TY - GEN

T1 - Statistical Shape Analysis Reveals Specific Features of Left Ventricular Wall Motion in Patients with Chronic Heart Failure

AU - Rokeakh, Roman

AU - Chumarnaya, Tatiana

AU - Zubarev, Stepan

AU - Lebedev, Dmitry

AU - Solovyova, Olga

N1 - This work was supported by Russian Science Foundation Grant No. 19-14-00134.

PY - 2023/9/28

Y1 - 2023/9/28

N2 - Ventricular myocardial tissue is a heterogeneous system which is finely tuned in space and time and provides rich kinematics of the ventricular wall. Cardiac pathology leads to specific changes in regional kinetics and deformations of the ventricular walls. Evaluation of the left ventricular (LV) wall motion is of great practical importance in the clinical diagnosis of cardiovascular pathology. Novel data analysis techniques are used to obtain more information about LV wall motion and deformation during the contractile cycle. Recently, the statistical shape analysis approach has become widespread. We tested this approach on echocardiographic imaging data from healthy subjects and patients with chronic heart failure (CHF). The main objective was to assess the predictive potential of features derived from LV wall motion analysis for the stratification of healthy and CHF hearts. Features describing LV shape and motion were extracted in an unsupervised manner. Using visual inspection and correlation analysis, we obtained a number of clinically relevant interpretations for the extracted features. We showed that the derived features are able to discriminate between healthy individuals and CHF patients with a great accuracy. Our results suggest that the statistical shape analysis approach is promising for further use as a diagnostic tool in cardiology.

AB - Ventricular myocardial tissue is a heterogeneous system which is finely tuned in space and time and provides rich kinematics of the ventricular wall. Cardiac pathology leads to specific changes in regional kinetics and deformations of the ventricular walls. Evaluation of the left ventricular (LV) wall motion is of great practical importance in the clinical diagnosis of cardiovascular pathology. Novel data analysis techniques are used to obtain more information about LV wall motion and deformation during the contractile cycle. Recently, the statistical shape analysis approach has become widespread. We tested this approach on echocardiographic imaging data from healthy subjects and patients with chronic heart failure (CHF). The main objective was to assess the predictive potential of features derived from LV wall motion analysis for the stratification of healthy and CHF hearts. Features describing LV shape and motion were extracted in an unsupervised manner. Using visual inspection and correlation analysis, we obtained a number of clinically relevant interpretations for the extracted features. We showed that the derived features are able to discriminate between healthy individuals and CHF patients with a great accuracy. Our results suggest that the statistical shape analysis approach is promising for further use as a diagnostic tool in cardiology.

UR - http://www.scopus.com/inward/record.url?partnerID=8YFLogxK&scp=85180372368

U2 - 10.1109/CSGB60362.2023.10329847

DO - 10.1109/CSGB60362.2023.10329847

M3 - Conference contribution

SN - 979-835030797-9

SP - 231

EP - 235

BT - 2023 IEEE Ural-Siberian Conference on Computational Technologies in Cognitive Science, Genomics and Biomedicine, CSGB 2023 - Proceedings

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 2023 IEEE Ural-Siberian Conference on Computational Technologies in Cognitive Science, Genomics and Biomedicine (CSGB)

Y2 - 28 September 2023 through 30 September 2023

ER -