Yibin Zhang¹, Yan Wang¹, Hongliu Yu²¹School of Medical Devices, Shanghai University of Medicine and Health Sciences, Shanghai 201318, China. ²School of Health Science and Engineering, University of Shanghai for Science and Technology, Shanghai 200093, China.

Address correspondence to: Hongliu Yu, School of Health Science and Engineering, University of Shanghai for Science and Technology, Shanghai city Jungong road 516, Shanghai 200093, China. E-mail: yhl98@hotmail.com.

DOI：https://doi.org/10.61189/961030gznunx

Received June 21, 2024; Accepted November 20, 2024; Published December 31, 2024

Highlights

● In prosthetics, using AI algorithms to identify the fused sensor data as known walking patterns has extremely strong expandability. Moreover, as the learning data continues to expand, the robustness of the model itself also increases accordingly.● There are numerous AI algorithms currently available. The effective utilization of algorithm combination techniques to learn from each other’s strengths can significantly improve the accuracy of identification. The combined model of convolutional neural networks (CNN) and bidirectional long short term memory (LSTM) attempted in this paper has witnessed a significant improvement in its comprehensive recognition rate.● In the practical application of prosthetics, the real-time performance during the mode switching transition period is particularly important as it can reflect the flexibility of the prosthetics. In this paper, the algorithm optimized by the AI model has controlled the delay rate within one gait cycle, greatly enhancing the safety and reliability of pro-sthetics in actual use.

Research Article |Published on: 31 December 2024

[Progress in Medical Devices] 2024; 2 (4): 144-152

Mingxia Wei, Qingyun Meng

School of Health Science and Engineering, University of Shanghai for Science and Technology, Shanghai 200093, China.

Address correspondence to: Qingyun Meng, School of Health Science and Engineering, University of Shanghai for Science and Technology, Yangpu District, Shanghai 200093, China. Tel: +86-13761813609. E-mail: mengqy@sumhs.edu.cn.

DOI: https://doi.org/10.61189/521889ygxkmc

Received May 15, 2024; Accepted June 25, 2024; Published December 31, 2024

Highlights

● A comprehensive overview of Visual-Inertial Navigation Systems.● Exploration of key technologies, including image processing methods for visual odometry.

Review Article |Published on: 31 December 2024

[Progress in Medical Devices] 2024; 2 (4): 153-160

Haonan Geng¹, Xudong Guo¹, Haibo Lin¹, Youguo Hao², Guojie Zhang³

¹School of Health Science and Engineering, University of Shanghai for Science and Technology, Shanghai 200093, China. ²Shanghai Putuo District People’s Hospital, Shanghai 200060, China. ³LingYuan Iron and Steel CO., LTD, Lingyuan 122500, Liaoning Province, China.

Address correspondence to: Xudong Guo, School of Health Science and Engineering, University of Shanghai for Science and Technology, No.516 Jungong Road, Yangpu District, Shanghai 200093, China. Email: guoxd@usst.edu.cn; Youguo Hao, Shanghai Putuo District Central Hospital, No.1291 Jiangning Road, Putuo District, Shanghai, 200060, China. Email: youguohao6@163.com.

DOI: https://doi.org/10.61189/673672yizrwd

Received September 8, 2024; Accepted November 6, 2024; Published December 31,2024

Highlights

●Gait prediction relies on multimodal sensor data, and the acquisition of multimodal information, such as physical sensors and bioelectrical signal sensors, is introduced in order to monitor and analyze the lower limb movement in real time, and provide a data basis for prediction.● The application of machine learning algorithms in gait prediction technology, such as Support Vector Machine, Random Forest, and Back Propagation Neural Network, is reviewed to construct an optimized gait prediction model, which provides effective support for the intelligent control of exoskeleton.● Compared with machine learning algorithms, the article summarizes the researchers’ efforts to extract and un derstand the hidden patterns in gait data by constructing neural network models related to different deep learning algorithms, which are used to improve the accuracy and robustness of gait prediction.

Review Article |Published on: 31 December 2024

[Progress in Medical Devices] 2024; 2 (4): 161-173

Fei Liu¹, Haipo Cui¹, Fujia Sun², Shuhao Hou³, Peng Yue⁴ 

Schools of ¹Health Science and Engineering, ²Mechanical Engineering, ³Materials and Chemistry, University of Shanghai for Science and Technology, Shanghai 200093, China. ⁴Shanghai Guoyan Medical Device Testing Cen ter Co., Ltd., Shanghai 200000, China. 

Address correspondence to: Haipo Cui, School of Health Science and Engineering, University of Shanghai for Science and Technology, NO.334, Jungong Road, Shanghai 200093, China. Tel: +86 21-55271290, E-mail: hpcui@usst.edu.cn; Fujia Sun, School of Mechanical Engineering, University of Shanghai for Science and Technology, NO.516, Jungong Road, Shanghai 200093, China. Tel: +86 13621773624, E-mail: chinasfj@126.com.

