Real-life clinical validation of AI-assisted echocardiography for aortic root and LVOT measurements against human readers and cardiac magnetic resonance

Q: What are the key findings of this study?

AI-assisted echocardiography demonstrated comparable agreement to expert readers for aortic root and LVOT measurements against cardiac magnetic resonance, achieving an ICC of 0.70 for LVOT.

Q: What is the clinical evidence from this study?

Study design: Observational. Population: Suspected myocarditis, heart failure, or cardiomyopathies (n=169). Intervention: AI-assisted echocardiography (US2.ai) vs. Expert manual echocardiography and cardiac magnetic resonance. Primary outcome: Agreement (Intraclass Correlation Coefficient) between AI and CMR for LVOT diameter (ICC 0.70, 95% CI 0.61-0.77, p=<0.001).

Q: What does this research mean for the field?

AI-assisted echocardiography demonstrates measurement agreement with cardiac magnetic resonance comparable to expert human readers for aortic root and LVOT diameters, though clinically significant discrepancies occur in up to 36.7% of cases requiring physician oversight. Novelty: ClaimNovelty.CONFIRMATORY. Consensus alignment: ConsensusAlignment.NEUTRAL.

Q: What question did this study set out to answer?

To evaluate the agreement between AI-assisted transthoracic echocardiography, expert manual measurements, and cardiac magnetic resonance as a reference standard for aortic root and left ventricular outflow tract diameters.

Patrycja Mołek-Dziadosz; Oleksandra Bondarchuk; Aleksandra Woźniak; Maria Królikowska; Agnieszka Mirek; Anna Furman-Niedziejko; Karol Mazur; Elżbieta Ostrowska-Kaim; Karolina Golińska‐Grzybała; Aleksander Siniarski; Joanna Szachowicz-Jaworska; Tomasz Miszalski-Jamka; Radosław Rychlak; Marc R. Dweck; Jadwiga Nessler; Andrzej Gackowski

doi:10.1093/ehjdh/ztag126

August 5, 2026European Heart Journal - Digital HealthOpen Access

Real-life clinical validation of AI-assisted echocardiography for aortic root and LVOT measurements against human readers and cardiac magnetic resonance

View Full Paper

Ask AI

Bookmark

Share

Key result

AI-assisted echocardiography matches expert accuracy for LVOT measurements against CMR with an ICC of 0.70.

ICC 0.70
95% CI 0.61-0.77
P<0.001
n=169

Figures

Graphical abstract published by European Heart Journal - Digital Health — Graphical abstract · Published by European Heart Journal - Digital Health · CC BY-NC

Why the study?

AI algorithms may improve echocardiographic measurement standardization, but direct validation against CMR remains limited.

Does AI-assisted echocardiography provide comparable agreement to expert manual echocardiography and CMR for measuring aortic root and LVOT diameters in adult patients?

Key Points

To evaluate the agreement between AI-assisted transthoracic echocardiography, expert manual measurements, and cardiac magnetic resonance as a reference standard for aortic root and left ventricular outflow tract diameters.
Analyzed 169 adult patients with complete AI-derived measurements of the left ventricular outflow tract (LVOT), sinotubular junction (STJ), and sinus of Valsalva (SoV) who underwent transthoracic echocardiography (TTE) and cardiovascular magnetic resonance (CMR) within seven days.
Compared automated measurements from the US2.AI platform against those of two expert echocardiographers and CMR reference standards using intraclass correlation coefficients (ICC) and Bland–Altman analysis.
AI-derived measurements showed agreement with CMR, with an ICC of 0.70 (95% CI: 0.61–0.77) for LVOT, 0.77 (95% CI: 0.70–0.83) for STJ, and 0.77 (95% CI: 0.70–0.83) for SoV.
Echo experts showed ICCs against CMR ranging from 0.51 (95% CI: 0.36–0.65) to 0.63 (95% CI: 0.50–0.73) for LVOT, 0.71 (95% CI: 0.63–0.78) to 0.82 (95% CI: 0.73–0.87) for STJ, and 0.85 (95% CI: 0.78–0.89) to 0.85 (95% CI: 0.79–0.89) for SoV.
Clinically significant differences between AI and CMR measurements occurred in 36.7% of LVOT (difference ≥2 mm), 20% of STJ (difference ≥4 mm), and 11.8% of SoV (difference ≥4 mm) measurements.

Study Design

Type

Observational (n=169)

Blinding

Blinded

Multicenter

No

Structured PICO

Does AI-assisted echocardiography provide comparable agreement to expert manual echocardiography and CMR for measuring aortic root and LVOT diameters in adult patients?

P

Population

169 adult patients (median age 52, 30% female) who underwent both transthoracic echocardiography and cardiac magnetic resonance imaging within 7 days.

