[an error occurred while processing this directive]

Advanced Ultrasound in Diagnosis and Therapy ›› 2025, Vol. 9 ›› Issue (3): 298-306.doi: 10.26599/AUDT.2025.240067

• • 上一篇    

  

  • 收稿日期:2024-12-22 修回日期:2025-04-12 接受日期:2025-05-13 出版日期:2025-09-30 发布日期:2025-10-13

Advanced Diagnosis of Aortic Stenosis Disease Based on Ultrasound Images: A Novel Artificial Intelligence Approach

Elkouahy Fatima Ezzahraa,*(), Bennis Ahmedc, Merke Nicolasb, Ouahid Hajard, Malali Hamid Ela, Taleb Lhoucine Bena, Mouhsen Azeddinea   

  1. aEnergy-Materials-Instrumentation and Telecom Laboratory (EMIT), Faculty of Science and Technology, University Hassan 1st, Settat, Morocco
    bDepartment of Cardiothoracic and Vascular Surgery, German Heart Center Charité Berlin, Berlin, Germany
    cDepartment of Cardiology, University Hospital Hassan 2, Casablanca, Morocco
    dFaculty of Medicine and Pharmacy, Cadi Ayyad University, Bioscience, and Health Research Laboratory, Marrakech, Morocco
  • Received:2024-12-22 Revised:2025-04-12 Accepted:2025-05-13 Online:2025-09-30 Published:2025-10-13
  • Contact: Laboratory of Radiation-Matter and Instrumentation, Faculty of Science and Technology, University Hassan 1st, Settat, Morocco. e-mail:elkouahyfat.fst@uhp.ac.ma(PE E)

RichHTML

2

PDF (PC)

2114

Abstract:

Objective Aortic stenosis (AS), a prevalent valvular disease, demands accurate diagnosis. Current methods, notably Doppler echocardiography, face limitations like dynamic image challenges and reliance on cardiologist experience. To assess aortic stenosis, measuring the LVOT diameter is critical, as a 1 mm difference can result in a 10% variation in stroke volume. Accurate Doppler beam alignment and LVOT VTI measurement are also essential to avoid errors. Our study, utilizing the TMED 2 dataset, introduces a novel artificial intelligence program for precise aortic stenosis diagnosis. By leveraging AI, we aim to overcome existing constraints and significantly enhance diagnostic accuracy.
Methods a novel method that involves using convolutional neural networks (CNNs), were used to grade AS based on various views of transthoracic echocardiography (TTE) images from the TMED 2 dataset. This innovative method aimed to take advantage of CNN’s abilities to recognize detailed patterns in echocardiographic data, making AS diagnosis more accurate. We evaluated the performance of our CNN models using confusion metrics and the area under the receiver operator curve (AUROC).
Results Our CNN networks were trained on a dataset comprising view_and_diagnosis_labeled_set, which included 599 studies from 577 unique patients (some with multiple studies on distinct days). For classification, we chose three classes: no aortic stenosis, aortic stenosis, and mild aortic stenosis. The detection of aortic stenosis achieved an accuracy of 85.74%. External validation using three views (PLAX, PSAX, and A4C) of outpatient transthoracic echocardiograms demonstrated effective screening for AS, yielding respective AUROCs of 0.81, 0.88, and 0.48.
Conclusion Our novel CNN-based approach achieved an 85,74% accuracy in AS detection using diverse views from the TMED 2 dataset. External validation on outpatient echocardiograms demonstrated robust screening capabilities, with AUROCs of 0.81, 0.88, and 0.48 for PLAX, PSAX, and A4C views, respectively. These promising results suggest the potential of AI in improving AS diagnosis for clinical applications. Moving forward, our future endeavors will focus on addressing data imbalances and detecting the view of images, in addition to assessing the severity of aortic stenosis, to further refine and optimize our diagnostic approach.

Key words: Aortic stenosis, Convolutional neural networks, Doppler echocardiography, Parasternal long-axis, Artificial intelligence.

"

"

"

BiomarkersTrue severeLow flow, low gradientPseudo-low flow, low gradient
Aortic valve area (cm2/Indexed cm2/m2)≤ 1.0 / ≤ 0.6≤ 1.0 / ≤ 0.6≤ 1.0 / ≤ 0.6
Aortic velocity (m/s)> 4< 4< 4
Mean pressure drop (mmHg)> 40< 40< 40
Stroke volume index (mL/m2)> 35< 35< 35
Augmentation of flow parameters under inotropic stimulationNAYesNo Peak

"

"

"

"

"

"

"

"

"

"

