory diseases place a significant burden on global health systems and affect more than one billion people with chronic diseases such as asthma and COPD (Chronic Obstructive Pulmonary Disease). This underscores the urgency of the fight against lung diseases. Electrical impedance tomography is an established medical technique that uses physical principles to image tissue impedance. It involves placing electrodes on specific areas of the body and using small elec- trical currents to create images of the area of interest. This technology is characterized by its affordability and ease of implementation, and can be used for a variety of medical and clinical applications. Its main objective is to monitor lung function in real time and provide continuous information. This report describes the development and results obtained from the validation, reconstruction and analysis of lung images from thoracic Electrical Impedance Tomography measurements, part of a larger project to manufacture a device for diagnostic support. The development of the system is detailed, high- lighting the validation of images from a resistor phantom, data conversion and management. With these data, images are reconstructed from the EIDORS software, which through the management of parameters such as solver, hyperparameter and optimization algorithms, reconstructs images in a lung model using FEM (Finite Element Method). On the other hand, the implementation and design of an application for the reconstruction and analysis of lung images made in MATLAB App Designer, with low processing times, is shown. The results indicate that the images are as expected, corresponding to the impedance measurements. It is possible to observe how the images vary over time, representing the breathing of 20 volunteers. The analysis of the variation of a pixel over time shows changes in impedance, correlated with the airflow corresponding to pulmonary respiration by 87.31 % for one of the volunteers. This process of recons- truction of the total data was performed in 90.9 seconds on average, being 3 times shorter than the data acquisition time in the volunteers. In conclusion, the results of this report show the effectiveness of the implemented EIT system, the possibility of having an application for visualization and analysis, and an adequate reconstruction speed, which would allow its use in a Point of Care device.
Respiratory diseases place a significant burden on global health systems and affect more than one billion people with chronic diseases such as asthma and COPD (Chronic Obstructive Pulmonary Disease). This underscores the urgency of the fight against lung diseases. Electrical impedance tomography is an established medical technique that uses physical principles to image tissue impedance. It involves placing electrodes on specific areas of the body and using small elec- trical currents to create images of the area of interest. This technology is characterized by its affordability and ease of implementation, and can be used for a variety of medical and clinical applications. Its main objective is to monitor lung function in real time and provide continuous information. This report describes the development and results obtained from the validation, reconstruction and analysis of lung images from thoracic Electrical Impedance Tomography measurements, part of a larger project to manufacture a device for diagnostic support. The development of the system is detailed, high- lighting the validation of images from a resistor phantom, data conversion and management. With these data, images are reconstructed from the EIDORS software, which through the management of parameters such as solver, hyperparameter and optimization algorithms, reconstructs images in a lung model using FEM (Finite Element Method). On the other hand, the implementation and design of an application for the reconstruction and analysis of lung images made in MATLAB App Designer, with low processing times, is shown. The results indicate that the images are as expected, corresponding to the impedance measurements. It is possible to observe how the images vary over time, representing the breathing of 20 volunteers. The analysis of the variation of a pixel over time shows changes in impedance, correlated with the airflow corresponding to pulmonary respiration by 87.31 % for one of the volunteers. This process of recons- truction of the total data was performed in 90.9 seconds on average, being 3 times shorter than the data acquisition time in the volunteers. In conclusion, the results of this report show the effectiveness of the implemented EIT system, the possibility of having an application for visualization and analysis, and an adequate reconstruction speed, which would allow its use in a Point of Care device.
Description
Tesis presentada para optar al título de Ingeniero/a Civil Biomédico.
Keywords
Tomografía, Análisis de imagen, Enfermedades pulmonares