Majid Roshanfar
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Majid Roshanfar has a total of 32 co-authors. Here are the top co-authors by number of shared publications:
- Amir Hooshiar
McGill University
5 shared publications - Bobak Mosadegh
Dalio Institute of Cardiovascular Imaging, Department of Radiology, Weill Cornell Medicine, New York, USA
4 shared publications - Javad Dargahi
Surgical Robotics Laboratory (SRL), Department of Mechanical Engineering, Gina Cody School of Engineering, Concordia University, Montreal, QC H3G 1M8, Canada
4 shared publications - Albert J. Sinusas
Yale University School of Medicine
3 shared publications - Mohammadhossein Salimi
Department of Mechanical Engineering, York University, Toronto, ON M3J 1P3, Canada
3 shared publications - Sun-Joo Jang
Section of Cardiovascular Medicine, Department of Medicine, Yale University School of Medicine, New Haven, CT 06519, USA
3 shared publications
Latest Publications
- Open Access
Neural Minimum-Distance Estimation for Collision-Aware Operation of Multi-Arm Laparoscopy Surgical Robots Through Learning-from-Simulation
This study presents an integrated framework for enhancing the safety and operational efficiency of robotic arms in laparoscopic surgery by addressing minimum distance estimation between multi-arm manipulators and the associated collision-aware warning. By combining analytical modeling, real time simulation, and machine learning, the framework offers a robust solution for ensuring safe robotic operations. An analytical model was developed to estimate the minimum distances between robotic arms based on their joint configurations, offering theoretical calculations that serve as both a validation tool and a benchmark. To complement this, a 3D simulation environment was created to model two 7 DOF Kinova robotic arms (Kinova Inc., Boisbriand, QC, Canada), generating a diverse dataset of configurations for distance estimation and collision warning. Using these insights, a deep residual neural network model was trained with joint configurations as inputs. On the held out validation set, the model achieves , RMSE mm, MAE mm, and a near zero mean bias, demonstrating strong predictive accuracy and consistent generalization across the workspace. The framework is intended as an early collision warning layer, where a warning is triggered when the predicted inter-arm distance falls below a 0.2 m threshold, which corresponds to a surface to surface clearance of approximately 50 mm given the Kinova Gen3 (Kinova Inc., Boisbriand, QC, Canada) cross sectional radius. This work demonstrates the effectiveness of combining analytical modeling with machine learning to enhance the precision and reliability of multi-arm robotic systems.Open AccessNeural Minimum-Distance Estimation for Collision-Aware Operation of Multi-Arm Laparoscopy Surgical Robots Through Learning-from-Simulation
This study presents an integrated framework for enhancing the safety and operational efficiency of robotic arms in laparoscopic surgery by addressing minimum distance estimation between multi-arm manipulators and the associated collision-aware warning. By combining analytical modeling, real time simulation, and machine learning, the framework offers a robust solution for ensuring safe robotic operations. An analytical model was developed to estimate the minimum distances between robotic arms based on their joint configurations, offering theoretical calculations that serve as both a validation tool and a benchmark. To complement this, a 3D simulation environment was created to model two 7 DOF Kinova robotic arms (Kinova Inc., Boisbriand, QC, Canada), generating a diverse dataset of configurations for distance estimation and collision warning. Using these insights, a deep residual neural network model was trained with joint configurations as inputs. On the held out validation set, the model achieves , RMSE mm, MAE mm, and a near zero mean bias, demonstrating strong predictive accuracy and consistent generalization across the workspace. The framework is intended as an early collision warning layer, where a warning is triggered when the predicted inter-arm distance falls below a 0.2 m threshold, which corresponds to a surface to surface clearance of approximately 50 mm given the Kinova Gen3 (Kinova Inc., Boisbriand, QC, Canada) cross sectional radius. This work demonstrates the effectiveness of combining analytical modeling with machine learning to enhance the precision and reliability of multi-arm robotic systems. - 4 February 2026
- Published by Elsevier in Mechatronics
- Vol. 116
Learning-based modeling of a magnetically steerable soft suction device for endoscopic endonasal interventions
- 4 February 2026
- Published by Elsevier in Mechatronics
- Vol. 116
Learning-based modeling of a magnetically steerable soft suction device for endoscopic endonasal interventions
- Open Access
Benchmarking Robust AI for Microrobot Detection with Ultrasound Imaging