[1] Zhenyi L, Ya C, Xinqi W, Lan Y, Anni C, Zhaojun L, et al. Left and right ventricular interaction: insight from echocardiography imaging. Advanced Ultrasound in Diagnosis and Therapy 2024; 8: 195-204.
[2] Vahanian A, Beyersdorf F, Praz F, Milojevic M, Baldus S, Bauersachs J, et al. 2021 ESC/EACTS Guidelines for the management of valvular heart disease. Eur Heart J 2022; 43: 561-632.
[3] Aluru JS, Barsouk A, Saginala K, Rawla P, Barsouk A. Valvular heart disease epidemiology. Med Sci (Basel) 2022; 10: 32
[4] Guedeney P, Collet JP. Aortic stenosis: an update. Rev Med Interne 2022; 43: 145-151.
[5] Chen W, Xie Y, Zhang Z, Zhu Y, Zhang Y, Zhu S, et al. Artificial intelligence-assisted medical imaging in interventional management of valvular heart disease. Advanced Ultrasound in Diagnosis & Therapy 2023; 7: 217-227.
[6] Pachulski RT, Chan KL. Progression of aortic valve dysfunction in 51 adult patients with congenital bicuspid aortic valve: assessment and follow up by Doppler echocardiography. Br Heart J 1993; 69: 237-240.
[7] Muhammad Y, Tahir M, Hayat M, Chong KT. Early and accurate detection and diagnosis of heart disease using intelligent computational model. Sci Rep 2020; 10: 19747.
[8] Tribouilloy C, Rusinaru D, Maréchaux S, Castel AL, Debry N, Maizel J, et al. Low-gradient, low-flow severe aortic stenosis with preserved left ventricular ejection fraction: characteristics, outcome, and implications for surgery. J Am Coll Cardiol 2015; 65: 55-66.
[9] Zoghbi WA, Galan A, Quinones MA. Accurate assessment of aortic stenosis severity by Doppler echocardiography independent of aortic jet velocity. Am Heart J 1988; 116: 855-863.
[10] Bang W C, Seong Y K, Lee J. The impact of deep learning on ultrasound in diagnosis and therapy: enhancing clinical decision support, workflow efficiency, quantification, image registration, and real-time assistance. Advanced Ultrasound in Diagnosis & Therapy (AUDT) 2023; 7: 204-216.
[11] Hagendorff A, Knebel F, Helfen A, Knierim J, Sinning C, Stöbe S, et al. Expert consensus document on the assessment of the severity of aortic valve stenosis by echocardiography to provide diagnostic conclusiveness by standardized verifiable documentation. Clin Res Cardiol 2020; 109: 271-288.
[12] Chernyshov A, Grue JF, Nyberg J, Grenne B, Dalen H, Aase SA, et al. Automated segmentation and quantification of the right ventricle in 2-D echocardiography. Ultrasound Med Biol 2024; 50: 540-548.
[13] Gill H, Fernandes J, Chehab O, Prendergast B, Redwood S, Chiribiri A, et al. Evaluation of aortic stenosis: from Bernoulli and Doppler to Navier-Stokes. Trends Cardiovasc Med 2023; 33: 32-43.
[14] Guzzetti E, Capoulade R, Tastet L, Garcia J, Le Ven F, Arsenault M, et al. Estimation of stroke volume and aortic valve area in patients with aortic stenosis: a comparison of echocardiography versus cardiovascular magnetic resonance. J Am Soc Echocardiogr 2020; 33: 953-963.e5.
[15] Saikrishnan N, Kumar G, Sawaya FJ, Lerakis S, Yoganathan AP. Accurate assessment of aortic stenosis: a review of diagnostic modalities and hemodynamics. Circulation 2014; 129: 244-253.
[16] Wessler BS, Huang Z, Long GM Jr, Pacifici S, Prashar N, Karmiy S, et al. Automated detection of aortic stenosis using machine learning. J Am Soc Echocardiogr 2023; 36: 411-420.
[17] Sun X, Yin Y, Yang Q, Huo T. Artificial intelligence in cardiovascular diseases: diagnostic and therapeutic perspectives. Eur J Med Res 2023; 28: 242.
[18] Milosevic M, Jin Q, Singh A, Amal S. Applications of artificial intelligence in multi-modal imaging for cardiovascular disease. Front Radiol 2024; 3: 1294068.
[19] Chen X, Yang F, Zhang P, Lin X, Wang W, Pu H, et al. Artificial intelligence–assisted left ventricular diastolic function assessment and grading: multiview versus single view. J Am Soc Echocardiogr 2023; 36: 1064-1078.
[20] Holste G, Oikonomou EK, Mortazavi BJ, Coppi A, Faridi KF, Miller EJ, et al. Severe aortic stenosis detection by deep learning applied to echocardiography. Eur Heart J 2023; 44: 4592-4604.
[21] Krishna H, Desai K, Slostad B, Bhayani S, Arnold JH, Ouwerkerk W, et al. Fully automated artificial intelligence assessment of aortic stenosis by echocardiography. J Am Soc Echocardiogr 2023; 36: 769-777.
[22] Huang Z, Long G, Wessler BS, Hughes MC. Tmed 2: a dataset for semi-supervised classification of echocardiograms. In DataPerf: Benchmarking Data for Data-Centric AI Workshop 2022.
[23] Huang Z, Sidhom MJ, Wessler BS, Hughes MC. Fix-A-Step: semi-supervised learning from uncurated unlabeled data. arXiv preprint arXiv:2208.11870 2023.
[24] Huang Z, Long G, Wessler B, Hughes MC. A new semi-supervised learning benchmark for classifying view and diagnosing aortic stenosis from echocardiograms. Machine Learning for Healthcare Conference 2021; 149: 614-647.
No related articles found!
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
No Suggested Reading articles found!
[an error occurred while processing this directive]