From high blood pressure to diabetes. Identification of risk factors through machine learning with heterogeneous data (Q3179048): Difference between revisions
Jump to navigation
Jump to search
(Created claim: summary (P836): Chronic pathologies account for 75 % of health expenditure. This fact, coupled with the increase in life expectancy, makes it very interesting to study what factors influence the progression of chronic pathologies. Hypertension (HTA) is one of the most prevalent chronic pathologies, and may be associated with the onset of other chronic conditions such as type 2 diabetes mellitus (DM-2). In addition to the comorbidities of AH and DM-2, their simu...) |
(Changed label, description and/or aliases in en: translated_label) |
||
| label / en | label / en | ||
From high blood pressure to diabetes. Identification of risk factors through machine learning with heterogeneous data | |||
Revision as of 20:26, 12 October 2021
Project Q3179048 in Spain
| Language | Label | Description | Also known as |
|---|---|---|---|
| English | From high blood pressure to diabetes. Identification of risk factors through machine learning with heterogeneous data |
Project Q3179048 in Spain |
Statements
12,750.0 Euro
0 references
25,500.0 Euro
0 references
50.0 percent
0 references
1 January 2018
0 references
31 March 2021
0 references
UNIVERSIDAD REY JUAN CARLOS
0 references
40°16'56.93"N, 3°47'32.42"W
0 references
28058
0 references
Las patologías crónicas representan el 75% del gasto sanitario. Este hecho, unido al aumento de la esperanza de vida, hace muy interesante estudiar qué factores influyen en la progresión de patologías crónicas. La hipertensión arterial (HTA) es una de las patologías crónicas de mayor prevalencia, pudiendo estar asociada al inicio de otras condiciones crónicas como la diabetes mellitus tipo 2 (DM-2). Además de las comorbilidades de HTA y DM-2, su ocurrencia simultánea aumenta significativamente el riesgo de eventos cardiovasculares. Puesto que no existe un registro unificado de toda la información asistencial y farmacológica, se utilizarán datos heterogéneos recogidos por el Hospital Universitario de Móstoles (HUM) y el Hospital Universitario de Fuenlabrada (HUF), apoyados tecnológicamente por la Universidad Rey Juan Carlos. Por un lado, los datos asociados al HUM proceden de una población derivada a la Unidad de Hipertensión y fueron recogidos durante el seguimiento dentro de esta unidad. Por otro lado, los datos clínicos aportados por el HUF corresponden a diagnósticos y dispensación farmacológica de toda la población adscrita al hospital. Los métodos de aprendizaje automático han mostrado su potencial para identificar variables relevantes y realizar inferencia a partir de datos suficientemente representativos del problema. El análisis de las inter-relaciones esperablemente complejas entre los factores de riesgo, así como la heterogeneidad de los datos intra e inter bases de datos, necesitará sin duda de la creación de herramientas de análisis avanzadas y adaptadas a ambos escenarios clínicos y a su explotación transversal. El objetivo principal del proyecto es diseñar nuevas herramientas de aprendizaje automático que permitan identificar factores de riesgo para caracterizar la progresión de un paciente hipertenso a DM-2, así como determinar su potencial relación con eventos cardiovasculares. (Spanish)
0 references
Chronic pathologies account for 75 % of health expenditure. This fact, coupled with the increase in life expectancy, makes it very interesting to study what factors influence the progression of chronic pathologies. Hypertension (HTA) is one of the most prevalent chronic pathologies, and may be associated with the onset of other chronic conditions such as type 2 diabetes mellitus (DM-2). In addition to the comorbidities of AH and DM-2, their simultaneous occurrence significantly increases the risk of cardiovascular events. Since there is no unified record of all care and pharmacological information, heterogeneous data collected by the University Hospital of Móstoles (HUM) and the University Hospital of Fuenlabrada (HUF), technologically supported by the Rey Juan Carlos University, will be used. On the one hand, the data associated with HUM come from a population derived from the Hypertension Unit and were collected during follow-up within this unit. On the other hand, the clinical data provided by the HUF correspond to diagnosis and pharmacological dispensing of the entire population attached to the hospital. Machine learning methods have shown their potential to identify relevant variables and make inference from sufficiently representative data of the problem. The analysis of the expected complex inter-relationships between risk factors, as well as the heterogeneity of intra and inter-database data, will undoubtedly require the creation of advanced analysis tools adapted to both clinical scenarios and their cross-sectional exploitation. The main objective of the project is to design new machine learning tools to identify risk factors to characterise the progression of a hypertensive patient to DM-2, as well as to determine their potential relationship with cardiovascular events. (English)
12 October 2021
0 references
Fuenlabrada
0 references
Identifiers
DTS17_00158
0 references