Microrobots are emerging as transformative tools in minimally invasive medicine, with applications in non-invasive therapy, real-time diagnosis, and targeted drug delivery. Effective use of these systems critically depends on accurate detection and tracking of microrobots within the body. Among commonly used imaging modalities, including MRI, CT, and optical imaging, ultrasound (US) offers an advantageous balance of portability, low cost, non-ionizing safety, and high temporal resolution, making it particularly suitable for real-time microrobot monitoring. This study reviews current detection strategies and presents a comparative evaluation of six advanced AI-based multi-object detectors, including ConvNeXt, Res2NeXt-101, ResNeSt-269, U-Net, and the latest YOLO variants (v11, v12), being applied to microrobot detection in US imaging. Performance is assessed using standard metrics (AP50–95, precision, recall, F1-score) and robustness to four visual perturbations: blur, brightness variation, occlusion, and speckle noise. Additionally, feature-level sensitivity analyses are conducted to identify the contributions of different visual cues. Computational efficiency is also measured to assess suitability for real-time deployment. Results show that ResNeSt-269 achieved the highest detection accuracy, followed by Res2NeXt-101 and ConvNeXt, while YOLO-based detectors provided superior computational efficiency. These findings offer actionable insights for developing robust and efficient microrobot tracking systems with strong potential in diagnostic and therapeutic healthcare applications.
Open AccessBenchmarking Robust AI for Microrobot Detection with Ultrasound Imaging
Microrobots are emerging as transformative tools in minimally invasive medicine, with applications in non-invasive therapy, real-time diagnosis, and targeted drug delivery. Effective use of these systems critically depends on accurate detection and tracking of microrobots within the body. Among commonly used imaging modalities, including MRI, CT, and optical imaging, ultrasound (US) offers an advantageous balance of portability, low cost, non-ionizing safety, and high temporal resolution, making it particularly suitable for real-time microrobot monitoring. This study reviews current detection strategies and presents a comparative evaluation of six advanced AI-based multi-object detectors, including ConvNeXt, Res2NeXt-101, ResNeSt-269, U-Net, and the latest YOLO variants (v11, v12), being applied to microrobot detection in US imaging. Performance is assessed using standard metrics (AP50–95, precision, recall, F1-score) and robustness to four visual perturbations: blur, brightness variation, occlusion, and speckle noise. Additionally, feature-level sensitivity analyses are conducted to identify the contributions of different visual cues. Computational efficiency is also measured to assess suitability for real-time deployment. Results show that ResNeSt-269 achieved the highest detection accuracy, followed by Res2NeXt-101 and ConvNeXt, while YOLO-based detectors provided superior computational efficiency. These findings offer actionable insights for developing robust and efficient microrobot tracking systems with strong potential in diagnostic and therapeutic healthcare applications.
- 12 September 2025
- Published by Springer Nature in Acta Mechanica
- Vol. 236  (12) , 6957-6973
Random vibration analysis of pieozoelectric nanobeams using a unified nonlocal shear deformation beam theory
- 12 September 2025
- Published by Springer Nature in Acta Mechanica
- Vol. 236  (12) , 6957-6973
Random vibration analysis of pieozoelectric nanobeams using a unified nonlocal shear deformation beam theory
- Open Access
- 2 May 2025
- Published by MDPI AG in Bioengineering
- Vol. 12  (5) , 488
Emerging Image-Guided Navigation Techniques for Cardiovascular Interventions: A Scoping Review
Background: Image-guided navigation has revolutionized precision cardiac interventions, yet current technologies face critical limitations in real-time guidance and procedural accuracy. Method: Here, we comprehensively evaluate state-of-the-art imaging modalities, from conventional fluoroscopy to emerging hybrid systems, analyzing their applications across coronary, structural, and electrophysiological interventions. Results: We demonstrate that novel approaches combining optical coherence tomography with near-infrared spectroscopy or fluorescence achieve unprecedented plaque characterization and procedural guidance through simultaneous structural and molecular imaging. Our analysis reveals key challenges, including imaging artifacts and resolution constraints, while highlighting recent technological solutions incorporating artificial intelligence and robotics. We show that non-imaging alternatives, such as fiber optic real-shape sensing and electromagnetic tracking, complement traditional techniques by providing real-time navigation without radiation exposure. This paper also discusses the integration of image-guided navigation techniques into augmented reality systems and patient-specific modeling, highlighting initial clinical studies that demonstrate their significant promise in reducing procedural times and improving accuracy. These findings establish a framework for next-generation cardiac interventions, emphasizing the critical role of multimodal imaging platforms enhanced by AI-driven decision support. Conclusions: We conclude that continued innovation in hybrid imaging systems, coupled with advances in automation, will be essential for optimizing procedural outcomes and expanding access to complex cardiac interventions.
- 2 May 2025
- Published by MDPI AG in Bioengineering
- Vol. 12  (5) , 488
Open AccessEmerging Image-Guided Navigation Techniques for Cardiovascular Interventions: A Scoping Review
Background: Image-guided navigation has revolutionized precision cardiac interventions, yet current technologies face critical limitations in real-time guidance and procedural accuracy. Method: Here, we comprehensively evaluate state-of-the-art imaging modalities, from conventional fluoroscopy to emerging hybrid systems, analyzing their applications across coronary, structural, and electrophysiological interventions. Results: We demonstrate that novel approaches combining optical coherence tomography with near-infrared spectroscopy or fluorescence achieve unprecedented plaque characterization and procedural guidance through simultaneous structural and molecular imaging. Our analysis reveals key challenges, including imaging artifacts and resolution constraints, while highlighting recent technological solutions incorporating artificial intelligence and robotics. We show that non-imaging alternatives, such as fiber optic real-shape sensing and electromagnetic tracking, complement traditional techniques by providing real-time navigation without radiation exposure. This paper also discusses the integration of image-guided navigation techniques into augmented reality systems and patient-specific modeling, highlighting initial clinical studies that demonstrate their significant promise in reducing procedural times and improving accuracy. These findings establish a framework for next-generation cardiac interventions, emphasizing the critical role of multimodal imaging platforms enhanced by AI-driven decision support. Conclusions: We conclude that continued innovation in hybrid imaging systems, coupled with advances in automation, will be essential for optimizing procedural outcomes and expanding access to complex cardiac interventions.
- Open Access
- 22 March 2025
- Published by MDPI AG in Micromachines
- Vol. 16  (4) , 363
Advanced Robotics for the Next-Generation of Cardiac Interventions
With an increasing number of elderly individuals, the demand for advanced technologies to treat cardiac diseases has become more critical than ever. Additionally, there is a pressing need to reduce the learning curve for cardiac interventionalists to keep pace with the rapid development of new types of procedures and devices and to expand the adoption of established procedures in more hospitals. This comprehensive review aims to shed light on recent advancements in novel robotic systems for cardiac interventions. To do so, this review provides a brief overview of the history of previously developed robotic systems and describes the necessity for advanced technologies for cardiac interventions to address the technological limitations of current systems. Moreover, this review explores the potential of cutting-edge technologies and methods in developing the next generation of intra-procedure autonomous navigation. Each highlighted topic undergoes a critical analysis to evaluate its technical limitations and the challenges that must be addressed for successful clinical implementation.
- 22 March 2025
- Published by MDPI AG in Micromachines
- Vol. 16  (4) , 363
Open AccessAdvanced Robotics for the Next-Generation of Cardiac Interventions
With an increasing number of elderly individuals, the demand for advanced technologies to treat cardiac diseases has become more critical than ever. Additionally, there is a pressing need to reduce the learning curve for cardiac interventionalists to keep pace with the rapid development of new types of procedures and devices and to expand the adoption of established procedures in more hospitals. This comprehensive review aims to shed light on recent advancements in novel robotic systems for cardiac interventions. To do so, this review provides a brief overview of the history of previously developed robotic systems and describes the necessity for advanced technologies for cardiac interventions to address the technological limitations of current systems. Moreover, this review explores the potential of cutting-edge technologies and methods in developing the next generation of intra-procedure autonomous navigation. Each highlighted topic undergoes a critical analysis to evaluate its technical limitations and the challenges that must be addressed for successful clinical implementation.
