Motion Estimation by Deep Learning in 2D Echocardiography: Synthetic Dataset and Validation

Evain, Ewan; Sun, Yueming; Faraz, Khuram; Garcia, Damien; Saloux, Éric; Gerber, Bernhard; Craene, Mathieu De; Bernard, Olivier

doi:10.1109/tmi.2022.3151606

Cited by 13 publications

(3 citation statements)

References 40 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The dataset is a substantial element of every DL-based intrusion detection scheme [ 52 ]. Selecting an effective and proportionate dataset significantly strengthens the IDS.…”

Section: Methodsmentioning

confidence: 99%

Fog-Assisted Deep-Learning-Empowered Intrusion Detection System for RPL-Based Resource-Constrained Smart Industries

Attique

Wang

2022

Sensors

View full text Add to dashboard Cite

The Internet of Things (IoT) is a prominent and advanced network communication technology that has familiarized the world with smart industries. The conveniently acquirable nature of IoT makes it susceptible to a diversified range of potential security threats. The literature has brought forth a plethora of solutions for ensuring secure communications in IoT-based smart industries. However, resource-constrained sectors still demand significant attention. We have proposed a fog-assisted deep learning (DL)-empowered intrusion detection system (IDS) for resource-constrained smart industries. The proposed Cuda–deep neural network gated recurrent unit (Cu-DNNGRU) framework was trained on the N-BaIoT dataset and was evaluated on judicious performance metrics, including accuracy, precision, recall, and F1-score. Additionally, the Cu-DNNGRU was empirically investigated alongside state-of-the-art classifiers, including Cu-LSTMDNN, Cu-BLSTM, and Cu-GRU. An extensive performance comparison was also undertaken among the proposed IDS and some outstanding solutions from the literature. The simulation results showed ample strength with respect to the validation of the proposed framework. The proposed Cu-DNNGRU achieved 99.39% accuracy, 99.09% precision, 98.89% recall, and an F1-score of 99.21%. In the performance comparison, the values were substantially higher than those of the benchmarked schemes, as well as competitive security solutions from the literature.

show abstract

“…The dataset is a substantial element of every DL-based intrusion detection scheme [ 52 ]. Selecting an effective and proportionate dataset significantly strengthens the IDS.…”

Section: Methodsmentioning

confidence: 99%

Fog-Assisted Deep-Learning-Empowered Intrusion Detection System for RPL-Based Resource-Constrained Smart Industries

Attique

Wang

2022

Sensors

View full text Add to dashboard Cite

show abstract

“…Evain et al [44] developed the EchoPWC-Net model for motion estimation using echocardiography. A Convolutional Neural Sub-Network (CNSN) estimator and normalized crosscorrelations from 2DE images are used by the model to forecast dense displacement maps.…”

Section: B Survey On DL Interpretation Of Echocardiography For Cardia...mentioning

confidence: 99%

Predicting Cardiac Issues From Echocardiograms: A Literature Review Using Deep Learning and Machine Learning Techniques

Valanrani

2024

ijarcs

View full text Add to dashboard Cite

Cardiovascular disease (CVD) has a substantial impact on overall health, well-being, and life-expectancy. Echocardiography is a widely used imaging technique in cardiovascular medicine, utilizing various medical imaging technology to visualize heart chambers and valve’s motion activity. In order to diagnose and treat complicated cardiovascular problems, it takes high-resolution images of the heart and its surroundings. However, it has limitations such as long procedure times, multiple measurement values, complex analyses, individualized assessments, operator subjectivity, and wide observation ranges. This makes it challenging for sonographers to accurately detect and diagnose heart diseases. In recent days, Deep Learning (DL) is increasingly used in clinical computer-assisted systems for disease detection, feature segmentation, functional evaluation, and diagnosis. It is an alternate technique for accurate detection and treatment of cardiovascular disorders; it improves the diagnostic capacities of echocardiography by identifying pathological conditions, extracting anatomically significant data, measuring cardio-motion, and calculating echo image quality. This paper presents a detailed review of various DL frameworks developed to analyse different cardiac views using echocardiography for improving the prediction and diagnosis of CVD. At the outset, a variety of echocardiography systems linked with DL-based segmentation and classification are reviewed briefly. Afterwards, a comparison research is carried out to gain insight into the shortcomings of those algorithms and provide a fresh approach to improve the accuracy of cardiac view categorization in echocardiography systems.

show abstract

“…Evain E. et al [ 36 ] developed a new optical flow estimation algorithm based on PWC-Net. The results showed that the algorithm accurately assessed GLS in the echocardiography sequences.…”

Section: Ai’s Application In Left Ventricular Systolic Function—glsmentioning

confidence: 99%

Artificial Intelligence-Enhanced Echocardiography for Systolic Function Assessment

Zhang

Zhu

Liu

et al. 2022

JCM

View full text Add to dashboard Cite

The accurate assessment of left ventricular systolic function is crucial in the diagnosis and treatment of cardiovascular diseases. Left ventricular ejection fraction (LVEF) and global longitudinal strain (GLS) are the most critical indexes of cardiac systolic function. Echocardiography has become the mainstay of cardiac imaging for measuring LVEF and GLS because it is non-invasive, radiation-free, and allows for bedside operation and real-time processing. However, the human assessment of cardiac function depends on the sonographer’s experience, and despite their years of training, inter-observer variability exists. In addition, GLS requires post-processing, which is time consuming and shows variability across different devices. Researchers have turned to artificial intelligence (AI) to address these challenges. The powerful learning capabilities of AI enable feature extraction, which helps to achieve accurate identification of cardiac structures and reliable estimation of the ventricular volume and myocardial motion. Hence, the automatic output of systolic function indexes can be achieved based on echocardiographic images. This review attempts to thoroughly explain the latest progress of AI in assessing left ventricular systolic function and differential diagnosis of heart diseases by echocardiography and discusses the challenges and promises of this new field.

show abstract

Motion Estimation by Deep Learning in 2D Echocardiography: Synthetic Dataset and Validation

Cited by 13 publications

References 40 publications

Fog-Assisted Deep-Learning-Empowered Intrusion Detection System for RPL-Based Resource-Constrained Smart Industries

Fog-Assisted Deep-Learning-Empowered Intrusion Detection System for RPL-Based Resource-Constrained Smart Industries

Predicting Cardiac Issues From Echocardiograms: A Literature Review Using Deep Learning and Machine Learning Techniques

Artificial Intelligence-Enhanced Echocardiography for Systolic Function Assessment

Contact Info

Product

Resources

About