Antonela Zanuttini
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Antonela Zanuttini has a total of 34 co-authors. Here are the top co-authors by number of shared publications:
- Philippe Pibarot
Université Laval
7 shared publications - Amanda Spilkin
Department of Mechanical, Industrial and Aerospace Engineering, Concordia University, Montreal, QC, Canada
5 shared publications - Hanae Elmekki
Institute for Information Systems Engineering, Concordia University, Montreal, QC, Canada
5 shared publications - Jamal Bentahar
Khalifa University
5 shared publications - Lyes Kadem
Concordia University
5 shared publications - Wen-Fang Xie
Concordia University
5 shared publications
Latest Publications
- 1 October 2026
- Published by Wolters Kluwer Health in Circulation: Cardiovascular Imaging
- Vol. 19 (10) , e018670
Cardiac Imaging in Clinical Trials of Intervention in Aortic Valve Stenosis
Cardiac imaging and in particular transthoracic echocardiography and computed tomography play a major role in the selection of the patients for surgical or transcatheter aortic valve replacement, for the assessment or procedural success and early prosthetic valve hemodynamics following aortic valve replacement, and for the evaluation and follow-up of the prosthetic valve structure and function in the longer-term, which is key to demonstrate the valve durability. The purpose of this review article is thus to present the role of cardiac imaging, and particularly transthoracic echocardiography and computed tomography, in: (1) patient selection for intervention; (2) assessment of procedural and device success, and of intended performance of the valve; and (3) assessment of the long-term success, valve durability, and prognosis, for clinical trials of intervention in patients with aortic stenosis. Transthoracic echocardiography is the primary imaging modality to detect and stage bioprosthetic valve dysfunction. However, multimodality imaging, including transesophageal echocardiography and computed tomography, is often necessary to determine the cause of bioprosthetic valve dysfunction and make the differential diagnosis between prosthesis-patient mismatch, structural valve deterioration, thrombosis, pannus, or endocarditis. The clinical trials in the field of structural heart disease, and particularly in the field of aortic valve intervention, include imaging end points as part of the primary or key secondary end points. Standardized methods and definitions should be applied to adjudicate these imaging end points, and ideally, these trial end points should be analyzed by independent imaging core labs.
- 1 October 2026
- Published by Wolters Kluwer Health in Circulation: Cardiovascular Imaging
- Vol. 19 (10) , e018670
Cardiac Imaging in Clinical Trials of Intervention in Aortic Valve Stenosis
Cardiac imaging and in particular transthoracic echocardiography and computed tomography play a major role in the selection of the patients for surgical or transcatheter aortic valve replacement, for the assessment or procedural success and early prosthetic valve hemodynamics following aortic valve replacement, and for the evaluation and follow-up of the prosthetic valve structure and function in the longer-term, which is key to demonstrate the valve durability. The purpose of this review article is thus to present the role of cardiac imaging, and particularly transthoracic echocardiography and computed tomography, in: (1) patient selection for intervention; (2) assessment of procedural and device success, and of intended performance of the valve; and (3) assessment of the long-term success, valve durability, and prognosis, for clinical trials of intervention in patients with aortic stenosis. Transthoracic echocardiography is the primary imaging modality to detect and stage bioprosthetic valve dysfunction. However, multimodality imaging, including transesophageal echocardiography and computed tomography, is often necessary to determine the cause of bioprosthetic valve dysfunction and make the differential diagnosis between prosthesis-patient mismatch, structural valve deterioration, thrombosis, pannus, or endocarditis. The clinical trials in the field of structural heart disease, and particularly in the field of aortic valve intervention, include imaging end points as part of the primary or key secondary end points. Standardized methods and definitions should be applied to adjudicate these imaging end points, and ideally, these trial end points should be analyzed by independent imaging core labs.
