Key result
Machine learning LVOTd estimation matches expert accuracy while boosting stroke volume calculation yield to 92%.
Why the study?
The accuracy of machine learning computer model versus human expert measurement of LVOTd in estimating stroke volume relative to a pulmonary artery catheter was unknown.
Does a machine learning computer model for estimating LVOTd improve the yield and maintain the accuracy of stroke volume measurement with echocardiography compared to human expert measurement in patients with a PAC?
Observational (n=84)
No
Does a machine learning computer model for estimating LVOTd improve the yield and maintain the accuracy of stroke volume measurement with echocardiography compared to human expert measurement in patients with a PAC?
A machine learning model estimating LVOTd allows accurate calculation of stroke volume using only VTI, increasing measurement yield from 70% to 92% compared to human expert measurement.
Answers are grounded in this paper and cite their sources.
CM-derived LVOTd raises SV measurement yield to 92% with bias comparable to expert LVOTd versus PAC; leaves open adoption pending prospective validation.
The stroke volume (SV) can be measured by a human expert (HE) using the left ventricular outflow tract diameter (LVOTd) and its velocity time integral (VTI) with echo. If the SV is known, the cardiac output and systemic vascular resistance can be calculated. We have previously described a machine learning computer model (CM) to estimate the LVOTd. The accuracy of LVOTd CM vs. LVOTd HEM in estimation of the SV relative to a pulmonary artery catheter (PAC) is unknown. Over 20 months, a prospective observational study was done in patients with a PAC placed for clinical indications. Participants underwent an echocardiogram, and at the same time the SV was measured by PAC. The SV with echo was calculated using the LVOTd HEM and LVOTd CM and the VTI. The bias and agreement between the metrics and PAC were determined with Bland-Altman analysis. Computer modeling improves measurement of stroke volume with echocardiography. Eighty-four patients with a PAC were enrolled. With human expert measurement of the LVOTd and the VTI, the SV could be calculated in 59 (70%), and with the LVOTd CM and the VTI in 78 (92%). Bland Altman analysis comparing PAC and HEM yielded a mean bias of 3.1 with 95% limits of agreement (LOA) -28.8 and 34.9. When comparing PAC and CM, bias was 1.75 and LOA of -29 and 32.5. A machine learning model of the LVOTd allows accurate calculation of the SV using only the VTI, simplifying the assessment while increasing the yield by 22%. Machine learning tools can lower the bar for obtaining quantitative metrics with echo. Adding objective and repeatable measures will improve the ability of echo to guide therapy in the critically ill.
No takes yet. Share an insight, caveat, or question.
Murthi et al. (2026) conducted an observational in Critically ill patients requiring hemodynamic monitoring (n=84). Machine learning computer model for left ventricular outflow tract diameter (LVOTdCM) vs. Human expert measurement (LVOTdHEM) was evaluated on Mean bias in stroke volume measurement compared to pulmonary artery catheter (PAC) (mean bias 1.75, 95% CI -29 to 32.5). A machine learning computer model for estimating left ventricular outflow tract diameter yielded a mean bias of 1.75 compared to pulmonary artery catheter, similar to human expert measurement, while increasing the yield of stroke volume calculation from 70% to 92%.
Synapse has enriched 10 related papers. Consider them for comparative context: