Mengzhe Xu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Mengzhe Xu has a total of 39 co-authors. Here are the top co-authors by number of shared publications:
- Angela Lin Wang2 shared publications
- Carlos Gutiérrez SanRomán
Servicio de Cirugía Pediátrica, Hospital Universitario y Politécnico La Fe, Valencia, España
2 shared publications - Chen Shi2 shared publications
- Chenrun Wang2 shared publications
- Fanrui Zhang
University of Science and Technology of China
2 shared publications - Gen Li
Department of Orthopaedics, Shanghai Key Laboratory for Prevention and Treatment of Bone and Joint Diseases, Shanghai Institute of Traumatology and Orthopaedics, Ruijin Hospital, Shanghai Jiao Tong University School of Medicine, 197 Ruijin 2nd Road, Shanghai 200025, P. R. China
2 shared publications
Latest Publications
- Open Access
- 6 February 2026
- Published by Elsevier in Expert Systems with Applications
- Vol. 313
CARE: A clinical agentic reasoning engine to enhance real-World diagnostic accuracy via structured medical reasoning
- 6 February 2026
- Published by Elsevier in Expert Systems with Applications
- Vol. 313
Open AccessCARE: A clinical agentic reasoning engine to enhance real-World diagnostic accuracy via structured medical reasoning
- 19 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Robotics & Automation Magazine
- Vol. 33 (1) , 52-62
From Patient-Specific Digital Twin to Real-World Phantom: Autonomous Right Heart Catheterization
Right heart catheterization (RHC) is a critical procedure for diagnosing and managing cardiovascular diseases (CVDs) such as heart failure, congenital heart disease, pulmonary edema, and pulmonary hypertension. However, currently prevalent manual RHC procedures require continuous communication of clinicians between the main control room and the operating room, leading to navigation inaccuracies and increased physical workload for clinicians during prolonged procedures. To overcome these challenges, this article introduces a robotic system that enables autonomous RHC (Auto-RHC) by transferring a catheter decision-making model from patient-specific (PS) digital twins to real-world robotic intervention using deep learning (DL) algorithms. By creating a PS digital twin using the Simulation Open Framework Architecture (SOFA) and conducting virtual RHC interventions, images capturing the catheter balloon’s position and aligned behavioral datasets were collected and utilized as input for a convolutional neural network (CNN) architecture. The trained catheter decision-making model derived from the digital twin was then transferred to real-world implementations of robot-assisted Auto-RHC. Experimental results validated the performance of the digital twin and demonstrated that the real-world robotic Auto-RHC achieved a high success rate across both static (≥96%) and dynamic heartbeat (≥94%) PS cardiac phantoms. Furthermore, Auto-RHC enhanced navigation consistency by ≥34.63% compared to expert manual operation.
- 19 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Robotics & Automation Magazine
- Vol. 33 (1) , 52-62
From Patient-Specific Digital Twin to Real-World Phantom: Autonomous Right Heart Catheterization
Right heart catheterization (RHC) is a critical procedure for diagnosing and managing cardiovascular diseases (CVDs) such as heart failure, congenital heart disease, pulmonary edema, and pulmonary hypertension. However, currently prevalent manual RHC procedures require continuous communication of clinicians between the main control room and the operating room, leading to navigation inaccuracies and increased physical workload for clinicians during prolonged procedures. To overcome these challenges, this article introduces a robotic system that enables autonomous RHC (Auto-RHC) by transferring a catheter decision-making model from patient-specific (PS) digital twins to real-world robotic intervention using deep learning (DL) algorithms. By creating a PS digital twin using the Simulation Open Framework Architecture (SOFA) and conducting virtual RHC interventions, images capturing the catheter balloon’s position and aligned behavioral datasets were collected and utilized as input for a convolutional neural network (CNN) architecture. The trained catheter decision-making model derived from the digital twin was then transferred to real-world implementations of robot-assisted Auto-RHC. Experimental results validated the performance of the digital twin and demonstrated that the real-world robotic Auto-RHC achieved a high success rate across both static (≥96%) and dynamic heartbeat (≥94%) PS cardiac phantoms. Furthermore, Auto-RHC enhanced navigation consistency by ≥34.63% compared to expert manual operation.
- Open Access
- 5 December 2025
- Published by Elsevier in Information Fusion
- Vol. 129
CX-Mind: a pioneering multimodal large language model for interleaved reasoning in chest X-ray via curriculum-guided reinforcement learning
- 5 December 2025
- Published by Elsevier in Information Fusion
- Vol. 129
Open AccessCX-Mind: a pioneering multimodal large language model for interleaved reasoning in chest X-ray via curriculum-guided reinforcement learning
- 18 December 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 73 (00189456) , 1-8
A High-Resolution Image Reconstruction Method of Lung Nodules With B-Spline-Based Whale Optimization Algorithm (B-WOA) for Electrical Impedance Tomography
A B-spline based Whale Optimization Algorithm (B-WOA) is proposed to realize a high-resolution Electrical Impedance Tomography (EIT) image reconstruction of lung nodules. The method achieves the mapping between the target position and the conductivity distribution by B-spline curve, and establishes the optimized objective function by the sensitivity matrix method, based on which the lung nodule image reconstruction is realized by the WOA. The quality of images reconstructed using WOA and three conventional algorithms is compared by numerical simulation in the randomly generated lung nodule model. The simulation results show that the average value of image correlation coefficient (ICC) of the reconstructed images using the WOA is improved by 4.8% compared with the Tikhonov regularization algorithm, which performs best among the traditional algorithms. In addition, an experimental model of lung nodules was established to validate the proposed method. Biological tissues of different diameters were used to simulate the lung nodule region. The performance of the WOA, the Tikhonov regularization algorithm, and the GA were investigated in the dual-target imaging experiment. The experimental results show that the average ICC values of the reconstructed images using the WOA are improved by 4.7% and 3.5% over the Tikhonov regularization algorithm and the GA, respectively. Therefore, the proposed B-WOA proposed can achieve high-resolution imaging of the distribution of tiny targets, which is expected to be used for the clinical detection of lung nodules.
- 18 December 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 73 (00189456) , 1-8
A High-Resolution Image Reconstruction Method of Lung Nodules With B-Spline-Based Whale Optimization Algorithm (B-WOA) for Electrical Impedance Tomography
A B-spline based Whale Optimization Algorithm (B-WOA) is proposed to realize a high-resolution Electrical Impedance Tomography (EIT) image reconstruction of lung nodules. The method achieves the mapping between the target position and the conductivity distribution by B-spline curve, and establishes the optimized objective function by the sensitivity matrix method, based on which the lung nodule image reconstruction is realized by the WOA. The quality of images reconstructed using WOA and three conventional algorithms is compared by numerical simulation in the randomly generated lung nodule model. The simulation results show that the average value of image correlation coefficient (ICC) of the reconstructed images using the WOA is improved by 4.8% compared with the Tikhonov regularization algorithm, which performs best among the traditional algorithms. In addition, an experimental model of lung nodules was established to validate the proposed method. Biological tissues of different diameters were used to simulate the lung nodule region. The performance of the WOA, the Tikhonov regularization algorithm, and the GA were investigated in the dual-target imaging experiment. The experimental results show that the average ICC values of the reconstructed images using the WOA are improved by 4.7% and 3.5% over the Tikhonov regularization algorithm and the GA, respectively. Therefore, the proposed B-WOA proposed can achieve high-resolution imaging of the distribution of tiny targets, which is expected to be used for the clinical detection of lung nodules.