- Open Access
Addressing Peri-Device Leaks in Next-Generation Transcatheter Left Atrial Appendage Occluders: An Open Question
With FDA-approved devices, left atrial appendage (LAA) occlusion has emerged as a well-established and rapidly growing approach to stroke prevention in patients with non-valvular atrial fibrillation. These devices are indicated for use in patients who are at increased risk of stroke and systemic embolism, as determined by CHA2DS2-VASc scores, and are suitable for anticoagulation therapy, with an appropriate rationale for seeking a non-pharmacologic alternative. This includes patients who may be unsuitable for long-term anticoagulation due to contra-indications. These devices, generally consisting of a nitinol-framed structure with a circular cross-section, are positioned within the LAA to obstruct the ostium, effectively preventing the thrombus from embolizing the brain. The initial clinical data from pivotal trials and observational registries indicated no strong correlation between peri-device leaks (PDLs) and adverse events. However, recent studies have shown that PDLs are associated with a higher risk of thrombo-embolic events, leading to renewed interest in managing PDLs. This paper reviews the occurrence of PDLs after percutaneous LAA occlusion using current FDA-approved devices, highlighting the need for non-circular occluders to better-accommodate the inherent variability in LAA anatomy. It also compares the benefits and limitations of emerging approaches still under investigation, focusing on addressing PDLs.
Open AccessAddressing Peri-Device Leaks in Next-Generation Transcatheter Left Atrial Appendage Occluders: An Open Question
With FDA-approved devices, left atrial appendage (LAA) occlusion has emerged as a well-established and rapidly growing approach to stroke prevention in patients with non-valvular atrial fibrillation. These devices are indicated for use in patients who are at increased risk of stroke and systemic embolism, as determined by CHA2DS2-VASc scores, and are suitable for anticoagulation therapy, with an appropriate rationale for seeking a non-pharmacologic alternative. This includes patients who may be unsuitable for long-term anticoagulation due to contra-indications. These devices, generally consisting of a nitinol-framed structure with a circular cross-section, are positioned within the LAA to obstruct the ostium, effectively preventing the thrombus from embolizing the brain. The initial clinical data from pivotal trials and observational registries indicated no strong correlation between peri-device leaks (PDLs) and adverse events. However, recent studies have shown that PDLs are associated with a higher risk of thrombo-embolic events, leading to renewed interest in managing PDLs. This paper reviews the occurrence of PDLs after percutaneous LAA occlusion using current FDA-approved devices, highlighting the need for non-circular occluders to better-accommodate the inherent variability in LAA anatomy. It also compares the benefits and limitations of emerging approaches still under investigation, focusing on addressing PDLs.
- Open Access
- 21 January 2024
- Published by MDPI AG in Biomimetics
- Vol. 9  (1) , 59
Design Optimization of a Hybrid-Driven Soft Surgical Robot with Biomimetic Constraints
The current study investigated the geometry optimization of a hybrid-driven (based on the combination of air pressure and tendon tension) soft robot for use in robot-assisted intra-bronchial intervention. Soft robots, made from compliant materials, have gained popularity for use in surgical interventions due to their dexterity and safety. The current study aimed to design a catheter-like soft robot with an improved performance by minimizing radial expansion during inflation and increasing the force exerted on targeted tissues through geometry optimization. To do so, a finite element analysis (FEA) was employed to optimize the soft robot’s geometry, considering a multi-objective goal function that incorporated factors such as chamber pressures, tendon tensions, and the cross-sectional area. To accomplish this, a cylindrical soft robot with three air chambers, three tendons, and a central working channel was considered. Then, the dimensions of the soft robot, including the length of the air chambers, the diameter of the air chambers, and the offsets of the air chambers and tendon routes, were optimized to minimize the goal function in an in-plane bending scenario. To accurately simulate the behavior of the soft robot, Ecoflex 00-50 samples were tested based on ISO 7743, and a hyperplastic model was fitted on the compression test data. The FEA simulations were performed using the response surface optimization (RSO) module in ANSYS software, which iteratively explored the design space based on defined objectives and constraints. Using RSO, 45 points of experiments were generated based on the geometrical and loading constraints. During the simulations, tendon force was applied to the tip of the soft robot, while simultaneously, air pressure was applied inside the chamber. Following the optimization of the geometry, a prototype of the soft robot with the optimized values was fabricated and tested in a phantom model, mimicking simulated surgical conditions. The decreased actuation effort and radial expansion of the soft robot resulting from the optimization process have the potential to increase the performance of the manipulator. This advancement led to improved control over the soft robot while additionally minimizing unnecessary cross-sectional expansion. The study demonstrates the effectiveness of the optimization methodology for refining the soft robot’s design and highlights its potential for enhancing surgical interventions.
