Quantification of left ventricular (LV) size and function plays a crucial role in evaluating various cardiac conditions and is the most basic and frequent indication for echocardiography. Three-dimensional echocardiography (3DE) has been demonstrated to accurately and reproducibly quantify LV volumes and function and is recommended by current guidelines [1,2]. However, conventional manual 3DE has not been routinely applied in clinical practice because it is time-consuming and requires expertise [3].

Over the past decade, the development of AI approaches has resulted in automated border detection and automation of 3DE, making 3DE quantification easier, faster, and more accurate. Several commercially available automated quantitative software programs have been introduced and preliminary used. Among them, speckle tracking-based 4D LV-Analysis is a widely utilized 3DE quantitative software package. Its normal values have been established based on several multicenter, large-sample studies worldwide [[4], [5], [6]]. Recently, a novel machine-learning-based fully automated 3DE program, Dynamic Heart Model (DHM), was developed and widely studied [7]. This model-based program achieves quantification by automatically matching and adapting “heart models” trained on approximately one thousand 3D datasets with different heart sizes, shapes, and image quality [8]. Studies have demonstrated that DHM further minimizes the difference between 3DE and cardiovascular magnetic resonance (CMR), providing a new rapid, accurate, and reproducible method for LV quantification [9]. However, no studies have been conducted to establish normal values for DHM, and because of the different algorithm, it remains unclear whether the reference values for speckle tracking-based 4D LV-Analysis can be directly applied to model-based DHM [10].

Accordingly, this study aimed to determine the quantitative differences of LV volumes and LVEF between DHM and 4D LV-Analysis. If necessary, the study will also establish normal reference values for LV parameters measured by DHM and clarify its influencing factors.