AI-based system for the analysis of vegetations and risk assessment of embolism in patients with infective endocarditis
Article Sidebar
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.
Main Article Content
https://orcid.org/0000-0002-3297-9060
Abstract
The aim – to enhance the efficiency of infective endocarditis (IE) diagnosis and assess embolism risks by employing an intelligent computer-based diagnostic system.
Materials and methods. The study utilized intelligent computer processing of echocardiographic images from 20 patients (15 in the training group and 5 in the reference group) diagnosed with IE. The dataset comprised 668 images with pathologies (vegetations and abscesses) and 632 «clean» frames without pathological changes, in total 1,300 images in parasternal and apical views. The images were extracted from echocardiograms in DICOM format. Preprocessing steps included cropping, normalization, and contrast enhancement. To ensure the model’s quality, training, validation, and test sets contained images from different patients.
Results. The developed AI-based automated diagnostic system effectively identified vegetations on heart structures and determined their volume almost instantly, eliminating the potential for human error. This approach improves the accuracy, reliability, and speed of embolism risk assessment, enabling the optimization of the IE diagnostic protocol. The developed system was tested on images of a reference group of 5 patients with various IE progression states and in different projections. The system correctly predicted the presence of vegetation in each of the images where it was present, and reliably calculated its volume.
Conclusions. The proposed AI-based system significantly enhances the individualization and impartiality of the IE diagnostic process, improving its quality and reducing its duration. This provides the potential to enhance the protocol for IE examination and diagnosis.
Article Details
Keywords:
References
Hammond-Haleya M, Hartley A, Al-Khayatt BM, et al. Trends in the incidence and mortality of infective endocarditis in high-income countries between 1990 and 2019. Intern J Cardiol. 2023;371:441-51. https://doi.org/10.1016/j.ijcard.2022.09.061 DOI: https://doi.org/10.1016/j.ijcard.2022.09.061
Delgado V, Marsan NA, de Waha S, et al from ESC Scientific Document Group 2023 ESC Guidelines for the management of endocarditis: Developed by the task force on the management of endocarditis of the European Society of Cardiology (ESC) Endorsed by the European Association for Cardio-Thoracic Surgery (EACTS) and the European Association of Nuclear Medicine (EANM). Eur Heart J. 2023;44:3948-4042. https://doi.org/10.1093/eurheartj/ehad193 DOI: https://doi.org/10.1093/eurheartj/ehad625
Fernandez-Felix BM, Barca LV, Garcia-Esquinas E, et al. Prognostic models for mortality after cardiac surgery in patients with infective endocarditis: a systematic review and aggregation of prediction models. Clin Microbiol and Infection. 2021;27:1422-30. https://doi.org/10.1016/j.cmi.2021.05.051 DOI: https://doi.org/10.1016/j.cmi.2021.05.051
Cuervo G, Escrihuela-Vidal F, Gudiol C, et al. Current Challenges in the Management of Infective Endocarditis. Front. Med. 2021;8:641243. https://doi.org/10.3389/fmed.2021.641243 DOI: https://doi.org/10.3389/fmed.2021.641243
Nappi F, Moon MR, Martuscelli G, et al. Infective Endocarditis in High-Income Countries. Metabolites. 2022;12:682-717. https://doi.org/10.3390/metabo12080682 DOI: https://doi.org/10.3390/metabo12080682
Seung Jun Song, Jung Ho Kim, Nam Su Ku, et al. Vegetation Size, Multiplicity, and Position in Patients With Infective Endocarditis. Ann Thorac Surg. 2022;114:2253-61. https://doi.org/10.1016/j.athoracsur.2021.10.071 DOI: https://doi.org/10.1016/j.athoracsur.2021.10.071
Mihos CG, Nappi F. A narrative review of echocardiography in infective endocarditis of the right heart. Ann Transl Med. 2020;8:1622-33. http://dx.doi.org/10.21037/atm-20-5198 DOI: https://doi.org/10.21037/atm-20-5198
Tischler MD, Vaitkus PT. The ability of vegetation size on echocardiography to predict clinical complications: A meta-analysis. J Am Society Echocardiography. 1997;10(5):562-8. https://doi.org/10.1016/S0894-7317(97)70011-7. DOI: https://doi.org/10.1016/S0894-7317(97)70011-7
Paras ML, Jassar AS. Vegetation Size in Patients with Infective Endocarditis: Does Size Matter? 2022;35(6):576-8. https://doi.org/10.1016/j.echo.2022.02.007 DOI: https://doi.org/10.1016/j.echo.2022.02.007
Luaces M, Vilacosta I, Fernández C, et al. Vegetation size at diagnosis in infective endocarditis: Influencing factors and prognostic implications. Intern J Cardiol. 2009;137(1):76-8. https://doi.org/10.1016/j.ijcard.2008.05.011. DOI: https://doi.org/10.1016/j.ijcard.2008.05.011
Ghaffar Nia N, Kaplanoglu E, Nasab A. Evaluation of artificial intelligence techniques in disease diagnosis and prediction. Discov Artif Intell. 2023;3(1):5. https://doi.org/10.1007/s44163-023-00049-5. DOI: https://doi.org/10.1007/s44163-023-00049-5
Sineglazov V, Kot A. Design of hybrid neural networks of the ensemble structure. Eastern-European J Enterprise Technologies. 2021 Feb;1:31-45. https://doi.org/10.15587/1729-4061.2021.225301. DOI: https://doi.org/10.15587/1729-4061.2021.225301
Zgurovsky M, Sineglazov V, Chumachenko E. Classification and Analysis of Multicriteria Optimization Methods. In: Artificial Intelligence Systems Based on Hybrid Neural Networks. Vol. 904. Studies in Computational Intelligence. Springer, 2021. ISBN: 978-3-030-48452-1. https://doi.org/10.1007/978-3-030-48453-8_2. DOI: https://doi.org/10.1007/978-3-030-48453-8_2
Sineglazov V, Ryazanovskiy K, Chumachenko O. Multicriteria conditional optimization based on genetic algorithms. In: System research and information technologies. Dec. 2020, pp. 89-104. https://doi.org/10.20535/SRIT.2308-8893.2020.3.07. DOI: https://doi.org/10.20535/SRIT.2308-8893.2020.3.07
Muller D, Soto-Rey I, Kramer F. An Analysis on Ensemble Learning Optimized Medical Image Classification With Deep Convolutional Neural Networks. In: IEEE Access. 10 (2022), pp. 66467-80. https://doi.org/10.48550/arXiv.2201.11440. DOI: https://doi.org/10.1109/ACCESS.2022.3182399
Thambawita V, Hicks SA, Halvorsen P. Divergent Nets: Medical Image Segmentation by Network Ensemble. In: EndoCV@ISBI. 2021. https://doi.org/10.48550/arXiv.2107.00283.