- 21 January 2024
- Published by MDPI AG in Biomimetics
- Vol. 9  (1) , 59
Open AccessDesign Optimization of a Hybrid-Driven Soft Surgical Robot with Biomimetic Constraints
The current study investigated the geometry optimization of a hybrid-driven (based on the combination of air pressure and tendon tension) soft robot for use in robot-assisted intra-bronchial intervention. Soft robots, made from compliant materials, have gained popularity for use in surgical interventions due to their dexterity and safety. The current study aimed to design a catheter-like soft robot with an improved performance by minimizing radial expansion during inflation and increasing the force exerted on targeted tissues through geometry optimization. To do so, a finite element analysis (FEA) was employed to optimize the soft robot’s geometry, considering a multi-objective goal function that incorporated factors such as chamber pressures, tendon tensions, and the cross-sectional area. To accomplish this, a cylindrical soft robot with three air chambers, three tendons, and a central working channel was considered. Then, the dimensions of the soft robot, including the length of the air chambers, the diameter of the air chambers, and the offsets of the air chambers and tendon routes, were optimized to minimize the goal function in an in-plane bending scenario. To accurately simulate the behavior of the soft robot, Ecoflex 00-50 samples were tested based on ISO 7743, and a hyperplastic model was fitted on the compression test data. The FEA simulations were performed using the response surface optimization (RSO) module in ANSYS software, which iteratively explored the design space based on defined objectives and constraints. Using RSO, 45 points of experiments were generated based on the geometrical and loading constraints. During the simulations, tendon force was applied to the tip of the soft robot, while simultaneously, air pressure was applied inside the chamber. Following the optimization of the geometry, a prototype of the soft robot with the optimized values was fabricated and tested in a phantom model, mimicking simulated surgical conditions. The decreased actuation effort and radial expansion of the soft robot resulting from the optimization process have the potential to increase the performance of the manipulator. This advancement led to improved control over the soft robot while additionally minimizing unnecessary cross-sectional expansion. The study demonstrates the effectiveness of the optimization methodology for refining the soft robot’s design and highlights its potential for enhancing surgical interventions.
- Open Access
- 26 December 2023
- Published by MDPI AG in Journal of Personalized Medicine
- Vol. 14  (1) , 33
Machine Learning-Assisted Short-Wave InfraRed (SWIR) Techniques for Biomedical Applications: Towards Personalized Medicine
Personalized medicine transforms healthcare by adapting interventions to individuals’ unique genetic, molecular, and clinical profiles. To maximize diagnostic and/or therapeutic efficacy, personalized medicine requires advanced imaging devices and sensors for accurate assessment and monitoring of individual patient conditions or responses to therapeutics. In the field of biomedical optics, short-wave infrared (SWIR) techniques offer an array of capabilities that hold promise to significantly enhance diagnostics, imaging, and therapeutic interventions. SWIR techniques provide in vivo information, which was previously inaccessible, by making use of its capacity to penetrate biological tissues with reduced attenuation and enable researchers and clinicians to delve deeper into anatomical structures, physiological processes, and molecular interactions. Combining SWIR techniques with machine learning (ML), which is a powerful tool for analyzing information, holds the potential to provide unprecedented accuracy for disease detection, precision in treatment guidance, and correlations of complex biological features, opening the way for the data-driven personalized medicine field. Despite numerous biomedical demonstrations that utilize cutting-edge SWIR techniques, the clinical potential of this approach has remained significantly underexplored. This paper demonstrates how the synergy between SWIR imaging and ML is reshaping biomedical research and clinical applications. As the paper showcases the growing significance of SWIR imaging techniques that are empowered by ML, it calls for continued collaboration between researchers, engineers, and clinicians to boost the translation of this technology into clinics, ultimately bridging the gap between cutting-edge technology and its potential for personalized medicine.
