Computerized Artificial Intelligence for Automatic Monitoring of Left Ventricular Function by Echocardiography
Abstract
Heart disease is a leading cause of mortality. In the perioperative and intensive care setting, left ventricular dysfunction is common and timely detection and monitoring of cardiac function are crucial to reduce associated complications. Perioperative and postoperative cardiac monitoring is valuable to guide therapy and vital to obtain early detection of disease, which is decisive in reducing cardiovascular complications and heart failure. This doctoral thesis introduced innovative methods for the automatic estimation of left ventricular function in critically ill patients through transesophageal echocardiography. The traditional method of manual and qualitative evaluation of cardiac ultrasound imaging, while effective, demands considerable time and expertise. In contrast, the proposed automated approach leverages state-of-the-art machine learning algorithms to facilitate rapid, consistent, and quantitative assessment, thereby enhancing the accessibility and efficiency of cardiac monitoring in clinical settings.
The core of this research involved the development of a robust artificial intelligence system capable of continuous, real-time monitoring of global and regional left ventricular function using transesophageal echocardiography data. By focusing on key functional parameters such as mitral annular plane systolic excursion and segmental longitudinal strain, the system aimed to offer precise and automated measurements. The culmination of this process was the estimation of critical metrics for the monitoring of patients during major surgeries or interventions with significant hemodynamic implications.
A noteworthy aspect of this research was the integration of the software developed with standard ultrasound equipment used in operating rooms and intensive care units. The method has the potential to improve the way cardiac function is monitored in these critical settings, offering a blend of precision, efficiency, and accessibility. The thesis not only delves into technological advances in artificial intelligence and medical imaging but also explores the practical implications of such an automated system in improving patient outcomes in critical care environments. The research, therefore, sits at the intersection of computer science, artificial intelligence, and medical imaging, offering a novel solution to a longstanding challenge in cardiac healthcare.
Has parts
Paper 1: Taskén, Anders Austlid; Yu, Jinyang; Berg, Erik Andreas Rye; Grenne, Bjørnar Leangen; Holte, Espen; Dalen, Håvard; Stølen, Stian Bergseng; Lindseth, Frank; Aakhus, Svend; Kiss, Gabriel Hanssen. Automatic Detection and Tracking of Anatomical Landmarks in Transesophageal Echocardiography for Quantification of Left Ventricular Function. Ultrasound in Medicine and Biology 2024. Published by Elsevier Inc. on behalf of World Federation for Ultrasound in Medicine & Biology. This is an open access article under the CC BY license. Available at: https://doi.org/10.1016/j.ultrasmedbio.2024.01.017Paper 2: Taskén, Anders Austlid; Berg, Erik Andreas Rye; Grenne, Bjørnar Leangen; Holte, Espen; Dalen, Håvard; Stølen, Stian Bergseng; Lindseth, Frank; Aakhus, Svend; Kiss, Gabriel Hanssen. Automated estimation of mitral annular plane systolic excursion by artificial intelligence from 3D ultrasound recordings. Artificial Intelligence in Medicine 2023 ;Volum 144. Published by Elsevier B.V. This is an open access article under the CC BY license. Available at: http://dx.doi.org/10.1016/j.artmed.2023.102646
Paper 3: Taskén, Anders Austlid; Judge, Thierry; Berg, Erik Andreas Rye; Yu, Jinyang; Grenne, Bjørnar; Lindseth, Frank; Aakhus, Svend; Jodoin, Pierre-Marc; Duchateau, Nicolas; Bernard, Olivier; Kiss, Gabriel. Estimation of Segmental Longitudinal Strain in Transesophageal Echocardiography by Deep Learning. This paper is submitted for publication and is therefore not included.
Paper 4: Berg, Erik Andreas Rye; Taskén, Anders Austlid; Nordal, Trym; Grenne, Bjørnar Leangen; Espeland, Torvald; Garstad, Idar Kirkeby; Dalen, Håvard; Holte, Espen; Stølen, Stian Bergseng; Aakhus, Svend; Kiss, Gabriel Hanssen. Fully automatic estimation of global left ventricular systolic function using deep learning in transesophageal echocardiography. European Heart Journal – Imaging Methods and Practice (EHJ-IMP) 2023 ;Volum 1.(1) Published by Oxford University Press on behalf of the European Society of Cardiology. This is an Open Access article distributed under the terms of the Creative Commons Attribution License CC BY. Available at: http://dx.doi.org/10.1093/ehjimp/qyad007
Paper 5: Yu, Jinyang; Taskén, Anders Austlid; Flade, Hans Martin; Skogvoll, Eirik; Berg, Erik Andreas Rye; Grenne, Bjørnar Leangen; Rimehaug, Audun Eskeland; Kirkeby-Garstad, Idar; Aakhus, Svend. Automatic assessment of left ventricular function for hemodynamic monitoring using artificial intelligence and transesophageal echocardiography. Journal of clinical monitoring and computing 2024 s. 281-291. © The Author(s), under exclusive licence to Springer Nature B.V. 2024. Available at: http://dx.doi.org/10.1007/s10877-023-01118-x
Paper 6: Yu, Jinyang; Taskén, Anders Austlid; Berg, Erik Andreas Rye; Tannvik, Tomas Dybos; Slagsvold, Katrine Hordnes; Kirkeby-Garstad, Idar; Grenne, Bjørnar Leangen; Kiss, Gabriel Hanssen; Aakhus, Svend. Continuous monitoring of left ventricular function in postoperative intensive care patients using artificial intelligence and transesophageal echocardiography. Intensive Care Medicine Experimental 2024 ;Volum 12.(1). Published by Springer. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License CC BY. Available at: http://dx.doi.org/10.1186/s40635-024-00640-9