- Open Access
- 6 January 2026
- Published by Wolters Kluwer Health in Journal of the American Heart Association
- Vol. 15 (1) , e047196
Echocardiographic Results of Transcatheter Versus Surgical Aortic Valve Replacement in Women With Severe Aortic Stenosis: The RHEIA Trial
Background: In the RHEIA (Randomized Research in Women All Comers With Aortic Stenosis) trial, the incidence of the primary end point of death, stroke, or rehospitalization at 1 year was lower with transcatheter aortic valve implantation (TAVI) than with surgical aortic valve replacement. The objective of this substudy was to compare echocardiographic findings in women with severe aortic stenosis following surgical aortic valve replacement or TAVI. Methods: At 48 European centers, 443 women underwent randomization 1:1, and 420 were treated as randomized. Echocardiograms were available in 356 patients and were analyzed by a core laboratory. Results: Rates of or greater moderate paravalvular regurgitation was low (P =1) was low and not different between groups. Valve hemodynamics were stable at 1 year. The rate of residual left ventricular hypertrophy (45.3 versus 28.6%; P =0.004) at 1 year was significantly higher with TAVI, whereas the rate of right ventricular systolic dysfunction (14.5 versus 40.7%; P P =0.001) were better with TAVI. Conclusions: Among women with severe aortic stenosis, both TAVI and surgical aortic valve replacement achieve excellent valve hemodynamic results with low and similar rates of moderate or greater paravalvular regurgitation or severe patient–prosthesis mismatch. Surgical aortic valve replacement was associated with lower gradients and more pronounced regression of left ventricular hypertrophy, whereas TAVI was associated with better right ventricular systolic function and evolution of cardiac damage stage. REGISTRATION: URL: https://clinicaltrials.gov/study/NCT04160130 ; Unique Identifier: NCT04160130.
- 6 January 2026
- Published by Wolters Kluwer Health in Journal of the American Heart Association
- Vol. 15 (1) , e047196
Open AccessEchocardiographic Results of Transcatheter Versus Surgical Aortic Valve Replacement in Women With Severe Aortic Stenosis: The RHEIA Trial
Background: In the RHEIA (Randomized Research in Women All Comers With Aortic Stenosis) trial, the incidence of the primary end point of death, stroke, or rehospitalization at 1 year was lower with transcatheter aortic valve implantation (TAVI) than with surgical aortic valve replacement. The objective of this substudy was to compare echocardiographic findings in women with severe aortic stenosis following surgical aortic valve replacement or TAVI. Methods: At 48 European centers, 443 women underwent randomization 1:1, and 420 were treated as randomized. Echocardiograms were available in 356 patients and were analyzed by a core laboratory. Results: Rates of or greater moderate paravalvular regurgitation was low (P =1) was low and not different between groups. Valve hemodynamics were stable at 1 year. The rate of residual left ventricular hypertrophy (45.3 versus 28.6%; P =0.004) at 1 year was significantly higher with TAVI, whereas the rate of right ventricular systolic dysfunction (14.5 versus 40.7%; P P =0.001) were better with TAVI. Conclusions: Among women with severe aortic stenosis, both TAVI and surgical aortic valve replacement achieve excellent valve hemodynamic results with low and similar rates of moderate or greater paravalvular regurgitation or severe patient–prosthesis mismatch. Surgical aortic valve replacement was associated with lower gradients and more pronounced regression of left ventricular hypertrophy, whereas TAVI was associated with better right ventricular systolic function and evolution of cardiac damage stage. REGISTRATION: URL: https://clinicaltrials.gov/study/NCT04160130 ; Unique Identifier: NCT04160130.
- 22 December 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Medical Robotics and Bionics
- Vol. 8 (1) , 151-166
Deep Reinforcement Learning-Based Ultrasound Visual Servoing Scheme for Autonomous Robotic Echocardiography
This study develops a deep reinforcement learning (DRL)-based ultrasound visual servoing (DBUVS) scheme for an autonomous robotic echocardiography system. The proposed approach aims to replace remote cardiac sonographers with a newly developed AI agent, Cardiac Sonographer Net (CS-Net), trained using DRL based on the Soft Actor-Critic (SAC) algorithm. To train CS-Net, a robotic echocardiography environment is simulated in GAZEBO simulator using a two-stage generative AI (GenAI) approach to produce highfidelity synthetic ultrasound images. In the first stage, multiple convolutional neural networks (CNNs) generate initial ultrasound images based on different parameter settings from a given probe pose. A fuzzy inference system (FIS) then fuses these images into a single low-quality representation. In the second stage, a super-resolution generative adversarial network with gradient penalty (SRGAN-GP) enhances image quality. Compared with low-quality images, the GenAI-based outputs show an 11.37 perceptual image patch similarity (LPIPS) and higher resolution (256W256 to 500W500 pixels), closely matching real ultrasound images. CS-Net is initially trained in the simulation environment and deployed on the real experimental robotic system with an ultrasound probe mounted on the end effector and a cardiac phantom for testing, using sim-to-real transfer learning. Experimental results demonstrate that the robotic echocardiography system powered by CS-Net performs autonomous scanning with higher accuracy and efficiency than the echocardiography remotely operated by the sonographer. Specifically, the system achieves faster convergence by reaching an image feature error norm of 0.176 in 25 seconds, compared with 0.253 in 50 seconds for remote operation.
