Sio Hang Pun

Dr. Sio Hang Pun


Research Focus & Metrics

Academic Career Length

22Years (2005-Present)

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

Representative Publications

View All Publications
4 February 2022 in IEEE Journal of Biomedical and Health Informatics
Verified
Journal Article
Closed access
JLJiawen Li
Jiawen Li
SBShovan Barma
Shovan Barma
PMPeng Un Mak
Peng Un Mak
FCFei Chen
Fei Chen
CLCheng Li
Cheng Li
Sio Hang PunSio Hang Pun
Sio Hang Pun
Recently, electroencephalography (EEG) signals have shown great potential for emotion recognition. Nevertheless, multichannel EEG recordings lead to redundant data, computational burden, and hardware complexity. Hence, efficient channel selection, especially single-channel selection, is vital. For this purpose, a technique termed brain rhythm sequencing (BRS) that interprets EEG based on a dominant brain rhythm having the maximum instantaneous power at each 0.2 s timestamp has been proposed. Then, dynamic time warping (DTW) is used for rhythm sequence classification through the similarity measure. After evaluating the rhythm sequences for the emotion recognition task, the representative channel that produces impressive accuracy can be found, which realizes single-channel selection accordingly. In addition, the appropriate time segment for emotion recognition is estimated during the assessments. The results from the music emotion recognition (MER) experiment and three emotional datasets (SEED, DEAP, and MAHNOB) indicate that the classification accuracies achieve 70–82% by single-channel data with a 10 s time length. Such performances are remarkable when considering minimum data sources as the primary concerns. Furthermore, the individual characteristics in emotion recognition are investigated based on the channels and times found. Therefore, this study provides a novel method to solve single-channel selection for emotion recognition.

15 June 2022 in IEEE Access
Verified
Journal Article
Open access
NYNing Yang
Ning Yang
DLDe Feng Liu
De Feng Liu
TLTao Liu
Tao Liu
THTianyuan Han
Tianyuan Han
PZPingyue Zhang
Pingyue Zhang
JZJianguo Zhang
Jianguo Zhang
Parkinson’s disease (PD) is a nervous disorder that can cause motor impairment. PD severity assessment based on a series of motor movements illustrated in Unified Parkinson’s Disease Rating Scale (UPDRS) is an important part of clinical PD diagnosis. However, the current quantifying method heavily relies on human observation, which is time-consuming and subjective. Therefore, automatic severity estimation stemming from machine learning methods is receiving an increasing amount of research attention. However, these advances are still limited by data availability and interpretability. In this paper, we release a large PD motor dataset of over 300 real PD patients collected under doctors’ instructions and propose a pipeline to automatically quantify the motor severity of PD in finger tapping and postural stability. These two selected movements are representative of local and global motor control, exhibiting great clinical importance. The pipeline contains three-stage: pose estimation, domain knowledge extraction, and classification stage. The pose estimation uses deep-learning-based methods to extract 21 and 17 key points for finger tapping and postural stability respectively. The domain knowledge extraction stage extracts several explicit features pre-defined by experienced neuro-physicians. Finally, a classifier is trained to infer PD severity under MDS-UPDRS. To combine deep-learning-based features from pose estimation and domain features from the expert, the pipeline achieves a better trade-off between the model efficiency and clinical interpretability. Experiments show that our method achieves a micro average f1-score of 88%, 84%, and 84%, respectively on left finger tapping, right finger tapping, and postural stability, outperforming previous methods by a large margin. In addition, involving expert knowledge in the feature extraction stage greatly improves our model’s interpretability, which is essential in automatic PD detection.

12 January 2018 in IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Control
Verified
Journal Article
Open access
Sio Hang PunSio Hang Pun
Sio Hang Pun
YYYuanyu Yu
Yuanyu Yu
JZJian Zhang
Jian Zhang
JWJiujiang Wang
Jiujiang Wang
CCChing-Hsiang Cheng
Ching-Hsiang Cheng
PMPeng Un Mak
Peng Un Mak
Among the biomedical imaging modalities, photoacoustic computed tomography (PACT) is one of the emerging hybrid techniques in recent years. In designing the PACT imaging system, a finite bandwidth transducer is one of the limited factors for the overall performance. As the target size is inversely proportional to the dominant frequency components of the generated photoacoustic (PA) signal, a broad bandwidth transducer is desired for different scales imaging. In the current study, a monolithic multi-band CMUT array was designed and fabricated for the reception of the wide band PA signals so as to provide high resolution images with high frequency CMUT arrays and present the high signal-to-noise-ratio major structure with low frequency CMUT arrays. To demonstrate its performance, a phantom experiment was conducted to show and evaluate the various qualities of multi-resolution images. In addition, an in vivo mouse model experiment was also carried out for revealing the multi-scale photoacoustic imaging capability with the multi-band CMUTs on biological tissues. From the obtained results, the images from different CMUT arrays could show the structures of the mouse brain in different scales. In addition, the images from the high frequency CMUT arrays were able to reveal the major blood vasculatures, whereas, the images from low frequency CMUT arrays showed the gross macroscopic anatomy of the brain with higher contrast.

