YZYao Zhang

Yao Zhang

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Yao Zhang has a total of 37 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Deep-Learning-Based Control of a Decoupled Two-Segment Continuum Robot for Endoscopic Submucosal Dissection

    by Yuancheng Shao, Yao Zhang, Jia Gu, Zixi Chen, Di Wu, Yuqiao Chen, Bo Lu, Wenjie Liu, Cesare Stefanini, Peng Qi

    Manual endoscopic submucosal dissection (ESD) is technically demanding, and existing single-segment robotic tools offer limited dexterity. These limitations motivate the development of more advanced solutions. To address this, decoupled endoscopic submucosal dissection robot (DESectBot), a novel dual-segment continuum robot with a decoupled structure and integrated surgical forceps, enabling six degrees of freedom tip dexterity for improved lesion targeting in ESD, was developed in this work. Deep learning controllers based on gated recurrent units (GRUs) for simultaneous tip position and orientation control, effectively handling the nonlinear coupling between continuum segments, were proposed. The GRU controller was benchmarked against Jacobian-based inverse kinematics, model predictive control (MPC), a feedforward neural network, and a long short-term memory (LSTM) network. In addition, a systematic ablation study under identical training conditions showed that, for the considered control task, the GRU achieves better generalization, faster convergence, and higher computational efficiency than the LSTM. In nested-rectangle and Lissajous trajectory tracking tasks, the GRU achieved the lowest position/orientation root-mean-square errors (RMSEs): 1.11 mm/ 4.62$^{\circ }$ and 0.81 mm/ 2.59$^{\circ }$, respectively. For orientation control at a fixed position (four target poses), the GRU attained a mean RMSE of 0.14 mm and 0.72$^{\circ }$, outperforming all alternatives. In a peg transfer task, the GRU achieved a 100% success rate (120 success/120 attempts) with an average transfer time of 11.8 s, the standard deviation significantly outperforms no vice-controlled systems. In addition, an ex vivo ESD demonstration—grasping, elevating, and resecting tissue as the scalpel completed the cut—confirmed that DESectBot provides sufficient stiffness to divide thick gastric mucosa and an operative workspace adequate for large lesions. These results confirm that GRU-based control significantly enhances precision, reliability, and usability in ESD surgical training scenarios.

  • Electrical Bioimpedance Sensing-Based Proximity Distance Estimation for Needle Interventions: Two-Layer Model and Validation

    by Yao Zhang, Zhuoqi Cheng, Mouloud Ourak, Thiusius Rajeeth Savarimuthu, Emmanuel Vander Poorten

    Needle-based interventions are routine in clinical practice, yet clinicians lack reliable, precise, real-time estimates of the tip-to-target distance—–a key determinant of safety and efficiency. To bridge this critical clinical gap, physics-based models that maps electrical bio-impedance (EBI) measurements to the absolute tip-to-target distance by leveraging tissue-intrinsic electrical properties encoded in EBI are presented in this work. A tripolar electrode configuration is adopted to ensure tool-independence and broad compatibility across instruments and anatomical sites. The optimal layout of electrode radius, inter-electrode spacing, and placement was identified through a systematic parametric study. The model was validated across finite-element simulations, saline solution experiments, and ex vivo layered tissue. In all settings, proximity distance estimates were accurate and robust across different insertion angles. In simulation, it achieved 0.55 mm RMSE/1.06 mm MAE when the target layer was less conductive than the first layer and 0.40 mm RMSE/1.21 mm MAE when it was more conductive. In saline, four insertion angles yielded millimeter errors with no angle dependence. In ex vivo layered tissue, accuracy remained near the millimeter level across angles (RMSE 1.00–1.13 mm and MAE 1.30–1.58 mm), which is slightly lower than saline yet consistent across orientations. Together, these results establish a general, instrument-agnostic framework for absolute proximity distance sensing during needle interventions with the potential to enhance safety and efficiency in minimally invasive procedures. Furthermore, by first focusing on a two-layer model, this work provides a clear pathway to multi-layer tissue modeling and broader anatomical generalization.