- 22 December 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Medical Robotics and Bionics
- Vol. 8 (1) , 151-166
Deep Reinforcement Learning-Based Ultrasound Visual Servoing Scheme for Autonomous Robotic Echocardiography
This study develops a deep reinforcement learning (DRL)-based ultrasound visual servoing (DBUVS) scheme for an autonomous robotic echocardiography system. The proposed approach aims to replace remote cardiac sonographers with a newly developed AI agent, Cardiac Sonographer Net (CS-Net), trained using DRL based on the Soft Actor-Critic (SAC) algorithm. To train CS-Net, a robotic echocardiography environment is simulated in GAZEBO simulator using a two-stage generative AI (GenAI) approach to produce highfidelity synthetic ultrasound images. In the first stage, multiple convolutional neural networks (CNNs) generate initial ultrasound images based on different parameter settings from a given probe pose. A fuzzy inference system (FIS) then fuses these images into a single low-quality representation. In the second stage, a super-resolution generative adversarial network with gradient penalty (SRGAN-GP) enhances image quality. Compared with low-quality images, the GenAI-based outputs show an 11.37 perceptual image patch similarity (LPIPS) and higher resolution (256W256 to 500W500 pixels), closely matching real ultrasound images. CS-Net is initially trained in the simulation environment and deployed on the real experimental robotic system with an ultrasound probe mounted on the end effector and a cardiac phantom for testing, using sim-to-real transfer learning. Experimental results demonstrate that the robotic echocardiography system powered by CS-Net performs autonomous scanning with higher accuracy and efficiency than the echocardiography remotely operated by the sonographer. Specifically, the system achieves faster convergence by reaching an image feature error norm of 0.176 in 25 seconds, compared with 0.253 in 50 seconds for remote operation.
- Open Access
- 18 March 2025
- Published by Elsevier in Computers in Biology and Medicine
- Vol. 190, 110003
CACTUS: An open dataset and framework for automated Cardiac Assessment and Classification of Ultrasound images using deep transfer learning
- 18 March 2025
- Published by Elsevier in Computers in Biology and Medicine
- Vol. 190, 110003
Open AccessCACTUS: An open dataset and framework for automated Cardiac Assessment and Classification of Ultrasound images using deep transfer learning
- 11 February 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE/ASME Transactions on Mechatronics
- Vol. 30 (6) , 6407-6419
Robust Deep Feature Ultrasound Image-Based Visual Servoing: Focus on Cardiac Examination
This article introduces a robust deep feature ultrasound image-based visual servoing (UIBVS) technique for an ultrasound robot focusing on automatic cardiac examination. To this end, a convolutional neural network named ultrasound-cardiac-feature-net (UCF-Net) is developed, which is trained in a supervised manner to process ultrasound images and generate a set of six image features referred to as deep ultrasound image features. To enhance the robustness of UCF-Net against the variables that affect the ultrasound image quality, such as interaction normal force, scan depth, dynamic range, power, and gain, several datasets with different sets of parameters are gathered for training. Deep ultrasound image features enable an eye-in-hand robot to interact with the human body through UIBVS. To implement UIBVS, a filtered integral quasi-super-twisting algorithm (FIQSTA) is synthesized as the primary controller. Interaction force control is also considered within a hybrid vision/force control framework, providing compliance with the body and increasing the safety of the interaction. The proof of the robustness and stability of FIQSTA is also investigated. Experimental results on a cardiac phantom for four main views, i.e., parasternal short axis, parasternal long axis, subcostal, and apical four chambers views, and a trajectory passing through the main views demonstrate the feasibility of the proposed method for cardiac examination and the superior performance of the main controller to other well-known methods, including proportional (P) controller, sliding mode controller, super-twisting algorithm (STA), and integral quasi-STA.
