Research

Research Interests

• Brain-computer Interface

• Pain

• Neuromodulation

• Artificial Intelligence

• Brain connectivity network

• Cognitive neuroscience

Current project

• Neural signatures of experimental pain in chronic pain patients

• Adaptive neuromodulation to chronic pain based on deep brain stimulation (DBS)

Previous projects

• Relative contributions of formants in Mandarin Chinese (Supervisor: Prof. Fei Chen): It was widely demonstrated that human perception to phoneme mostly depends on the first three formants in speech, and some studies proved that the first two formants contribute more than the third one respectively. In this study, the relative contributions of the first three formants in Mandarin were evaluated.

• Forensic Camera Model Identification Challenge (For Signal Processing Cup 2018, organized by Signal Processing Society, IEEE)

• The influencing mechanism among spatial, visual and auditory perceptions in virtual reality (Supervisor: Prof. Fei Chen, Sponsored by SUSTech SITP grant for CNY 10000)

• Development of a novel approach to neurophysiological pain assessment in unresponsive patients (Supervisor: Dr. Sebastian Halder and Dr. Elia Valentini)

  • Question 1: Which feature can be used to predict pain?  Due to the complexity of the brain's responses to pain, the neural marker predicting pain using EEG must involve the integration across brain regions or frequency bands. Among such features, functional connectivity from the alpha band is quite powerful and efficient to be extracted.

  • Question 2: How can we predict pain with machine learning? We developed a CNN model for this purpose, which can predict pain with an accuracy above 94% from non-pain conditions. More importantly, it can be applied across participants (users in the realistic scenario).

  • Question 3: Is it practical to apply the pain prediction model to unresponsive users? Though until now, the developed model cannot assess pain from unlabelled data, the functional connectivity from the alpha band has been proven as a marker of individual specificity as well. Fortunately, an adversarial model under development will enlighten this question, please look forward to that.

Publications

• Han, Y., Valentini, E., & Halder, S. (2023) Validation of Cross-Individual Pain Assessment with Individual Recognition Model from Electroencephalogram. 45th Annual International Conference of the IEEE Engineering in Medicine &Biology Society (EMBC). Sydney, Australia, July 24-27, 2023. 

• Han, Y., Valentini, E., & Halder, S. (2022) Classification of Tonic Pain Experience based on Phase Connectivity in the Alpha Frequency Band of the Electroencephalogram using Convolutional Neural Networks. 44th Annual International Conference of the IEEE Engineering in Medicine &Biology Society (EMBC), 3542-3545. Glasgow, UK, July 11-15, 2022. (https://ieeexplore.ieee.org/document/9871353)

• Han, Y., Ziebell, P., Riccio, A., & Halder, S. (2022) Two sides of the same coin: adaptation of BCIs to internal states with user-centered design and electrophysiological features. Brain-Computer Interfaces, 9(2), 102-114. (https://www.tandfonline.com/doi/full/10.1080/2326263X.2022.2041294)

• Han, Y., & Chen, F. (2019). Minimum Audible Movement Angle in Virtual Auditory Environment: Effect of Stimulus Frequency. IEEE 2nd International Conference on Multimedia Information Processing and Retrieval. (https://ieeexplore.ieee.org/document/8695358 )

• Han, Y., & Chen, F. (2017). Relative contributions of formants to the intelligibility of sine-wave sentences in Mandarin Chinese. Journal of the Acoustical Society of America, 141(6), EL495.  (https://asa.scitation.org/doi/10.1121/1.4983747 )