Zhihong Wang

Dr. Zhihong Wang


Research Focus & Metrics

Academic Career Length

15Years (2012-Present)

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

Representative Publications

View All Publications
23 May 2025 in IEEE Journal of Biomedical and Health Informatics
Verified
Journal Article
Closed access
Songlu LinSonglu Lin
Songlu Lin
ZWZhihong Wang
Zhihong Wang
HGHans van Gorp
Hans van Gorp
MXMengzhu Xu
Mengzhu Xu
MGMerel M van Gilst
Merel M van Gilst
XLXi Long
Xi Long
Automatic sleep staging from single-channel electroencephalography (EEG) using artificial intelligence (AI) is emerging as an alternative to costly and time-consuming manual scoring using multi-channel polysomnography. However, current AI methods, mainly deep learning models such as convolutional neural network (CNN) and long short-term memory (LSTM), struggle to detect the N1 sleep stage, which is challenging due to its rarity and ambiguous nature compared to other stages. Here we propose SSC-SleepNet, an automatic sleep staging algorithm aimed at improving the learning of N1 sleep. SSC-SleepNet employs a pseudo-Siamese neural network architecture owing to its capability in one- or few-shot learning with contrastive loss. Which we selected due to its strong capability in one- or few-shot learning with a contrastive loss function. SSC-SleepNet consists of two branches of neural networks: a squeeze-and-excitation residual network branch and a CNN-LSTM branch. These two branches are used to generate latent features of the EEG epoch. The adaptive loss function of SSC-SleepNet uses a weighing factor to combine weighted cross-entropy loss and focal loss to specifically address the class imbalance issue inherent in sleep staging. The proposed new loss function dynamically assigns a higher penalty to misclassified N1 sleep stages, which can improve the model's learning capability for this minority class. Four datasets were used for sleep staging experiments. In the Sleep-EDF-SC, Sleep-EDF-X, Sleep Heart Health Study, and Haaglanden Medisch Centrum datasets, SSC-SleepNet achieved macro F1-scores of 84.5%, 89.6%, 89.5%, and 85.4% for all sleep stages, and N1 sleep stage F1-scores of 60.2%, 58.3%, 57.8%, and 55.2%, respectively. Our proposed deep learning model outperformed the most existing models in automatic sleep staging using single-channel EEG signals. In particular, N1 detection performance has been markedly improved compared to the state-of-art models.

21 August 2015 in Instrumentation Science & Technology
Verified
Journal Article
Closed access
JWJingru Wang
Jingru Wang
JLJie Liu
Jie Liu
GLGuangda Liu
Guangda Liu
ZWZhihong Wang
Zhihong Wang
Scanning near-infrared (NIR) spectrometers commonly employ analog lock-in amplifiers for demodulation that require complex circuitry and are sensitive to electromagnetic interference. Here, a new sampling system based on digital phase-sensitive detection is reported that used a quadrature sampling algorithm to establish the sampling frequency as an integral multiple of four times the optical signal frequency. A complex programmable logic device was used to control the analog-to-digital converter. A digital signal processor was used to receive data, complete the quadrature calculations, and transfer the results to a computer. Consequently, the circuitry was simplified with reduced electromagnetic interference. Spectra were obtained using instrumentation based on analog and digital sensitive detection that showed the latter provided more accurate results. The optimum signal-to-noise ratio of 66.34 dB was higher than the value obtained by the analog system.

1 September 2013
Verified
Conference Paper
Closed access
JZJing Zhao
Jing Zhao
MLMing Li
Ming Li
ZWZhihong Wang
Zhihong Wang
WXWenbo Xu
Wenbo Xu
To keep the balance between the global search and local search, a novel binary quantum-behaved particle swarm optimization algorithm with comprehensive learning and cooperative approach (CCBQPSO) is presented. In the proposed algorithm, all the particles' personal best position can participate in updating the local attractor firstly. Then all the particles' previous personal best position and swarm's global best position are performed in each dimension of the solution vector. Five test functions are used to test the performance of CCBQPSO. The results of experiment show that the proposed technique can increase diversity of swarm and converge more rapidly than other binary algorithms.

1 February 2018 in Review of Scientific Instruments
Verified
Journal Article
Closed access
PDF file
FZFudong Zhang
Fudong Zhang
CCChen Chen
Chen Chen
JLJie Liu
Jie Liu
ZWZhihong Wang
Zhihong Wang
This study focuses on improving the accuracy of a grating monochromator on the basis of the grating diffraction equation in combination with an analysis of the mechanical transmission relationship between the grating, the sine bar, and the screw of the scanning mechanism. First, the relationship between the mechanical error in the monochromator with the sine drive and the wavelength error is analyzed. Second, a mathematical model of the wavelength error and mechanical error is developed, and an accurate wavelength calibration method based on the sine bar’s length adjustment and error compensation is proposed. Based on the mathematical model and calibration method, experiments using a standard light source with known spectral lines and a pre-adjusted sine bar length are conducted. The model parameter equations are solved, and subsequent parameter optimization simulations are performed to determine the optimal length ratio. Lastly, the length of the sine bar is adjusted. The experimental results indicate that the wavelength accuracy is ±0.3 nm, which is better than the original accuracy of ±2.6 nm. The results confirm the validity of the error analysis of the mechanical system of the monochromator as well as the validity of the calibration method.

20 July 2025 in Applied Sciences
Verified
Journal Article
Open access
PDF file
Songlu LinSonglu Lin
Songlu Lin
RTRenzheng Tang
Renzheng Tang
YWYuzhe Wang
Yuzhe Wang
ZWZhihong Wang
Zhihong Wang
Accurate sleep staging and sleep-disordered breathing (SDB) severity prediction are critical for the early diagnosis and management of sleep disorders. However, real-world polysomnography (PSG) data often suffer from modality heterogeneity, label scarcity, and non-independent and identically distributed (non-IID) characteristics across institutions, posing significant challenges for model generalization and clinical deployment. To address these issues, we propose a federated multi-task learning (FMTL) framework that simultaneously performs sleep staging and SDB severity classification from seven multimodal physiological signals, including EEG, ECG, respiration, etc. The proposed framework is built upon a hybrid deep neural architecture that integrates convolutional layers (CNN) for spatial representation, bidirectional GRUs for temporal modeling, and multi-head self-attention for long-range dependency learning. A shared feature extractor is combined with task-specific heads to enable joint diagnosis, while the FedAvg algorithm is employed to facilitate decentralized training across multiple institutions without sharing raw data, thereby preserving privacy and addressing non-IID challenges. We evaluate the proposed method across three public datasets (APPLES, SHHS, and HMC) treated as independent clients. For sleep staging, the model achieves accuracies of 85.3% (APPLES), 87.1% (SHHS_rest), and 79.3% (HMC), with Cohen’s Kappa scores exceeding 0.71. For SDB severity classification, it obtains macro-F1 scores of 77.6%, 76.4%, and 79.1% on APPLES, SHHS_rest, and HMC, respectively. These results demonstrate that our unified FMTL framework effectively leverages multimodal PSG signals and federated training to deliver accurate and scalable sleep disorder assessment, paving the way for the development of a privacy-preserving, generalizable, and clinically applicable digital sleep monitoring system.

26 November 2023
Verified
Conference Paper
Closed access
SLSonglu Lin
Songlu Lin
MYMeicheng Yang
Meicheng Yang
YWYuzhe Wang
Yuzhe Wang
ZWZhihong Wang
Zhihong Wang