Key result
Automated deep learning measures left atrial volume within ~3 mL of manual 2-D assessments.
Why the study?
Echocardiographic assessment of left atrial volume often relies on manual or semi-automated measurements, which are time-consuming and operator-dependent.
Does a fully automated deep learning pipeline accurately measure 2-D and 3-D left atrial volume compared to manual expert assessments?
Does a fully automated deep learning pipeline accurately measure 2-D and 3-D left atrial volume compared to manual expert assessments?
Effect estimate: 2.9 mL (2-D) and 7.8 mL (3-D) mean difference
A fully automated deep learning pipeline can estimate 2-D and 3-D left atrial volume in real time with high accuracy comparable to manual expert assessments.
May enable real-time automated LA volumetry in echocardiography workflows; leaves open impact on clinical decision-making and outcomes.
OBJECTIVE: Echocardiography, a critical tool for assessing left atrial (LA) volume, often relies on manual or semi-automated measurements. This study introduces a fully automated, real-time method for measuring LA volume in both 2-D and 3-D imaging, in the aim of offering accuracy comparable to that of expert assessments while saving time and reducing operator variability. METHODS: We developed an automated pipeline comprising a network to identify the end-systole (ES) time point and robust 2-D and 3-D U-Nets for segmentation. We employed data sets of 789 2-D images and 286 3-D recordings and explored various training regimes, including recurrent networks and pseudo-labeling, to estimate volume curves. RESULTS: Our baseline results revealed an average volume difference of 2.9 mL for 2-D and 7.8 mL for 3-D, respectively, compared with manual methods. The application of pseudo-labeling to all frames in the cine loop generally led to more robust volume curves and notably improved ES measurement in cases with limited data. CONCLUSION: Our results highlight the potential of automated LA volume estimation in clinical practice. The proposed prototype application, capable of processing real-time data from a clinical ultrasound scanner, provides valuable temporal volume curve information in the echo lab.
No takes yet. Share an insight, caveat, or question.
Hu et al. (2023) studied Left atrial volume assessment. Automated 2-D and 3-D U-Nets for left atrial volume segmentation vs. Manual methods was evaluated on Average volume difference (2.9 mL (2-D) and 7.8 mL (3-D) mean difference). Automated deep learning-based left atrial volume measurement yielded an average volume difference of 2.9 mL for 2-D images and 7.8 mL for 3-D recordings compared with manual methods.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: