Keywords
Cardiac ejection fraction
Emergency department
Point-of-care ultrasound
Artificial intelligence
How to Cite
Abstract
Background: The integration of artificial intelligence (AI) in key cardiac function parameters, such as left ventricular ejection fraction (LVEF), can hold important value for clinicians, both in terms of time consumption and interobserver variability. However, the reproducibility between devices remains unknown.
Aim: The purpose of this study was to assess two ultrasound devices with their automated LVEF (auto-LVEF) measurements: the midrange GE venue (GEv), and the handheld Butterfly iQ+(Bfi); regarding correlation in ejection fraction (EF), time consumption, and image quality (IQ).
Method: Adult emergency room patients were included and scanned using both ultrasound devices by a novice operator. In each case, the objective was to acquire an apical four-chamber view and calculate the EF with each device’s pre-installed AI software. Out of those, 12 patients were rescanned by a physician experienced in cardiac ultrasound to evaluate the interoperator agreement.
Results: A total of 150 patients were included, with a median age of 64 years; 51% were female. The GEv and Bfi successfully generated auto-EF measurements in 73% (95% confidence interval [CI]: 65%–80%) and 52% (95% CI: 44–60%) of cases, respectively. The agreement in EF measurements between the GEv's real-time EF and the Bfi's Simpson monoplane method was high with a correlation coefficient r = 0.70 (0.60–0.77), p < 0.001. Bland-Altman analysis demonstrated a bias of 0.84% (95% upper and lower limits of agreement: 15.0% and -13.3%). The median scanning time in both apparatuses was 2 minutes (IQR GEv 1–2, IQR Bfi 1–3), the median IQ score was 4/5 (IQR 4–5) in GEv and 3.5/5 (IQR 3–4) in Bfi. The interobserver agreement was high, with a Kappa of κGEv = 0.75 and κBfi = 0.82.
Conclusion: In conclusion, Bfi had a lower success rate in calculating EF and a lower IQ than GEv. However, when auto-EF was successfully obtained, a strong correlation was observed between the machines.
References
Lang RM, Badano LP, Mor-Avi V, Afilalo J, Armstrong A, Ernande L, et al. Recommendations for cardiac chamber quantification by echocardiography in adults: an update from the American Society of Echocardiography and the European Association of Cardiovascular Imaging. Eur Heart J Cardiovasc Imaging. 2015;16(3):233-70. doi: 10.1093/ehjci/jev014.
Spahillari A, Colbert JA, Fox J, Singh J, Nayyar P, Chen S, et al. On-call transthoracic echocardiographic interpretation by first year cardiology fellows: comparison with attending cardiologists. BMC Med Educ. 2019;19(1):213. doi: 10.1186/s12909-019-1645-5.
Asch FM, Poilvert N, Abraham T, Jankowski M, Cleve J, Adams M, et al. Automated echocardiographic quantification of left ventricular ejection fraction without volume measurements using a machine learning algorithm mimicking a human expert. Circ Cardiovasc Imaging. 2019;12(9):e009303. doi: 10.1161/CIRCIMAGING.119.009303.
Hila Best PU. GE Healthcare. Venue™ Family Real-Time EF. 2022. Available from: https://www.gehealthcare.dk/. doi: not available. [Access date 19. september 2024].
Butterfly Network. Calculate Simpson's Ejection Fraction. 2022. Available from: https://www.butterflynetwork.com/. doi: not available. [Access date 19. september 2024].
Harris PA, Taylor R, Thielke R, Payne J, Gonzalez N, Conde JG. Research electronic data capture (REDCap) – a metadata-driven methodology and workflow process for providing translational research informatics support. J Biomed Inform. 2009;42(2):377-81. doi: 10.1016/j.jbi.2008.08.010.
Viera AJ, Garrett JM. Understanding interobserver agreement: the kappa statistic. Fam Med. 2005;37(5):360-3. doi: not available.
Barry T, Odreman L, Jain S, Campbell M, Nakatani S, Mihailescu SD. The role of artificial intelligence in echocardiography. J Imaging. 2023;9(2):40. doi: 10.3390/jimaging9020040.
Jafari MH, Habijan M, LeBel A, Tsai K, Fenster A. Automatic biplane left ventricular ejection fraction estimation with mobile point-of-care ultrasound using multi-task learning and adversarial training. Int J Comput Assist Radiol Surg. 2019;14(6):1027-37. doi: 10.1007/s11548-019-01955-9
Dadon Z, Eitan A, Raz A, Freund Y, Levy G. Use of artificial intelligence as a didactic tool to improve ejection fraction assessment in the emergency department: a randomized controlled pilot study. AEM Educ Train. 2022;6(2):e10738. doi: 10.1002/aet2.10738.
Moal O, Brunet M, Couronne R, D’Hooge J, Bernard O. Explicit and automatic ejection fraction assessment on 2D cardiac ultrasound with a deep learning-based approach. Comput Biol Med. 2022;146:105637. doi: 10.1016/j.compbiomed.2022.105637.
Asch FM, Jankowski M, Cleve J, Adams M, Markuzon N, Manovel A, et al. Deep learning-based automated echocardiographic quantification of left ventricular ejection fraction: A point-of-care solution. Circ Cardiovasc Imaging. 2021;14(6):e012293. doi: 10.1161/CIRCIMAGING.120.012293.
Varudo R, Bento L, Vitorino R, Vieira J, Miro V, Guerreiro R, et al. Machine learning for the real-time assessment of left ventricular ejection fraction in critically ill patients: a bedside evaluation by novices and experts in echocardiography. Crit Care. 2022;26(1):386. doi: 10.1186/s13054-022-04255-4.
Bacariza J, Sousa C, Teixeira C, Pinto J, Ribeiro L, Martins R, et al. Smartphone-based automatic assessment of left ventricular ejection fraction with a silicon chip ultrasound probe: a prospective comparison study in critically ill patients. Br J Anaesth. 2023;130(6):e485-7. doi: 10.1016/j.bja.2023.02.014.
Heinen A, Schulte J, Scherschel K, Meyer C, Kelm M, Kleinbongard P, et al. Echocardiographic analysis of cardiac function after infarction in mice: validation of single-plane long-axis view measurements and the bi-plane Simpson method. Ultrasound Med Biol. 2018;44(7):1544-55. doi: 10.1016/j.ultrasmedbio.2018.03.008.
Johri AM, Picard MH, Newell J, Marshall JE, King MEE, Hung J. Can a teaching intervention reduce interobserver variability in LVEF assessment: a quality control exercise in the echocardiography lab. JACC Cardiovascular imaging [Internet]. 2011 Aug 1 [cited 2023 Apr 17];4(8):821–9. doi: 10.1016/j.jcmg.2011.06.004.
This work is licensed under a Creative Commons Attribution 4.0 International License.
Copyright (c) 2025 Meryem Hamodi, Annmarie Touborg Lassen, Stefan Posth