14 March 2022 in IEEE Sensors Journal
Verified
Journal Article
Closed access
SASaeed Akbarzadeh
Saeed Akbarzadeh
HGHemant Ghayvat
Hemant Ghayvat
Chen ChenChen Chen
Chen Chen
XZXian Zhao
Xian Zhao
SHStephanie Hosier
Stephanie Hosier
Wei ChenWei Chen
Wei Chen
Tactile sensors are among the most important devices used in industrial and biomedical fields. Sensors’ profiles are significantly affected by their structures and material used. This article presents a robust, low-cost, low noise, accurate and simple fabrication capacitive tactile sensor as a single taxel fabricated on foam. This highly scalable design provides excellent noise immunity, accuracy, and due to a unique printable elastic conductor, it is flexible and stretchable with more than 200% strain. Furthermore, the taxel is based on the capacitive Wheatstone bridge. As a result, noise immunity and stability in case of temperature fluctuation is accomplished. Additionally, the sensor’s innovative, simple fabrication, made of Polyurethane foam and printable elastic conductor, allows the system to adapt and achieve relevant results necessary for the purpose of the sensor’s application. Therefore, the proposed sensor has potential applications in industrial and biomedical contexts, such as sleep monitoring, etc.

19 December 2023 in IEEE Transactions on Neural Networks and Learning Systems
Verified
Journal Article
Closed access
WYWei Yu
Wei Yu
NYNing Yang
Ning Yang
ZWZhijiong Wang
Zhijiong Wang
HLHung Chun Li
Hung Chun Li
AZAnguo Zhang
Anguo Zhang
Sio Hang PunSio Hang Pun
Sio Hang Pun
In this article, we proposed a novel fault-tolerant control scheme for quadrotor unmanned aerial vehicles (UAVs) based on spiking neural networks (SNNs), which leverages the inherent features of neural network computing to significantly enhance the reliability and robustness of UAV flight control. Traditional control methods are known to be inadequate in dealing with complex and real-time sensor data, which results in poor performance and reduced robustness in fault-tolerant control. In contrast, the temporal processing, parallelism, and nonlinear capacity of SNNs enable the fault-tolerant control scheme to process vast amounts of sensory data with the ability to accurately identify and respond to faults. Furthermore, SNNs can learn and adjust to new environments and fault conditions, providing effective and adaptive flight control. The proposed SNN-based fault-tolerant control scheme demonstrates significant improvements in control accuracy and robustness compared with conventional methods, indicating its potential applicability and suitability for a range of UAV flight control scenarios.

4 November 2019 in IEEE Access
Verified
Journal Article
Open access
JLJiawen Li
Jiawen Li
SBShovan Barma
Shovan Barma
PMPeng Un Mak
Peng Un Mak
Sio Hang PunSio Hang Pun
Sio Hang Pun
MVMang I Vai
Mang I Vai
A technique based on five brain rhythms (δ, θ, α, β, and γ) presented in the sequence for analyzing Electroencephalography (EEG) signals has been proposed. First, the production of the sequence has been accomplished by selecting the prominent brain rhythm having the maximum instantaneous power at specific timestamp consecutively throughout the EEG. To this purpose, the reassigned smoothed pseudo Wigner-Ville distribution (RSPWVD) has been employed. Then, in order to verify the proposed technique and evaluate its performance, a case study of seizure detection has been implemented. As experimental validation, 93 patients from the Karunya database have been investigated. Moreover, to characterize the brain rhythm sequence for seizure detection, two additional indices derived from the power discharge and synchronous behavior have been applied. Results show that the particular rhythm pattern during the seizure is usually one type (either δ, θ, or α) and it is subject-dependent. Hence, by focusing on the changes of such particular rhythm through the two indices, the time-related occurrences of seizures can be determined in detail. Meanwhile, the representative channels for seizure detection can be found by studying the similarity of sequences, which are helpful to reduce the number of applied channels. Finally, the proposed technique provides an accuracy of 98.9%, which demonstrates it is competent to detect the appearances of abnormal seizures from the EEG signals reliably. Consequently, the brain rhythm sequencing could open a new way to interpret and characterize the EEG in various applications such as for epileptic patients.