Jiaqi Xue

Jiaqi Xue

Auto-generated profile, based on publicly available information

Jiaqi Xue

Department of Biomedical Engineering City University of Hong Kong 83 Tat Chee Avenue, Kowloon Tong Hong Kong

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
10 April 2023 in ACS Applied Materials & Interfaces
Verified
Journal Article
Closed access
XZXiaoyang Zou
Xiaoyang Zou
JXJiaqi Xue
Jiaqi Xue
XLXiaoting Li
Xiaoting Li
CCColin Pak Yu Chan
Colin Pak Yu Chan
ZLZiqi Li
Ziqi Li
KLKing Wai Chiu Lai
King Wai Chiu Lai

8 September 2023 in Advanced Intelligent Systems
Verified
Journal Article
Open access
JXJiaqi Xue
Jiaqi Xue
KLKing Wai Chiu Lai
King Wai Chiu Lai
Currently, electromyography (EMG)‐driven lower limb control can be divided into computational intrinsic control and interactive extrinsic control based on the participation of neural information in the controllers. However, the former method lacks detailed measurements of the expected motion, whereas the latter is prone to errors in diverse movement situations. Based on this problem, a multi‐intent prediction scheme is proposed that analyzes both rhythmic locomotion and volitional movement intent for AI‐based intrinsic and extrinsic hybrid control (AIEC). A 1D residual shrinkage convolutional network is designed to extract EMG features. The motion state is continuously predicted with an accuracy of 91.66% and the angle completion is also estimated with of 0.9540 and 0.9456 for left and right knee angles, respectively. Additionally, the comparison test further indicates that the motion state classification is significantly improved in the multitask analysis compared with the single‐task approach. This work demonstrates and verifies a novel method in EMG studies that multi‐intent recognition not only compensates for the lack of information in the analysis of single rhythmic or volitional movement, but also enhances the comprehensive performance and makes AIEC feasible, which optimizes the current EMG‐driven control for ampler intent collection and more practical robotic control.

22 January 2024 in Advanced Intelligent Systems
Verified
Journal Article
Open access
JXJiaqi Xue
Jiaqi Xue
KLKing Wai Chiu Lai
King Wai Chiu Lai

14 May 2023
Verified
Conference Paper
Closed access
JXJiaqi Xue
Jiaqi Xue
XZXiaoyang Zou
Xiaoyang Zou
CCColin Pak Yu Chan
Colin Pak Yu Chan
KLKing Wai Chiu Lai
King Wai Chiu Lai
In recent years, the development of flexible and stretchable sensors has shown considerable potential in human information collection for wearable applications. With skin-fitting film electrodes, Electromyography (EMG) signals can be monitored stably and detected for effective control actuation in wearable systems. Traditionally in EMG-based activation, researchers usually focused on the design of EMG features, which is time-costing and difficult to always find the optimal combination. In this work, we have proposed a scheme to use convolutional neural network for complicated EMG feature extraction and accurate force classification. Four force levels were recognized by our model. The experimental result stated that the accuracy in each force level has reached 90.64%, 89.94%, 84.21% and 95.24%, respectively. In addition, the performance of our deep learning model has outperformed the traditional manual-feature-based methods, which utilized mean absolute value (MAV), waveform length (WL) and Willison amplitude (WAMP) for force identification. Actually, this work has verified the excellent effect of intelligent methods in EMG feature learning, and can be further applied in a real-time wearable system to promote its convenience and practicality.

5 December 2023
Verified
Conference Paper
Closed access
XZXiaoyang Zou
Xiaoyang Zou
XLXiaoting Li
Xiaoting Li
JXJiaqi Xue
Jiaqi Xue
CCColin Pak Yu Chan
Colin Pak Yu Chan
ZLZiqi Li
Ziqi Li
KLKing Wai Chiu Lai
King Wai Chiu Lai
The knee joint is one of the most important joints in our lower body, affecting humans’ lower limb activities. For instance, the knee joint can even bear three to four times the human body’s weight when running and climbing. The knee joint is prone to degenerative lesions for older people as they age, which significantly affects their daily lives. Therefore, long-term and real-time assessment of lower-limb motions in older people is crucial for their lower-limb health detection. In this work, we developed wearable knee sensors based on nanocomposite e-textile strain sensors for detecting knee joints’ range of motion (ROM) while older people perform lower-limb activities. A total of ten features were extracted to characterize their STS performance by detecting their knee joint range of motion through normal speed and fast speed sit-to-stand (STS). The elderly subjects were grouped by their age, gender, and body mass index (BMI) to assess the relationship of these three parameters with the extracted features. As a result, the stability, ROM, and gender of elderly subjects while performing fast speed STS had a significant relationship with age and BMI (P<0.05). The result demonstrates the generalizability of e-textile sensors for assessing lower-limb motions, which shows great potential for long-term and real-time detection of lower-limb health in older people.

5 December 2023
Verified
Conference Paper
Closed access
JXJiaqi Xue
Jiaqi Xue
XZXiaoyang Zou
Xiaoyang Zou
ZLZiqi Li
Ziqi Li
KLKing Wai Chiu Lai
King Wai Chiu Lai
Electromyography (EMG) is a bioelectrical signal to reflect human intention before actual motion occurs. EMG has been widely used in human-machine interaction such as robot control, rehabilitation and health monitoring. In this work, we have designed an intelligent approach for wrist motion classification based on EMG signals. Since commercial electrodes cannot maintain good contact with skin during deformation, we have utilized a new type of fabricated flexible electrodes. With these electrodes, high-quality signals can be acquired. And the machine learning methods have been utilized to classify the extracted feature sets. Four different feature conditions have been compared. In the condition of six EMG features, we have identified four wrist gestures including wrist flexion, extension, radial deviation, and ulnar deviation with the best accuracy of 92.26%.