Li Liu

Mr. Li Liu

orcid

Research Focus & Metrics

Academic Career Length

6Years (2021-Present)

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

Representative Publications

View All Publications
3 August 2021 in IEEE Sensors Journal
Verified
Journal Article
Closed access
LLLi Liu
Li Liu
XZXianghao Zhan
Xianghao Zhan
RWRumeng Wu
Rumeng Wu
XGXiaoqing Guan
Xiaoqing Guan
ZWZhan Wang
Zhan Wang
GLGuang Li
Guang Li
Electronic nose has been proven effective in alternative herbal medicine classification, but due to the nature of supervised learning, previous research heavily relies on the labelled training data, which are time-costly and labor-intensive to collect. To alleviate the critical dependency on the training data in real-world applications, this study aims to improve classification accuracy via data augmentation strategies. The effectiveness of five data augmentation strategies under different training data inadequacy are investigated in two scenarios: the noise-free scenario where different availabilities of unlabelled data were considered, and the noisy scenario where different levels of Gaussian noises and translational shift were added to represent sensor drift. The five augmentation strategies, namely noise-adding data augmentation, semi-supervised learning, classifier-based online learning, Inductive Conformal Prediction (ICP) online learning and our novel ensemble ICP online learning (EICP) proposed in this study, are compared against supervised learning baselines, with Linear Discriminant Analysis (LDA) and Support Vector Machine (SVM) as the classifiers. This study provides a systematic analysis of different augmentation strategies. Our novel strategy, EICP, outperforms the others by showing non-decreasing classification accuracy on all tasks and a significant improvement on most simulated tasks (25 out of 36 tasks, ${p} \leq 0.05$ ), which demonstrated both effectiveness and robustness in boosting the classification model generalizability. This strategy can be employed in other machine learning applications.

29 May 2023 in IEEE Transactions on Instrumentation and Measurement
Verified
Journal Article
Closed access
HGHan Gao
Han Gao
SZShuo Zhao
Shuo Zhao
HLHuiyan Li
Huiyan Li
LLLi Liu
Li Liu
HWHengyang Wang
Hengyang Wang
GLGuang Li
Guang Li
Taste sensation recognition is a keystone for taste-related Brain-Computer Interface (BCI). A commonly used measurement of brain activity in response to specific stimulation is through electroencephalography (EEG) signals. However, it remains challenging to develop accurate and generalizable EEG-based measurement for human taste sensation. This paper proposes EEG-MSRNet, a novel fully convolutional neural network for EEG-based classification of basic taste sensations (blank, sour, sweet, bitter, salty, umami). Firstly, a multi-scale temporal convolution operation with residual learning is designed to extract features in different frequencies from the down-sampled EEG signals. Subsequently, a multi-scale spatial convolution operation represents the features in a cross-channel manner. Finally, a convolutional layer and global average pooling (GAP) layer are introduced to make predictions with the feature representation instead of the commonly used fully connected layers for classification. An experimental procedure is developed to acquire the EEG signals under taste stimulation. Comparison experiments and ablation studies have proved the stable and generalizable recognition performance of EEG-MSRNet on our self-collected EEG dataset. The results suggest that our EEG-based system with EEG-MSRNet is effective and generalizable for taste sensation recognition, which provides a powerful measurement for taste-related BCI such as taste disorder diagnosis and virtual taste.

18 May 2022 in IEEE Transactions on Instrumentation and Measurement
Verified
Journal Article
Closed access
HWHengyang Wang
Hengyang Wang
XZXianghao Zhan
Xianghao Zhan
LLLi Liu
Li Liu
AUAsif Ullah
Asif Ullah
HLHuiyan Li
Huiyan Li
GLGuang Li
Guang Li
Human taste sensation can be qualitatively described with surface electromyography. However, the pattern recognition models trained on one subject (the source domain) do not generalize well on other subjects (the target domain). To improve the generalizability and transferability of taste sensation models developed with sEMG data, two methods were innovatively applied in this study: domain regularized component analysis (DRCA) and conformal prediction with shrunken centroids (CPSC). The effectiveness of these two methods was investigated independently in an unlabeled data augmentation process with the unlabeled data from the target domain, and the same crossuser adaptation pipeline was conducted on six subjects. The results show that DRCA improved the classification accuracy on six subjects (p < 0.05), compared with the baseline models trained only with the source domain data, while CPSC did not guarantee the accuracy improvement. Furthermore, the combination of DRCA and CPSC presented statistically significant improvement (p < 0.05) in classification accuracy on six subjects. The proposed strategy of combining DRCA and CPSC showed its effectiveness in addressing the cross-user data distribution drift in sEMGbased taste sensation recognition application. It also shows the potential for more cross-user adaptation applications.