E

Exposure

Fully automated AI-assisted transthoracic echocardiography (TTE) using the US2.ai software (US2.v2 version) for measurements of the left ventricular outflow tract (LVOT), sinus of Valsalva (SoV), and sinotubular junction (STJ).

C

Comparator

Expert manual TTE (performed by two independent experienced echocardiographers) and Cardiac Magnetic Resonance (CMR) as the reference standard.

O

Outcome

Agreement between AI-assisted TTE, expert manual TTE, and CMR for aortic root (SoV, STJ) and LVOT diameters, assessed using intraclass correlation coefficients (ICC) and Bland-Altman analysis.surrogate

Main Result

Effect estimate: ICC 0.70 (95% CI 0.61-0.77)

p-value: p=<0.001

Limitations

Retrospective, single-center study using one AI platform
CMR measurements obtained in 3-chamber view rather than 3D multiplanar reconstruction
Exclusion of patients with prior valve surgery or very poor image quality
Potential spectrum bias due to specific CMR referral indications
AI systems perform well under optimal conditions but lack the flexibility of human experts in borderline-quality cases
Errors due to incorrect cardiac phase selection (e.g., measuring LVOT on closed cusps)
Errors due to partial or anatomically confounded measurements
Errors due to poor image quality

Implication

AI-assisted echocardiography provides automated aortic root and LVOT measurements with agreement comparable to human experts and CMR, though physician oversight remains necessary for borderline-quality images.

Abstract

Aims AI algorithms may improve echocardiographic measurement standardization, but direct validation against cardiovascular magnetic resonance (CMR) remains limited. This study aimed to evaluate the agreement between AI-assisted transthoracic echocardiography (TTE), expert manual TTE, and CMR as a gold standard for aortic root and left ventricular outflow tract (LVOT) diameters. Methods and results Out of 183 patients with analyzable TTE and CMR recordings performed within seven days, 169 had complete measurements of LVOT, sinotubular junction (SoV) and sinus of Valsalva (STJ) by AI and were included in the primary analysis. Automated AI measurements were obtained using the US2.AI platform and compared with those of expert echocardiographers and CMR. Agreement was assessed using intraclass correlation coefficients (ICC) and Bland-Altman analysis. Agreement between Expert 1 and CMR, and Expert 2 and CMR, yielded ICCs: 0.63 (95% CI: 0.50–0.73) and 0.51 (95% CI: 0.36–0.65) for LVOT, 0.71 (95% CI: 0.63–0.78) and 0.82 (95% CI: 0.73–0.87) for STJ, and 0.85 (95% CI: 0.79–0.89) to 0.85 (95% CI: 0.78–0.89) for SoV, respectively. AI-derived measurements demonstrated comparable agreement with CMR: ICC 0.70 (95% CI: 0.61–0.77) for LVOT, 0.77 (95% 0.70–0.83) for STJ, and 0.77 (95% 0.70–0.83) for SoV. Clinically significant differences (≥2 mm for LVOT and ≥4 mm for STJ and SoV) between AI vs CMR measurements were observed in 36.7% of LVOT, 20% of STJ, and 11.8% of SoV measurements. Conclusion AI-assisted echocardiography showed comparable agreement for aortic root and LVOT measurements, but physician oversight remains necessary.

Discussion

No takes yet. Share an insight, caveat, or question.

Cite This Study

Mołek-Dziadosz et al. (2026) conducted an observational in Suspected myocarditis, heart failure, or cardiomyopathies (n=169). AI-assisted echocardiography (US2.ai) vs. Expert manual echocardiography and cardiac magnetic resonance was evaluated on Agreement (Intraclass Correlation Coefficient) between AI and CMR for LVOT diameter (ICC 0.70, 95% CI 0.61-0.77, p=<0.001). AI-assisted echocardiography demonstrated comparable agreement to expert readers for aortic root and LVOT measurements against cardiac magnetic resonance, achieving an ICC of 0.70 for LVOT.

synapsesocial.com/papers/6a76da8cf12abadc79814ef4 https://doi.org/10.1093/ehjdh/ztag126

View Full Paper

Ask AI

Bookmark

Share

Real-life clinical validation of AI-assisted echocardiography for aortic root and LVOT measurements against human readers and cardiac magnetic resonance

Figures

Key Points

Study Design

Structured PICO

Main Result

Limitations

Implication

Abstract

Discussion

Cite This Study

Real-life clinical validation of AI-assisted echocardiography for aortic root and LVOT measurements against human readers and cardiac magnetic resonance

Figures

Key Points

Study Design

Structured PICO

Main Result

Limitations

Implication

Abstract

Discussion

Cite This Study

Keep reading

Keep reading