  • Development and Control of a Macro–Micro Multi-Arm Robotic System for Open Spina Bifida Treatment

    by Xiaohang Zhang, Di Wu, Matthias Tummers, Yuyu Cai, Yao Zhang, Tom Vandebroek, Gianni Borghesan, Jan Deprest, Emmanuel Vander Poorten

    Open spina bifida (OSB), a congenital anomaly caused by incomplete spinal column closure, leads to progressive nerve damage. While multi-port minimally invasive procedures are being optimized clinically, single-port access (SPA) surgery presents a promising, less invasive alternative. Beyond reducing scarring, SPA may also decrease the risk of iatrogenic preterm premature rupture of membranes and enhance maternal recovery, as suggested by outcomes in other minimally invasive interventions. However, technical challenges persist in SPA implementations, particularly in instrument miniaturization and control precision. To facilitate this approach, a macro-micro multi-arm robotic system is under development. The system features an 11-mm-diameter insertion port to deploy four instrument arms. The arm design incorporates a hybrid backbone structure. A macro proximal section of concentric tubes provides stable deployment, while a micro distal flexible segment enables the dexterous manipulation. This paper presents dedicated modeling and control strategies to address actuator and transmission nonlinearities arising from stringent miniaturization requirements. A hybrid framework combining joint-level feedback with deep-learning-based feedforward compensation mitigates limitations imposed by pneumatic artificial muscles and tendon transmissions. The system is validated through trajectory-following experiments, each conducted over 10 trials. Preliminary results show an average root-mean-squared error of 0.28 mm, an average maximum absolute error of 0.85 mm, and an average standard deviation of 0.18 mm across both trajectories. These results demonstrate consistent sub-millimeter precision, satisfying the requirement for OSB repair. The feasibility of the system was then verified through a phantom study.


  • Electrical Bioimpedance Sensing-Based Microcatheter Alignment in CTO-PCI

    by Yao Zhang, Martina Polidoro, Mouloud Ourak, Eric Verschooten, Johan Bennett, Emmanuel Vander Poorten

    Percutaneous coronary intervention (PCI) is increasingly utilized for treating chronic total occlusion (CTO) coronary arteries. When direct passage through the occlusion is not feasible, the retrograde “kissing wire technique” (KWT) provides a valuable cannulation alternative. The KWT technique employs two microcatheters: one in the subintimal layer and the other in the true lumen on the opposite side of the occlusion. The latter microcatheter acts as a visual marker to guide the re-entry of the microcatheter located in the subintimal layer toward the true lumen. However, accurately determining the microcatheter’s orientation, which is crucial for ensuring the correct re-entry direction and avoiding the pericardial space, remains a significant challenge. This work presents a new method based on multi-electrical bio-impedance (EBI) sensing and cubic spline interpolation (CSI). The angular range for the correct re-entry orientation was calculated based on the dimensions of the blood vessel and the microcatheters. By measuring the EBI from each electrode around the microcatheter tip, the catheter position is estimated through the minimum EBI, indicating the optimal reentry point. Experiments conducted in pure saline and in vitro conditions demonstrate that the approach reduces errors by over 30% at three distinct roll angles compared to the angular range. Furthermore, experiments across all angles revealed a maximum estimation error approximately 60% lower than the clinical standard. These results indicate significant potential for enhancing the efficacy and reducing the risks of the KWT technique through the EBI-guided approach.


  • Physiological Motion Compensation for Neuroscience Research Based on Electrical Bio-Impedance Sensing

    by Yao Zhang, Eric Verschooten, Mouloud Ourak, Kaat Van Assche, Gianni Borghesan, Di Wu, Kenan Niu, Philip X. Joris, Emmanuel Vander Poorten

    Researchers have developed a large number of methods to study the brain’s function. One of the most effective techniques is in vivo whole-cell patch clamp recording which allows the recording of intracellular neuronal activity. A major issue that drastically reduces the efficiency of in vivo patch clamping is the excessive movement of the brain primarily caused by heartbeat and breathing, which can be larger than the size of the neurons under investigation. Motion compensation techniques are complicated due to the lack of sensors to reliably measure local physiologically-induced motion. This work proposes the use of Electrical Bio-impedance (EBI) to the existing patch electrodes in the patching pipette as a proximity sensor. The study further develops an Extended Kalman Filter (EKF) to estimate overall motion and then establishes a motion compensation algorithm for the patch pipette. The proposed method was developed on a custom lab benchtop setup and validated during actual in vivo experiments. The results of the lab experiments show a real-time compensatory performance exceeding 80%. The in vivo experiments achieved a performance of over 75%, confirming the ability to compensate for real physiologically induced motion. Moreover, the method demonstrated dynamic continuous motion compensation while the electrode was advanced to a neuron, contacting the neuron membrane without damage illustrating the ability to improve neuronal patch clamping. As far as the authors are aware this is the first time that physiologically induced motion can be compensated for this application and this solely relies on EBI.