DOI: https://doi.org/10.61189/273582nrnxmc

Received August 12, 2024; Accepted September 11, 2024; Published March 31, 2025

Highlights 

 ● Algorithms such as adaptive beamforming and synthetic aperture technology have significantly improved the quality of ultrasound images. 

 ● New algorithms, such as deep learning, can adapt to more complex signal environments at the expense of real-time performance. 

 ● Combining different algorithms can overcome the limitations of a single algorithm, thereby improving image resolution, contrast, and noise resistance.

Review Article |Published on: 31 March 2025

[Progress in Medical Devices] 2025; 3 (1): 26-42

Yilong Chen, Lin Mao, Zijie Zhou, Chengli Song 

Shanghai Institute for Minimally Invasive Therapy, School of Health Science and Engineering, University of Shanghai for Science and Technology, Shanghai 200093, China.

Address correspondence to: Lin Mao, Shanghai Institute for Minimally Invasive Therapy, School of  Health Science and Engineering, University of Shanghai for Science and Technology, Yangpu District,  Shanghai 200093, China. Tel: +86-21-55572159. E-mail: linmao@usst.edu.cn.

DOI: https://doi.org/10.61189/314845qnicsc

Received January 19, 2025; Accepted February 19, 2025; Published March 31, 2025

Highlights

● Continuous suturing in traditional manual suturing shortens operation time and reduces infection risk. Absorbable sutures are preferred for intestinal suturing and anastomosis to minimize foreign body reactions.

● Mechanical anastomosis with linear and circular metal staples offers distinct advantages, while new biodegradable staples demonstrate good performance. 

● Magnetopressure anastomosis, leveraging magnetic attraction, has shown success in specific scenarios, providing innovative approaches to intestinal anastomosis. 

● Radio frequency energy tissue welding technology enables rapid, seamless intestinal anastomosis, with   fewer complications and holds strong potential for future applications. 

● The support method for intestinal anastomosis, particularly the "degradable internal stent anastomosis" using a simple support method, shows significant promise in animal studies.

Review Article |Published on: 31 March 2025

[Progress in Medical Devices] 2025; 3 (1): 66-76

Haonan Geng¹, Xudong Guo¹, Fengqi Zhong², Haibo Lin¹, Guojie Zhang³, Qin Zhang⁴, Jiaheng Chen¹

¹School of Health Science and Engineering, University of Shanghai for Science and Technology, Shanghai 200093, China. ²CloudSemi, Pudong New Area, Shanghai 200120, China. ³LingYuan Iron and Steel CO., LTD, Lingyuan 122500, Liaoning Province, China. ⁴Medical Engineering Department of Northern Jiangsu People’s Hospital, Yangzhou 225001, Jiangsu Province, China.

Address correspondence to: Xudong Guo, School of Health Science and Engineering, University of Shanghai for Science and Technology, No.516 Jungong Road, Shanghai 200093, China. Email: guoxd@usst.edu.cn.

DOI: https://doi.org/10.61189/164995qvdasw

Received September 29, 2024; Accepted December 3, 2024; Published March 31, 2025

Highlights

● The paper develops a gait prediction control strategy for lower limb exoskeleton robots using a real-time adaptive Kalman filtering algorithm, with public gait data from a Clinical Gait Analysis serving as input.

● The model incorporates motor rotation angle, angular velocity, and angular acceleration as core parameters, calculated based on the principles of uniformly accelerated motion. It achieves gait prediction by initializing parameters, calculating Kalman gain, correcting measurements, and updating the covariance matrix.

● A control strategy guided by normal gait parameters enables the exoskeleton to transition efficiently into the desired motion state during startup and gait phase switching. The system employs a microcontroller and Raspberry Pi as its control core, integrated with Bluetooth communication for effective robot control.

Review Article |Published on: 31 March 2025

[Progress in Medical Devices] 2025; 3 (1): 57-65

Juxiao Wang, Lin Mao, Weiwei Fan, Chengli Song

Shanghai Institute for Minimally Invasive Therapy, School of Health Science and Engineering, University of Shanghai for Science and Technology, Shanghai 200093, China.

Address correspondence to: Lin Mao, Shanghai Institute for Minimally Invasive Therapy, School of Health Science and Engineering, NO.516 Jungong Road, Shanghai 200093, China. Tel: +86-21-55572159. E-mail: linmao@usst.edu.cn.

DOI：https://doi.org/10.61189/392182sdzooq

Received December 31, 2024; Accepted February 19, 2025; Published March 31, 2025

Highlights

●Three types of circular electrodes with varying thicknesses were designed to achieve weight reduction, accompanied by support structures for intestinal tissue on both sides.

●The mechanical properties of the three electrode configurations were systematically compared to determine the optimal thickness for welding.

●In vitro tissue experiments successfully welded the tissue, identifying the optimal welding parameters.

Research Article |Published on: 31 March 2025

[Progress in Medical Devices] 2025; 3 (1): 12-20