Dongyuan Liu

Dr. Dongyuan Liu


Research Focus & Metrics

Academic Career Length

25Years (2002-Present)

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

Representative Publications

View All Publications
25 May 2023 in Biomedical Optics Express
Verified
Journal Article
Closed access
YZYao Zhang
Yao Zhang
DLDongyuan Liu
Dongyuan Liu
TLTieni Li
Tieni Li
PZPengrui Zhang
Pengrui Zhang
ZLZhiyong Li
Zhiyong Li
FGFeng Gao
Feng Gao
Functional near-infrared spectroscopy (fNIRS) is increasingly used to investigate different mental tasks for brain-computer interface (BCI) control due to its excellent environmental and motion robustness. Feature extraction and classification strategy for fNIRS signal are essential to enhance the classification accuracy of voluntarily controlled BCI systems. The limitation of traditional machine learning classifiers (MLCs) lies in manual feature engineering, which is considered as one of the drawbacks that reduce accuracy. Since the fNIRS signal is a typical multivariate time series with multi-dimensionality and complexity, it makes the deep learning classifier (DLC) ideal for classifying neural activation patterns. However, the inherent bottleneck of DLCs is the requirement of substantial-scale, high-quality labeled training data and expensive computational resources to train deep networks. The existing DLCs for classifying mental tasks do not fully consider the temporal and spatial properties of fNIRS signals. Therefore, a specifically-designed DLC is desired to classify multi-tasks with high accuracy in fNIRS-BCI. To this end, we herein propose a novel data-augmented DLC to accurately classify mental tasks, which employs a convolution-based conditional generative adversarial network (CGAN) for data augmentation and a revised Inception-ResNet (rIRN) based DLC. The CGAN is utilized to generate class-specific synthetic fNIRS signals to augment the training dataset. The network architecture of rIRN is elaborately designed in accordance with the characteristics of the fNIRS signal, with serial multiple spatial and temporal feature extraction modules (FEMs), where each FEM performs deep and multi-scale feature extraction and fusion. The results of the paradigm experiments show that the proposed CGAN-rIRN approach improves the single-trial accuracy for mental arithmetic and mental singing tasks in both the data augmentation and classifier, as compared to the traditional MLCs and the commonly used DLCs. The proposed fully data-driven hybrid deep learning approach paves a promising way to improve the classification performance of volitional control fNIRS-BCI.

24 June 2019 in IEEE Access
Verified
Journal Article
Open access
DLDongyuan Liu
Dongyuan Liu
BWBingyuan Wang
Bingyuan Wang
TPTiantian Pan
Tiantian Pan
JLJiao Li
Jiao Li
ZQZhuanping Qin
Zhuanping Qin
GFGao Feng
Gao Feng
It is well acknowledged that the functional near-infrared spectroscopic (fNIRS) imaging of brain functions can be quantitatively enhanced by diffuse optical tomography (DOT). Unlike the widely-used conventional optical topography (optical mapping), an fNIRS-DOT instrument needs to possess both high sensitivity and large dynamic-range so that measurements at multiple source-detector separations are reliably and synchronously collected for improved depth-resolution, quantitation, and superficial signal reduction, the performances particularly crucial to the occipital cortex probing. For the goal, we implemented a novel 3-wavelength, 240-channel (20 sources and 12 detectors) continuous-wave fNIRS-DOT instrument. The system features the square-wave modulation lock-in photon-counting scheme that combines the multi-channel parallelization of the lock-in detection, the ultra-high sensitivity, and unprecedented dynamic-range of the photon-counting technique, as well as the reduced complexity of the hardware architecture. We have systematically investigated the key specifications of the instrument such as the detection linearity, stability, channel crosstalks and so on, under the practically-viable conditions of the incident power and signal-to-noise ratio, and accordingly proposed wavelength-switched multiple-field illumination paradigms for both the non-overlapping and overlapping DOT measurements, respectively. Finally, DOT experiments using the semi-three-dimensional image reconstruction has been conducted on a phantom with a practically low-contrasted and deep-sit absorption target, to evaluate the overall imaging performance of the system and to demonstrate the superiority of the model-based image reconstruction over the topographical mapping.