  • Open Access

    Feasibility of using an Extended Kalman Filter for Motion Estimation based on Electrical Bioimpedance Sensing

    by Yao Zhang, Kaat Van Assche, Mouloud Ourak, Eric Verschooten, Gianni Borghesan, Kenan Niu, Philip Joris, Emmanuel Vander Poorten

    Patch-clamp is a widely used technique to record the electrophysiological activity of neurons in vivo. However, establishing and maintaining a long-term stable recording is difficult due to neuronal motion induced by physiological motion. This abstract proposes an Extended Kalman Filter (EKF) method for motion estimation based on Electrical Bioimpedance (EBI) sensing. The results show that with EBI, the EKF could estimate the motion with high precision (RMSE = 0.022V) and robustness (STD = 0.0028V) in real-time.

  • Deep-Learning-Based Compliant Motion Control of a Pneumatically-Driven Robotic Catheter

    by Di Wu, Xuan Thao Ha, Yao Zhang, Mouloud Ourak, Gianni Borghesan, Kenan Niu, Fabian Trauzettel, Jenny Dankelman, Arianna Menciassi, Emmanuel Vander Poorten

    In cardiovascular interventions, when steering catheters and especially robotic catheters, great care should be paid to prevent applying too large forces on the vessel walls as this could dislodge calcifications, induce scars or even cause perforation. To address this challenge, this paper presents a novel compliant motion control algorithm that relies solely on position sensing of the catheter tip and knowledge of the catheter’s behavior. The proposed algorithm features a data-driven tip position controller. The controller is trained based on a so-called control Long Short-Term Memory Network (control-LSTM). Trajectory following experiments are conducted to validate the quality of the proposed control-LSTM. Results demonstrated superior positioning capability with sub-degree precision of the new approach in the presence of severe rate-dependent hysteresis. Experiments both in a simplified setup as well as in an aortic phantom further show that the proposed approach allows reducing the interaction forces with the environment by around 70%. This work shows how deep learning can be exploited advantageously to avoid tedious modeling that would be needed to precisely steer continuum robots in constrained environments such as the patient’s vasculature.


  • Hysteresis Modeling of Robotic Catheters Based on Long Short-Term Memory Network for Improved Environment Reconstruction

    by Di Wu, Yao Zhang, Mouloud Ourak, Kenan Niu, Jenny Dankelman, Emmanuel Vander Poorten

    Catheters are increasingly being used to tackle problems in the cardiovascular system. However, positioning precision of the catheter tip is negatively affected by hysteresis. To ensure tissue damage due to imprecise positioning is avoided, hysteresis is to be understood and compensated for. This work investigates the feasibility to model hysteresis with a Long Short-Term Memory (LSTM) network. A bench-top setup containing a catheter distal segment was developed for model evaluation. The LSTM was first tested using four groups of test datasets containing diverse patterns. To compare with the LSTM, a Deadband Rate-Dependent Prandtl-Ishlinskii (DRDPI) model and a Support Vector Regression (SVR) model were established. The results demonstrated that the LSTM is capable of predicting the tip bending angle with sub-degree precision. The LSTM outperformed the DRDPI model and the SVR model by 60.1% and 36.0%, respectively, in arbitrarily varying signals. Next, the LSTM was further validated in a 3D reconstruction experiment using Forward-Looking Optical Coherence Tomography (FL-OCT). The results revealed that the LSTM was able to accurately reconstruct the environment with a reconstruction error below 0.25 mm. Overall, the proposed LSTM enabled precise free-space control of a robotic catheter in the presence of severe hysteresis. The LSTM predicted the catheter tip response precisely based on proximal input pressure, minimizing the need to install sensors at the catheter tip for localization.