Xinze Li

Xinze Li

Auto-generated profile, based on publicly available information

Xinze Li

Suzhou Institute of Biomedical Engineering and Technology, Chinese Academy of Sciences, Suzhou, China

Research Focus & Metrics

Academic Career Length

3Years (2024-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
13 May 2024 in IEEE/ASME Transactions on Mechatronics
Verified
Journal Article
Closed access
ZYZhengxin Yang
Zhengxin Yang
LLLihao Liu
Lihao Liu
XLXinze Li
Xinze Li
JLJiaqi Li
Jiaqi Li
YJYang Jiao
Yang Jiao
YCYaoyao Cui
Yaoyao Cui
Atherosclerosis accounts for over 30% of global deaths annually. The intravascular ultrasound (IVUS) can provide detailed information on lesions, enabling accurate diagnosis, and treatment of vascular diseases. The existing IVUS catheters face challenges to work in tortuous blood vessels. This study presents a dual-mode magnetic IVUS robot, featuring by miniature size and high softness, which allows for intravascular imaging in hard-to-reach regions. First, the magnetic IVUS robot consisting of an acoustic-magnetic tip and a soft catheter is designed, which performs: 1) controllable bending under static magnetic fields to navigate in complex vasculature; 2) shaft-free acoustic beam scanning under dynamic magnetic fields to provide high-resolution ultrasound images. Then, the motion modeling of navigating mode and imaging mode are analyzed, respectively. Next, an integrated system and actuating strategy are established, realizing the dual-mode decoupling control of the magnetic IVUS robot. Finally, the effectiveness of the presented method is validated in vascular models, imaging phantoms, and in-vitro porcine artery vessel. The designed magnetic IVUS robot achieves dual objectives, including active angle steering at bifurcations and acoustic beam scanning at lesions, simultaneously improving the operating flexibility and imaging stability. The proposed design and control scheme demonstrate the feasibility of tortuous intravascular imaging, providing a potential solution for clinical applications.

9 April 2024 in Biomedical Signal Processing and Control
Verified
Journal Article
Closed access
XLXinze Li
Xinze Li
PSPeng Song
Peng Song
TLTiantian Lv
Tiantian Lv
YJYang Jiao
Yang Jiao
YGYunbo Guo
Yunbo Guo
YCYaoyao Cui
Yaoyao Cui

30 November 2025 in Clinical Endoscopy
Verified
Journal Article
Open access
ZFZijun Fan
Zijun Fan
ZGZhenyun Gong
Zhenyun Gong
RBRun Bao
Run Bao
QLQinkai Li
Qinkai Li
WWWei Wu
Wei Wu
DHDuanmin Hu
Duanmin Hu
Background/Aims: Lymphadenopathy presents diagnostic challenges, particularly for the mediastinal and intra-abdominal lymph nodes (LNs). Endoscopic ultrasonography (EUS) has emerged as a tool for LN detection; however, its accuracy varies. To enhance the diagnostic performance and minimize medical costs, assisting LN assessment using EUS is necessary. Machine learning (ML) offers potential for medical image analysis. This study aimed to develop an ML model for classifying mediastinal and intra-abdominal LNs using gastrointestinal EUS.Methods: EUS images of mediastinal and intra-abdominal LNs were randomly split into training and validation datasets. U-Net was selected for LN segmentation, and six deep-learning architectures were combined with the k-nearest-neighbor algorithm for LN classification. Physicians, comprising one expert group and one trainee group, reviewed the validation dataset and made individual diagnoses. A logistic regression model was generated based on LN features. We compared the diagnostic yields of ML, expert and trainee groups, logistic regression analysis, and a combination of the various methods mentioned above for diagnosing LNs.Results: In total, 93 patients were enrolled, providing 630 images. The ResNet-50+logistic regression analysis+expert group achieved the best F1 score and sensitivity of 0.89 and 100.0%, respectively. Paired comparisons revealed that the combination outperformed both experts and trainees in terms of the area under the curve (p<0.01).Conclusions: ML assists in predicting the mediastinal and intra-abdominal LNs based on gastrointestinal EUS images, particularly when combined with expert expertise and logistic regression models.

10 April 2026 in IEEE/ASME Transactions on Mechatronics
Verified
Journal Article
Closed access
ZYZhengxin Yang
Zhengxin Yang
LLLihao Liu
Lihao Liu
XLXinze Li
Xinze Li
JLJiaqi Li
Jiaqi Li
YCYaoyao Cui
Yaoyao Cui
YJYang Jiao
Yang Jiao
Localization of the magnetic capsule endoscope is crucial to accurate actuation. Despite progress with internal magnetic field sensors and external magnetic sources, achieving reliable localization remains challenging in the presence of a time-varying internal magnetic field (IMF). This study proposes a compound localization method for an ultrasound capsule endoscope (USCE) based on an IMF sensor array and an external permanent magnet source. The method enables simultaneous 6-degrees of freedom (DOF) pose detection for closed-loop magnetic navigation and real-time ultrasound (US) beam scanning angle sensing for distortion-free US image reconstruction. First, a USCE consisting of an internal permanent magnet, a US transducer, Hall sensors, and an anchoring balloon is designed, enabling both spiral structure-based locomotion and high-quality endoluminal US imaging. Next, a compound localization strategy is presented, which realizes the separation of the time-varying IMF and external magnetic field, facilitating synchronous 6-DOF pose detection and scanning angle sensing. Furthermore, a closed-loop magnetic actuation method is proposed that allows automatic propelling in unknown tubular environments. Finally, the effectiveness of the presented method is validated through simulations, localization tests, imaging tests, and actuation tests. The demonstrated static pose detection error is 4.08 $\pm$ 1.91 mm in position norm and 2.46 $\pm$ 1.31$^\circ$ in heading orientation, within a workspace shared with the robotic manipulator. In addition, the proposed scanning angle sensing successfully corrects US image distortion, providing subsurface structures for accurate diagnosis.