Zhuofu Li

Zhuofu Li

Auto-generated profile, based on publicly available information

Zhuofu Li

Department of Orthopaedics, Peking University Third Hospital, 49 North Garden Road, Haidian District, Beijing 100191, China

Research Focus & Metrics

Academic Career Length

4Years (2023-Present)

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

Representative Publications

View All Publications
21 March 2023 in Journal of Bone and Joint Surgery
Verified
Journal Article
Closed access
ZLZhuofu Li
Zhuofu Li
CWChengxia Wang
Chengxia Wang
XSXiongkang Song
Xiongkang Song
SLShanshan Liu
Shanshan Liu
YZYonghong Zhang
Yonghong Zhang
WLWeishi Li
Weishi Li
Background: The main function of robots in spine surgery is to assist with pedicle screw placement. Laminectomy, which is as important as pedicle screw placement, lacks a mature robot-assisted system. The aims of this study were to introduce the first autonomous laminectomy robot, to explore the feasibility of autonomous robotic laminectomy, and to validate its accuracy using a cadaveric model. Methods: Forty vertebrae from 4 cadavers were included in the study; 7 thoracic and 3 lumbar vertebrae were randomly selected in each cadaver. The surgeon was able to plan the laminectomy path based on computed tomographic (CT) data before the surgical procedure. The robot performed the laminectomy autonomously, and a postoperative CT scan was made. The deviation of each cutting plane from the plan was quantitatively analyzed, and the accuracy and safety were qualitatively evaluated. The time required for the laminectomy was also recorded. Results: Cuts were performed in 80 laminectomy planes (56 for thoracic vertebrae and 24 for lumbar vertebrae). The mean time for 1-sided laminectomy was 333.59 ± 116.49 seconds, which was shorter for thoracic vertebrae (284.41 ± 66.04 seconds) than lumbar vertebrae (448.33 ± 128.65 seconds) (p < 0.001). The mean time for single-level total laminectomy was 814.05 ± 302.23 seconds, which was also shorter for thoracic vertebrae (690.46 ± 165.74 seconds) than lumbar vertebrae (1,102.42 ± 356.13 seconds) (p = 0.002). The mean deviation of the cutting plane from the plan was 0.67 ± 0.30 mm for the most superior cutting point and 0.73 ± 0.31 mm for the most inferior point. There were no significant differences in the deviation between thoracic vertebrae (0.66 ± 0.26 mm) and lumbar vertebrae (0.67 ± 0.38 mm) at the superior cutting point (p = 0.908) and between thoracic vertebrae (0.72 ± 0.30 mm) and lumbar vertebrae (0.73 ± 0.33 mm) at the inferior cutting point (p = 0.923). In the qualitative analysis of the accuracy of the 80 laminectomy planes, 66 (83%) were classified as grade A, 14 (18%) were grade B, and none was grade C. In the safety analysis, 65 planes (81%) were considered safe and the safety of the other 15 planes (19%) was considered uncertain. Conclusions: The results confirmed the accuracy of this robotic system, supporting its use for laminectomy of thoracolumbar vertebrae. Level of Evidence: Therapeutic Level V. See Instructions for Authors for a complete description of levels of evidence.

1 January 2023 in International Journal of Intelligent Systems
Verified
Journal Article
Open access
PDF file
YZYonghong Zhang
Yonghong Zhang
XJXuquan Ji
Xuquan Ji
WLWenyong Liu
Wenyong Liu
ZLZhuofu Li
Zhuofu Li
JZJian Zhang
Jian Zhang
WLWeishi Li
Weishi Li
High‐precision image segmentation of the spine in computed tomography (CT) images is important for the diagnosis of spinal diseases and surgical path planning. Manual segmentation is often tedious and time consuming. Thus, an automatic segmentation algorithm is expected to solve this problem. However, because different areas are scanned, the number of spines in the original CT image and the coverage area are often different, making it extremely difficult to directly conduct a fully autonomous spine segmentation. In this study, we propose a two‐stage automatic spine segmentation method based on 3D Swin Transformer. In the first stage, the 3D Swin‐YoloX algorithm is used to achieve an accurate positioning of each spine segment in the CT images. In the second stage, 3D Swin‐UNet is used to achieve a high‐precision segmentation of the spine. Using an open dataset, the average Dice of our approach can reach 0.942 and the average Hausdorff distance can reach 6.24, indicating a higher accuracy in comparison with other published methods. Our proposed method can effectively eliminate any adverse effects of the different scanning areas on a spinal image segmentation and has a high application value.

1 January 2023 in International Journal of Intelligent Systems
Verified
Journal Article
Open access
PDF file
LHLei Hu
Lei Hu
YZYonghong Zhang
Yonghong Zhang
JZJian Zhang
Jian Zhang
TZTianyang Zhang
Tianyang Zhang
XSXiongkang Song
Xiongkang Song
WLWeishi Li
Weishi Li
Spinal surgery robots have a great application value in laminar decompression surgery. For a safe surgery, the robot needs to accurately identify the cutting state of the lamina. Therefore, it is very important to deal with various sensing signals in the form of a time series. However, various state recognition algorithms proposed so far cannot completely avoid cutting through the lamina, leaving hidden dangers for nerve thermal damage caused by high‐temperature liquid splashing. We propose a long time series prediction algorithm called STP‐Net, which combined with the existing algorithms for recognizing the lamina cutting state can stop the lamina cutting in advance, thus preserving a thin inner layer of cortical bone and blocking high‐temperature liquid splashing. STP‐Net has the following advantages: (1) Dimension reduction is first performed on Swin (shifted window) Transformer to obtain 1D Swin Transformer, which was used to develop STP‐Net for long time series prediction, which not only has high accuracy but also shows a linear relationship between computational complexity and input sequence length L, namely, O(L). (2) The token merging layer is proposed and applied to the encoder of STP‐Net, which reduces the computation cost and improves the global information extraction capability of the algorithm. (3) STP‐Net uses a generative decoder to output the prediction sequence directly through a single operation, which not only has high efficiency but also avoids the accumulation of errors. Taking force signals as an example, when 400 numbers are used to predict 100 numbers, the average mean square error (MSE) and mean absolute error (MAE) of standard STP‐Net are 1.22 × 10−3 and 2.42 × 10−2, respectively. STP‐Net, combined with the existing method for recognizing the lamina cutting status, can stop the robot from cutting in advance in 87.14% of the cases. In addition, the results of practical lamina cutting experiments confirmed the effectiveness of STP‐Net.

14 September 2026 in Spine
Verified
Journal Article
Closed access
WLWeishi Li
Weishi Li
ZSZhuoran Sun
Zhuoran Sun
WQWeipeng Qiu
Weipeng Qiu
SJShuai Jiang
Shuai Jiang
JLJialiang Lin
Jialiang Lin
TZTianyuan Zhao
Tianyuan Zhao
Was not provided for this article.