6 February 2024 in Optics Express
Verified
Journal Article
Closed access
WBWenxing Bai
Wenxing Bai
YDYihan Dong
Yihan Dong
YZYaru Zhang
Yaru Zhang
YWYe Wu
Ye Wu
MDMai Dan
Mai Dan
GFGao Feng
Gao Feng
We present a wide-field illumination time-domain (TD) diffusion optical tomography (DOT) for three-dimensional (3-D) reconstruction within a shallow region under the illuminated surface of the turbid medium. The methodological foundation is laid on the single-pixel spatial frequency domain (SFD) imaging that facilitates the adoption of the well-established time-correlated single-photon counting (TCSPC)-based TD detection and generalized pulse spectrum techniques (GPST)-based reconstruction. To ameliorate the defects of the conventional diffusion equation (DE) in the forward modeling of TD-SFD-DOT, mainly the low accuracy in the near-field region and in profiling early-photon migration, we propose a modified model employing the time-dependent δ-P1 approximation and verify its improved accuracy in comparison with both the Monte Carlo and DE-based ones. For a simplified inversion process, a modified GPST approach is extended to TD-SFD-DOT that enables the effective separation of the absorption and scattering coefficients using a steady-state equivalent strategy. Furthermore, we set up a single-pixel TD-SFD-DOT system that employs the TCSPC-based TD detection in the SFD imaging framework. For assessments of the reconstruction approach and the system performance, phantom experiments are performed for a series of scenarios. The results show the effectiveness of the proposed methodology for rapid 3-D reconstruction of the absorption and scattering coefficients within a depth range of about 5 mean free pathlengths.

24 September 2021 in IEEE Transactions on Computational Imaging
Verified
Journal Article
Closed access
DLDongyuan Liu
Dongyuan Liu
YZYao Zhang
Yao Zhang
LBLu Bai
Lu Bai
PZPengrui Zhang
Pengrui Zhang
GFGao Feng
Gao Feng
High-density (HD) diffuse optical tomography (DOT), as an advanced modality of functional near-infrared spectroscopy, is finding increasing applications in neuroimaging regime. However, it is a primary challenge that the superficial physiological interferences usually significantly contaminate the functional activation reconstruction. In addition, the random noises, majorly the photon-shot and instrumental ones, also cast negative influences on the measurements, and further distort the reconstructed image. To mitigate the adversities, we herein propose a combined scheme of two-layer semi-three-dimensional (S3D) reconstruction and multi-wavelength image fusion, which leverages a mathematical model with explicit physical significance, to suppress the physiological interferences and random noises in HD-DOT reconstruction, respectively. The approach is purely data-driven without additional auxiliary measurement, and comprised of two steps: First, the absorption perturbations are topographically reconstructed over both the scalp (SC) and cerebral-cortex (CC) layers using the two-layer S3D scheme, of which the superficial interferences are estimated from the SC reconstruction and adaptively filtered out from the CC one; Second, the interference-suppressed multi-wavelength CC-images are decomposed using the discrete wavelet transform, and fused at multi-resolutions into a mask for further removal of the random noises. We comprehensively validate the proposed scheme using simulations and phantom experiments, and demonstrate its sound effectiveness in suppressing the physiological interferences and random noises. The performance improvement rather than by more cycles or longer sampling time offers additional payoff: shorter measurement time or higher temporal resolution.

11 February 2023 in Journal of Biophotonics
Verified
Journal Article
Closed access
JAJingyi An
Jingyi An
QZQi Zhang
Qi Zhang
LZLimin Zhang
Limin Zhang
CLChenlu Liu
Chenlu Liu
DLDongyuan Liu
Dongyuan Liu
FGFeng Gao
Feng Gao

14 September 2023 in IEEE Transactions on Instrumentation and Measurement
Verified
Journal Article
Closed access
DLDongyuan Liu
Dongyuan Liu
PZPengrui Zhang
Pengrui Zhang
YZYao Zhang
Yao Zhang
TLTieni Li
Tieni Li
ZLZhiyong Li
Zhiyong Li
GFGao Feng
Gao Feng
With the aim of hyperscanning real-world interactions using functional near-infrared spectroscopy (fNIRS) technology, a distributed wearable multi-module system is implemented in the present study, enabling synchronous measurement of multiple subjects, wireless master-slave connections and lower power consumption. In order to achieve an integrated characteristic of miniaturization, movement freedom, high cost-effectiveness and hemodynamics profiling, the system is structured as a network of distributed fibreless optode modules, with each equipped with 10 dual-wavelength light emitting diodes and 4 monolithic silicon photodiodes with on-chip transimpedance amplifier. On-line digital lock-in detection based on multi-periodic reference-weighted counting algorithm is performed in each slave to separate the signals from the different source-detector combinations as well as the wavelengths. Phantom and breath-holding experiments are designed to assess the fundamental performances of the implemented system, and validate the good raw signal quality with the static measurement fluctuation of < 2%, dynamic optical range of > 72dB, and variation coefficient of ~ 5.39%, as 10 channels work simultaneously, as well as the feasible sensitivity for detecting task-evoked activation. Finally, the dynamic inter-brain synchrony is explored by simultaneously measuring hemodynamic responses of two subjects while they played a cooperation game. This preliminary task-oriented experiment suggests that, the proposed system has a capability of distilling the inter-brain activation patterns with acceptable temporal resolution, and bears high potential toward fNIRS-based social neuroimaging in real-world.