- 11 February 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE/ASME Transactions on Mechatronics
- Vol. 30 (6) , 6407-6419
Robust Deep Feature Ultrasound Image-Based Visual Servoing: Focus on Cardiac Examination
This article introduces a robust deep feature ultrasound image-based visual servoing (UIBVS) technique for an ultrasound robot focusing on automatic cardiac examination. To this end, a convolutional neural network named ultrasound-cardiac-feature-net (UCF-Net) is developed, which is trained in a supervised manner to process ultrasound images and generate a set of six image features referred to as deep ultrasound image features. To enhance the robustness of UCF-Net against the variables that affect the ultrasound image quality, such as interaction normal force, scan depth, dynamic range, power, and gain, several datasets with different sets of parameters are gathered for training. Deep ultrasound image features enable an eye-in-hand robot to interact with the human body through UIBVS. To implement UIBVS, a filtered integral quasi-super-twisting algorithm (FIQSTA) is synthesized as the primary controller. Interaction force control is also considered within a hybrid vision/force control framework, providing compliance with the body and increasing the safety of the interaction. The proof of the robustness and stability of FIQSTA is also investigated. Experimental results on a cardiac phantom for four main views, i.e., parasternal short axis, parasternal long axis, subcostal, and apical four chambers views, and a trajectory passing through the main views demonstrate the feasibility of the proposed method for cardiac examination and the superior performance of the main controller to other well-known methods, including proportional (P) controller, sliding mode controller, super-twisting algorithm (STA), and integral quasi-STA.
- 17 September 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Industrial Electronics
- Vol. 72 (4) , 3972-3983
AI-Powered Robust Interaction Force Control of a Cardiac Ultrasound Robotic System
This article introduces a novel intelligent robust interaction force control method for a cardiac ultrasound robotic system (CURS), exploiting dual control loops and artificial intelligence (AI)-driven image feedback to enhance both image quality and patient safety during cardiac examinations. Unlike existing systems that use a constant interaction force, the proposed method adjusts the force based on ultrasound image feedback, which is critical for adapting to different cardiac views. The system employs an internal control loop, where the force feedback generates control commands (low-level controller), and an external control loop, where the feedback is processed through a convolutional neural network (CNN), named ultrasound-cardiac-feature-net (UCF-Net), determines the optimal force values (high-level controller). An adaptive filtered quasi-sliding mode controller (AFQSMC) manages both interaction force and probe’s position within a hybrid position/force control context, ensuring robustness against uncertainties and disturbances. Experimental evaluations on a cardiac phantom navigating main cardiac views demonstrate the superiority of the proposed approach over traditional constant force control. Moreover, AFQSMC achieves significant improvements in interaction force control, with enhancements ranging from 21.87% to 68.25% over traditional FQSMC, sliding mode control (SMC), and proportional-integral (PI) controllers, across quantitative metrics such as root mean square (RMS), standard deviation (STD), and Max, confirming its potential for improving cardiac examination performance.
- 17 September 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Industrial Electronics
- Vol. 72 (4) , 3972-3983
AI-Powered Robust Interaction Force Control of a Cardiac Ultrasound Robotic System
This article introduces a novel intelligent robust interaction force control method for a cardiac ultrasound robotic system (CURS), exploiting dual control loops and artificial intelligence (AI)-driven image feedback to enhance both image quality and patient safety during cardiac examinations. Unlike existing systems that use a constant interaction force, the proposed method adjusts the force based on ultrasound image feedback, which is critical for adapting to different cardiac views. The system employs an internal control loop, where the force feedback generates control commands (low-level controller), and an external control loop, where the feedback is processed through a convolutional neural network (CNN), named ultrasound-cardiac-feature-net (UCF-Net), determines the optimal force values (high-level controller). An adaptive filtered quasi-sliding mode controller (AFQSMC) manages both interaction force and probe’s position within a hybrid position/force control context, ensuring robustness against uncertainties and disturbances. Experimental evaluations on a cardiac phantom navigating main cardiac views demonstrate the superiority of the proposed approach over traditional constant force control. Moreover, AFQSMC achieves significant improvements in interaction force control, with enhancements ranging from 21.87% to 68.25% over traditional FQSMC, sliding mode control (SMC), and proportional-integral (PI) controllers, across quantitative metrics such as root mean square (RMS), standard deviation (STD), and Max, confirming its potential for improving cardiac examination performance.