Volume 29, Issue 3 (5-2022)
RJMS 2022, 29(3): 221-229 |
Back to browse issues page
Research code: 0
Ethics code: 0
Clinical trials code: 0
Download citation:
BibTeX | RIS | EndNote | Medlars | ProCite | Reference Manager | RefWorks
Send citation to:
BibTeX | RIS | EndNote | Medlars | ProCite | Reference Manager | RefWorks
Send citation to:
Gerami R, Salehi A. The Impact of New Technologies in Improving the Diagnostic Capability of Echocardiography in the Diagnosis of Fetal Heart Disorders; a Review Study. RJMS 2022; 29 (3) :221-229
URL: http://rjms.iums.ac.ir/article-1-8482-en.html
URL: http://rjms.iums.ac.ir/article-1-8482-en.html
Assistant Professor, Department of Radiology, Faculty of Medicine, AJA University of medical science, Tehran, Iran , Rezagerami64@gmail.com
Abstract: (112 Views)
Background & Aims: Artificial intelligence (AI) plays an important role in the development of echocardiography for fetal heart disorders. Artificial intelligence, especially deep learning, has shown significant capabilities in reducing the time required for echocardiographic examinations, increasing diagnostic accuracy, and helping to identify anatomical changes and abnormalities in the fetal heart. In the field of fetal heart and blood vessels, artificial intelligence promises to improve prenatal diagnosis of congenital heart disease. This offers the potential to improve screening processes that lead to early diagnosis and intervention in cases of fetal heart disorders. Smart diagnosis based on echocardiography, along with artificial intelligence techniques such as heart segmentation and identification of standard heart parts, helps in more effective and accurate diagnosis. The integration of artificial intelligence in the perinatal diagnosis of congenital heart disease shows its application in improving diagnostic accuracy with continuous efforts in research to further increase its effectiveness. Prenatal diagnosis of congenital heart disease (CHD) in medicine seems to be a solved problem, although challenges continue. Factors affecting fetal congenital heart diseases (CHDs) are diverse and available data in this field are limited. A study of fetal circulatory physiology and brain development in individuals with fetal congenital heart disease provides valuable insights. Advances in prenatal management and intervention for congenital heart disease are the subject of ongoing research that discusses current knowledge, implications, and challenges. Additionally, ongoing investigations such as blood tests to detect dangerous fetal heart defects before birth show promising advances in diagnostic techniques. Over the past decade, there have been significant advances in the prenatal diagnosis of congenital heart defects. While the rate of prenatal diagnosis has increased significantly, some malformations with 3 abnormal vessels are challenging to identify prenatally. Advances in prenatal diagnostic techniques, such as fetal echocardiography, have played an important role in increasing the accuracy of assessing structural heart lesions and dysrhythmic mechanisms. The use of fetal echocardiography has contributed to the growing trend of prenatal diagnosis of congenital heart disease and highlights the impact of evolving diagnostic technologies. The majority of defects identified in fetal life are atrial and ventricular septal defects, and advances continue to address challenges in detecting minor defects. An analysis of the types and trends of prenatally diagnosed fetal heart disorders in the last decade provides insights into the prevalence and characteristics of different types of fetal heart disorders. The purpose of this review study is to evaluate how new technologies can improve the ability of echocardiography to diagnose fetal heart defects.
Methods: In order to thoroughly examine the effects of new technologies on the diagnostic capacities of fetal echocardiography, a full narrative review was conducted using a systematic methodology. We conducted an extensive literature search using well-known academic databases such as Web of Science, ScienceDirect, Scopus, Springer, and Google Scholar. The search approach included targeted keywords pertaining to fetal echocardiography, cutting-edge technology, and enhancements in diagnostics. In order to promote inclusion, we conducted a systematic search of national databases such as the Scientific Information Database (SID), NoorMags, Magiran, and the Islamic World Science Citation Database (ISC) to identify relevant works. The search criteria were limited to papers published until January 2023, encompassing both English and Persian language articles.
Results: In the field of fetal echocardiography, machine learning (ML) brings significant improvements through its application in automated measurements. ML algorithms are effective in automating the measurement of cardiac biometrics and provide accurate assessment of fetal heart structures such as heart chambers. This not only increases efficiency but also ensures accuracy and helps sonographers achieve reliable measurements. Beyond biometrics, ML plays an important role in quality control by evaluating fetal telemedicine audio-visual systems (FTAS) through score-based systems. In addition, ML helps assess the learning curves of sonographers and ensures the quality and consistency of fetal echocardiographic examinations. The versatility of ML programs is evident in fully automated fetal lung ultrasound analysis and shows its ability to deal with various aspects of fetal health monitoring. Additionally, ML is important in hemodynamic quantification, with integrated and automated tools that use ML algorithms to quantify clinically relevant parameters such as B-mode-based pressure and pulse-wave Doppler hemodynamics. These advances underscore the transformative impact of ML in increasing the accuracy, efficiency, and comprehensiveness of fetal echocardiography. Computerized examinations in fetal echocardiography have made significant progress through the integration of machine learning. Studies suggest deep learning-based computer systems for automated echocardiographic examination of the fetal heart. These systems use ML algorithms to predict standard fetal heart shapes, views, and sections, providing valuable insights into congenital heart defects. FetalNet, a deep learning model, improves the detection of congenital heart disease using computer-aided segmentation of standard heart views. In addition, artificial intelligence has shown potential in improving prenatal diagnosis of congenital heart disease and contributing to better prenatal care. The use of deep learning for real cardiac object detection demonstrates the powerful capabilities of computer-aided ML methods in fetal echocardiographic analysis. This investigation demonstrates that STIC, functioning as a dynamic 3D imaging method, enables the ongoing capture of volumetric data from the fetal heart, providing accurate and detailed pictures of cardiac structures and arteries. The incorporation of machine learning (ML) in fetal echocardiography improves the precision of biometric measures, since artificial intelligence systems are skilled at detecting congenital heart abnormalities using conventional images. Moreover, the application of automated assessments and deep learning displays their potential to carefully examine fetal cardiac systems. This integration of technology enables researchers and medical personnel to do more accurate and thorough assessments of fetal cardiac well-being.
