Monitoreo de la Concentración de la Glucosa Plasmática en Pacientes en Cuidados Intensivos usando Mediciones de Glucosa Intersticial

The glucose homeostasis is responsible for regulating the blood glucose concentration around 100 mg / dl. As soon as this mechanism is broken due to the inability of the pancreas to produce insulin, the blood glucose levels increase and patients are diagnosed with Type 1 Diabetes Mellitus (T1DM). Re...

Full description

Autores:
Aguirre-Zapata, Estefanía
García-Tirado, José Fernando
Tipo de recurso:
Fecha de publicación:
2016
Institución:
Universidad de San Buenaventura
Repositorio:
Repositorio USB
Idioma:
spa
OAI Identifier:
oai:bibliotecadigital.usb.edu.co:10819/6689
Acceso en línea:
http://hdl.handle.net/10819/6689
Palabra clave:
EKF
State estimation
Glucose homeostasis
insulin
Mathematical mode
T1DM
DMT1
Estimación de estado
FKE
Homeostasis de la glucosa
Insulina
Modelo matemático
Diabetes Mellitus
Rights
License
Atribución-NoComercial-SinDerivadas 2.5 Colombia
Description
Summary:The glucose homeostasis is responsible for regulating the blood glucose concentration around 100 mg / dl. As soon as this mechanism is broken due to the inability of the pancreas to produce insulin, the blood glucose levels increase and patients are diagnosed with Type 1 Diabetes Mellitus (T1DM). Research has been directed towards the creation of an artificial pancreas allowing the regulation of glucose levels in blood. However, one of the main difficulties is that not all internal variables of the mathematical model can be measured directly, preventing the implementation of automatic controllers. Therefore, it is necessary to use estimation schemes to reconstruct the unknown states by using the available measurements. In the case of T1DM, the delay between the blood glucose and the interstitial glucose has a negative effect on the performance of state estimators, so the treatment of this delay is necessary either from the modeling process or by a modification of the estimation techniques. Two scenarios are discussed in this contribution. First, the is assumed to be measured from continuous glucose monitors (CGM) which introduces a measurement delay affecting the performance of the studied filter. Then, in a second scenario, the dynamic relationship between and is added to prevent the filter to deal directly with the measurement delay, then giving better results.