3 January 2022 in IEEE Transactions on Instrumentation and Measurement
Verified
Journal Article
Closed access
LLLi Liu
Li Liu
XZXianghao Zhan
Xianghao Zhan
XYXikai Yang
Xikai Yang
XGXiaoqing Guan
Xiaoqing Guan
RWRumeng Wu
Rumeng Wu
GLGuang Li
Guang Li
In machine learning applications, the reliability of predictions is significant for assisted decision and risk control. As an effective framework to quantify the prediction reliability, conformal prediction (CP) was commonly developed with $k$ -nearest neighbor, known as CPKNN. However, the conventional CPKNN suffers from high variance and bias, and long computational time as the feature dimensionality increases. To address these limitations, a new CP framework-CP with shrunken centroids (CPSC) is proposed. It regularizes the class centroids to attenuate the irrelevant features and shrink the sample space for predictions and reliability quantification. We have also compared the CPSC with CP based on SVM, LightGBM, RF on the herbal medicine dataset, with the electronic nose as a case and assessed them in two tasks: 1) off-line prediction: the training set was fixed, and the accuracy on the testing set was evaluated and 2) online prediction with data augmentation: CPKNN and CPSC filtered unlabeled data to augment the training data based on the prediction reliability, and the final accuracy of the testing set was compared. The result shows that CPSC significantly outperformed CPKNN in both two tasks: 1) CPSC reached a significantly higher accuracy with lower computation cost, and with the same credibility output, CPSC generally achieved higher accuracy and 2) the data augmentation process with CPSC robustly manifested a statistically significant improvement in prediction accuracy with different reliability thresholds, and the augmented data were more balanced in classes. This novel CPSC provides higher prediction accuracy and better reliability quantification, which can assist in reliable decision support.

20 October 2021 in Sensors
Verified
Journal Article
Open access
PDF file
HWHengyang Wang
Hengyang Wang
DLDongcheng Lu
Dongcheng Lu
LLLi Liu
Li Liu
HGHan Gao
Han Gao
RWRumeng Wu
Rumeng Wu
GLGuang Li
Guang Li
A novel approach to quantitatively recognize the intensity of primary taste stimuli was explored based on surface electromyography (sEMG). We captured sEMG samples under stimuli of primary taste with different intensities and quantitatively recognized preprocessed samples with Support Vector Machine (SVM). The feasibility of quantitatively recognizing the intensity of Sour, Bitter, and Salty was verified. The sEMG signals were acquired under the stimuli of citric acid (aq), sucrose (aq), magnesium chloride (aq), sodium chloride (aq), and sodium glutamate (aq) with different concentrations, for five types of primary tastes: Sour, Sweet, Bitter, Salty, and Umami, whose order was fixed in this article. The acquired signals were processed with a method called Quadratic Variation Reduction to remove baseline wandering, and an adaptive notch to remove power frequency interference. After extracting 330 features for each sample, an SVM regressor with five-fold cross-validation was performed and the model reached R2 scores of 0.7277, 0.1963, 0.7450, 0.7642, and 0.5055 for five types of primary tastes, respectively, which manifested the feasibilities of the quantitative recognitions of Sour, Bitter, and Salty. To explore the facial responses to taste stimuli, we summarized and compared the muscle activities under stimuli of different taste types and taste intensities. To further simplify the model, we explored the impact of feature dimensionalities and optimized the feature combination for each taste in a channel-wise manner, and the feature dimensionality was reduced from 330 to 210, 120, 210, 260, 170 for five types of primary tastes, respectively. Lastly, we analyzed the model performance on multiple subjects and the relation between the model’s performance and the number of experiment subjects. This study can provide references for further research and applications on taste stimuli recognition with sEMG.

8 May 2023 in IEEE Transactions on Instrumentation and Measurement
Verified
Journal Article
Closed access
HLHuiyan Li
Huiyan Li
YLYihao Liang
Yihao Liang
HGHan Gao
Han Gao
LLLi Liu
Li Liu
YWYou Wang
You Wang
GLGuang Li
Guang Li
Silent Speech Interface (SSI) based on neuromuscular signals has become popular in Human-Computer Interaction. Articulatory neuromuscular activity can be captured with surface electromyography (sEMG). However, in the absence of acoustic sounds, it remains challenging to convert the articulatory neuromuscular signals of silent speakers into corresponding audio. This paper proposes a Mandarin-based SSI, a novel converter between articulatory neuromuscular from silent speakers and audio signals, by providing an auxiliary set of sEMG and audio pairs from vocal speakers. The proposed method is based on a hybrid framework combining convolutional neural networks and Transformer to utilize both global and local information. We experimentally validate the feasibility of the proposed method by obtaining an average objective character error rate (CER) of 10.69% using an automatic speech recognition (ASR) evaluation tool with four silent speakers. The results show that our Mandarin-based SSI with vocal speaker assistance facilitates the conversion from sEMG into audio in the cross-subject scenario.