Conclusion: The results clearly demonstrate that using modern technology in fetal echocardiography not only enhances diagnostic processes but also has a crucial impact on enhancing treatment and effectively managing fetal cardiac diseases. The integration of imaging technology and artificial intelligence has significant potential for improving diagnostic standards, therefore raising the overall quality of fetal care. The results emphasize the potential revolutionary influence of these technologies on the domain of fetal echocardiography.
Methods: In order to thoroughly examine the effects of new technologies on the diagnostic capacities of fetal echocardiography, a full narrative review was conducted using a systematic methodology. We conducted an extensive literature search using well-known academic databases such as Web of Science, ScienceDirect, Scopus, Springer, and Google Scholar. The search approach included targeted keywords pertaining to fetal echocardiography, cutting-edge technology, and enhancements in diagnostics. In order to promote inclusion, we conducted a systematic search of national databases such as the Scientific Information Database (SID), NoorMags, Magiran, and the Islamic World Science Citation Database (ISC) to identify relevant works. The search criteria were limited to papers published until January 2023, encompassing both English and Persian language articles.
Results: In the field of fetal echocardiography, machine learning (ML) brings significant improvements through its application in automated measurements. ML algorithms are effective in automating the measurement of cardiac biometrics and provide accurate assessment of fetal heart structures such as heart chambers. This not only increases efficiency but also ensures accuracy and helps sonographers achieve reliable measurements. Beyond biometrics, ML plays an important role in quality control by evaluating fetal telemedicine audio-visual systems (FTAS) through score-based systems. In addition, ML helps assess the learning curves of sonographers and ensures the quality and consistency of fetal echocardiographic examinations. The versatility of ML programs is evident in fully automated fetal lung ultrasound analysis and shows its ability to deal with various aspects of fetal health monitoring. Additionally, ML is important in hemodynamic quantification, with integrated and automated tools that use ML algorithms to quantify clinically relevant parameters such as B-mode-based pressure and pulse-wave Doppler hemodynamics. These advances underscore the transformative impact of ML in increasing the accuracy, efficiency, and comprehensiveness of fetal echocardiography. Computerized examinations in fetal echocardiography have made significant progress through the integration of machine learning. Studies suggest deep learning-based computer systems for automated echocardiographic examination of the fetal heart. These systems use ML algorithms to predict standard fetal heart shapes, views, and sections, providing valuable insights into congenital heart defects. FetalNet, a deep learning model, improves the detection of congenital heart disease using computer-aided segmentation of standard heart views. In addition, artificial intelligence has shown potential in improving prenatal diagnosis of congenital heart disease and contributing to better prenatal care. The use of deep learning for real cardiac object detection demonstrates the powerful capabilities of computer-aided ML methods in fetal echocardiographic analysis. This investigation demonstrates that STIC, functioning as a dynamic 3D imaging method, enables the ongoing capture of volumetric data from the fetal heart, providing accurate and detailed pictures of cardiac structures and arteries. The incorporation of machine learning (ML) in fetal echocardiography improves the precision of biometric measures, since artificial intelligence systems are skilled at detecting congenital heart abnormalities using conventional images. Moreover, the application of automated assessments and deep learning displays their potential to carefully examine fetal cardiac systems. This integration of technology enables researchers and medical personnel to do more accurate and thorough assessments of fetal cardiac well-being.
Conclusion: The results clearly demonstrate that using modern technology in fetal echocardiography not only enhances diagnostic processes but also has a crucial impact on enhancing treatment and effectively managing fetal cardiac diseases. The integration of imaging technology and artificial intelligence has significant potential for improving diagnostic standards, therefore raising the overall quality of fetal care. The results emphasize the potential revolutionary influence of these technologies on the domain of fetal echocardiography.
Keywords: Fetal Echocardiography, Fetal Heart Disorders, Machine Learning, Artificial Intelligence, Computerized Examinations
Type of Study: review article |
Subject:
Radiology
Send email to the article author
| Rights and permissions | |
 |
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License. |
