Echocardiographic Analysis by Artificial Intelligence in Hypertension

Conditions studied

Arterial Hypertension

About this study

The objective of this study is to develop artificial intelligence (AI) tools to characterize cardiac remodelling in arterial hypertension through the analysis of routine imaging data (echocardiography) coupled with patient data. More precisely, this study will make it possible (i) to represent in a relevant way the spectrum of the cardiac repercussions of a population affected by hypertension to better characterize the progress of the pathology, in particular vis-à-vis the more ambiguous subjects to be characterized ("grey" area); (ii) to develop tools for the automatic quantification of cardiac function (segmentation and robust monitoring of the myocardium during the cardiac cycle, for the dynamic analysis of the shape and global deformation of the heart) and therefore to extract descriptors richer in cardiac function than those currently used in clinical routine.

Interventions

• Other: Clinical cardiac parameters (Arterial pressure, number of anti-hypertensive medications, and the NT-proBNP measurements) — The degree of severity of the hypertension will have been previously diagnosed for each patient with the help of the blood pressure measurements, the number of antihypertensive treatments and the NT-proBNP measurement which is correlated with the left intraventricular pressure and the mass of the left ventricle. This information will be used (i) to guide the structuring of the pathology representation space during the learning phase from a sub-population (a portion of the collected database that will be used for training algorithms); (ii) to serve as an endpoint from a test sub-population (the remaining portion of the database collected).

Primary outcomes

• Arterial Blood Pressure (Baseline)
• Number of Antihypertensive Medications (Baseline)
• NT-proBNP Concentration (Baseline)

Eligibility information

Study locations

• Hopital Lyon Sud, Pierre-Bénite, 69310 France