- 26 December 2023
- Published by MDPI AG in Journal of Personalized Medicine
- Vol. 14  (1) , 33
Open AccessMachine Learning-Assisted Short-Wave InfraRed (SWIR) Techniques for Biomedical Applications: Towards Personalized Medicine
Personalized medicine transforms healthcare by adapting interventions to individuals’ unique genetic, molecular, and clinical profiles. To maximize diagnostic and/or therapeutic efficacy, personalized medicine requires advanced imaging devices and sensors for accurate assessment and monitoring of individual patient conditions or responses to therapeutics. In the field of biomedical optics, short-wave infrared (SWIR) techniques offer an array of capabilities that hold promise to significantly enhance diagnostics, imaging, and therapeutic interventions. SWIR techniques provide in vivo information, which was previously inaccessible, by making use of its capacity to penetrate biological tissues with reduced attenuation and enable researchers and clinicians to delve deeper into anatomical structures, physiological processes, and molecular interactions. Combining SWIR techniques with machine learning (ML), which is a powerful tool for analyzing information, holds the potential to provide unprecedented accuracy for disease detection, precision in treatment guidance, and correlations of complex biological features, opening the way for the data-driven personalized medicine field. Despite numerous biomedical demonstrations that utilize cutting-edge SWIR techniques, the clinical potential of this approach has remained significantly underexplored. This paper demonstrates how the synergy between SWIR imaging and ML is reshaping biomedical research and clinical applications. As the paper showcases the growing significance of SWIR imaging techniques that are empowered by ML, it calls for continued collaboration between researchers, engineers, and clinicians to boost the translation of this technology into clinics, ultimately bridging the gap between cutting-edge technology and its potential for personalized medicine.
- Open Access
Cosserat Rod-Based Dynamic Modeling of a Hybrid-Actuated Soft Robot for Robot-Assisted Cardiac Ablation
Soft robotics has emerged as a promising field due to the unique characteristics offered by compliant and flexible structures. Overcoming the challenge of precise position control is crucial in the development of such systems that require accurate modeling of soft robots. In response, a hybrid-actuated soft robot employing both air pressure and tendons was proposed, modeled, and validated using the dynamic Cosserat rod theory. This approach comprehensively addresses various aspects of deformation, including bending, torsion, shear, and extension. The designed robot was intended for robot-assisted cardiac ablation, a minimally invasive procedure that is used to treat cardiac arrhythmias. Within the framework of the Cosserat model, dynamic equations were discretized over time, and ordinary differential equations (ODEs) were solved at each time step. These equations of motion facilitated the prediction of the robot’s response to different control inputs, such as the air pressure and tension applied to the tendons. Experimental studies were conducted on a physical prototype to examine the accuracy of the model. The experiments covered a tension range of 0 to 3 N for each tendon and an air pressure range of 0 to 40 kPa for the central chamber. The results confirmed the accuracy of the model, demonstrating that the dynamic equations successfully predicted the robot’s motion in response to diverse control inputs.
Open AccessCosserat Rod-Based Dynamic Modeling of a Hybrid-Actuated Soft Robot for Robot-Assisted Cardiac Ablation
Soft robotics has emerged as a promising field due to the unique characteristics offered by compliant and flexible structures. Overcoming the challenge of precise position control is crucial in the development of such systems that require accurate modeling of soft robots. In response, a hybrid-actuated soft robot employing both air pressure and tendons was proposed, modeled, and validated using the dynamic Cosserat rod theory. This approach comprehensively addresses various aspects of deformation, including bending, torsion, shear, and extension. The designed robot was intended for robot-assisted cardiac ablation, a minimally invasive procedure that is used to treat cardiac arrhythmias. Within the framework of the Cosserat model, dynamic equations were discretized over time, and ordinary differential equations (ODEs) were solved at each time step. These equations of motion facilitated the prediction of the robot’s response to different control inputs, such as the air pressure and tension applied to the tendons. Experimental studies were conducted on a physical prototype to examine the accuracy of the model. The experiments covered a tension range of 0 to 3 N for each tendon and an air pressure range of 0 to 40 kPa for the central chamber. The results confirmed the accuracy of the model, demonstrating that the dynamic equations successfully predicted the robot’s motion in response to diverse control inputs.