Ultrasonic Texture Features for Assessing Cardiac Remodeling and Dysfunction

Quincy A Hathaway; Naveena Yanamala; Nanda K Siva; Donald A Adjeroh; John M Hollander; Partho P Sengupta

doi:10.1016/j.jacc.2022.09.036

Ultrasonic Texture Features for Assessing Cardiac Remodeling and Dysfunction

J Am Coll Cardiol. 2022 Dec 6;80(23):2187-2201. doi: 10.1016/j.jacc.2022.09.036.

Authors

Quincy A Hathaway¹, Naveena Yanamala², Nanda K Siva³, Donald A Adjeroh⁴, John M Hollander⁵, Partho P Sengupta⁶

Affiliations

¹ Heart and Vascular Institute, West Virginia University, Morgantown, West Virginia, USA. Electronic address: https://twitter.com/QuincyHathaway.
² Rutgers Robert Wood Johnson Medical School, New Brunswick, New Jersey, USA. Electronic address: https://twitter.com/YanamalaNaveena.
³ Heart and Vascular Institute, West Virginia University, Morgantown, West Virginia, USA.
⁴ Lane Department of Computer Science and Electrical Engineering, West Virginia University, Morgantown, West Virginia, USA.
⁵ Division of Exercise Physiology, West Virginia University, School of Medicine, Morgantown, West Virginia, USA.
⁶ Rutgers Robert Wood Johnson Medical School, New Brunswick, New Jersey, USA. Electronic address: partho.sengupta@rutgers.edu.

PMID: 36456049
DOI: 10.1016/j.jacc.2022.09.036

Abstract

Background: Changes in cardiac size, myocardial mass, cardiomyocyte appearance, and, ultimately, the function of the entire organ are interrelated features of cardiac remodeling that profoundly affect patient outcomes.

Objectives: This study proposes that the application of radiomics for extracting cardiac ultrasonic textural features (ultrasomics) can aid rapid, automated assessment of left ventricular (LV) structure and function without requiring manual measurements.

Methods: This study developed machine-learning models using cardiac ultrasound images from 1,915 subjects in 3 clinical cohorts: 1) an expert-annotated cardiac point-of-care-ultrasound (POCUS) registry (n = 943, 80% training/testing and 20% internal validation); 2) a prospective POCUS cohort for external validation (n = 275); and 3) a prospective external validation on high-end ultrasound systems (n = 484). In a type 2 diabetes murine model, echocardiography of wild-type (n = 10) and Leptr^-/- (n = 8) mice were assessed longitudinally at 3 and 25 weeks, and ultrasomics features were correlated with histopathological features of hypertrophy.

Results: The ultrasomics model predicted LV remodeling in the POCUS and high-end ultrasound external validation studies (area under the curve: 0.78 [95% CI: 0.68-0.88] and 0.79 [95% CI: 0.73-0.86], respectively). Similarly, the ultrasomics model predicted LV remodeling was significantly associated with major adverse cardiovascular events in both cohorts (P < 0.0001 and P = 0.0008, respectively). Moreover, on multivariate analysis, the ultrasomics probability score was an independent echocardiographic predictor of major adverse cardiovascular events in the high-end ultrasound cohort (HR: 8.53; 95% CI: 4.75-32.1; P = 0.0003). In the murine model, cardiomyocyte hypertrophy positively correlated with 2 ultrasomics biomarkers (R² = 0.57 and 0.52, Q < 0.05).

Conclusions: Cardiac ultrasomics-based biomarkers may aid development of machine-learning models that provide an expert-level assessment of LV structure and function.

Keywords: MACE; automated; machine-learning; mouse; radiomics; ultrasomics.

Publication types

Research Support, U.S. Gov't, Non-P.H.S.
Research Support, Non-U.S. Gov't

MeSH terms

Animals
Cardiovascular Diseases*
Diabetes Mellitus, Type 2*
Disease Models, Animal
Hypertrophy
Mice
Myocytes, Cardiac
Prospective Studies
Ultrasonics
Ventricular Remodeling