Qing Miao
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Qing Miao has a total of 10 co-authors. Here are the top co-authors by number of shared publications:
- Xianwei Zheng
School of Mathematics, Foshan University, Foshan, China
2 shared publications - Jiongqi Wang
College of Liberal Arts and Sciences, National University of Defense Technology, Changsha 410073, China
1 shared publication - Junqi Lin
School of Mathematics, Foshan University, Guangdong, Foshan 528000, P. R. China
1 shared publication Longtao Zhou
Guangzhou Yangchengtong Public Transportation Card Co. Ltd., Guangzhou 510080, China
1 shared publication- Weijie You
School of Mathematics, Foshan University, Guangdong, Foshan 528000, P. R. China
1 shared publication - Xutao Li
Shantou University
1 shared publication
Latest Publications
- 30 October 2024
- Published by World Scientific Pub Co Pte Ltd in International Journal of Wavelets, Multiresolution and Information Processing
- Vol. 23 (01)
DAB-LSTMNN: A novel respiratory disease diagnosis deep CNN framework based on attention scheme and BLSTM neural networks
Respiratory diseases have a significant impact on modern society, posing considerable public health risks. With the emergence of the COVID-19 pandemic, addressing respiratory issues has become increasingly urgent and important. Recently, artificial intelligence-based methods utilizing the acoustic recordings of suspected patients have shown promise in the diagnosis of various respiratory diseases, enabling localized treatment and containment of their spread. As the existing methods cannot efficiently extract subtle differences between the sound samples, thereby limiting their generalization ability. To improve the diagnosis accuracy, this paper proposes a novel multi-channel, multi-modal deep learning architecture based on the attention mechanism. The proposed framework combines a deep convolutional neural network (DCNN) with a bidirectional long short-term memory (BLSTM) network, and also utilizes the attention scheme to extract temporal and spectral features from different modalities of speech data (e.g. coughs, counting sounds and sustained vowel articulation). The proposed method effectively classifies COVID-19 patients, asthma patients and healthy individuals, with a test accuracy of 89.27 ± 0.1%, and an F1 score of 85.42 ± 0.2%. The experimental results validate the feasibility of our method, and also indicate that it is competitive with the existing deep networks.
- 30 October 2024
- Published by World Scientific Pub Co Pte Ltd in International Journal of Wavelets, Multiresolution and Information Processing
- Vol. 23 (01)
DAB-LSTMNN: A novel respiratory disease diagnosis deep CNN framework based on attention scheme and BLSTM neural networks
Respiratory diseases have a significant impact on modern society, posing considerable public health risks. With the emergence of the COVID-19 pandemic, addressing respiratory issues has become increasingly urgent and important. Recently, artificial intelligence-based methods utilizing the acoustic recordings of suspected patients have shown promise in the diagnosis of various respiratory diseases, enabling localized treatment and containment of their spread. As the existing methods cannot efficiently extract subtle differences between the sound samples, thereby limiting their generalization ability. To improve the diagnosis accuracy, this paper proposes a novel multi-channel, multi-modal deep learning architecture based on the attention mechanism. The proposed framework combines a deep convolutional neural network (DCNN) with a bidirectional long short-term memory (BLSTM) network, and also utilizes the attention scheme to extract temporal and spectral features from different modalities of speech data (e.g. coughs, counting sounds and sustained vowel articulation). The proposed method effectively classifies COVID-19 patients, asthma patients and healthy individuals, with a test accuracy of 89.27 ± 0.1%, and an F1 score of 85.42 ± 0.2%. The experimental results validate the feasibility of our method, and also indicate that it is competitive with the existing deep networks.
- 5 June 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Signal Processing Letters
- Vol. 31 (10709908) , 1630-1634
Spatiotemporal Gated Graph Transformer for EEG-Based Emotion Recognition
The availability of accurate and reliable electroencephalography (EEG) signal data makes emotion recognition feasible. In recent years, an increasing number of deep learning methods have been applied to emotion recognition tasks based on EEG signals, but they all have their own drawbacks, including the inability to simultaneously capture spatial and temporal information, the loss of spatial information, and even the improper selection of data scales. To address the above problems, this paper proposes a novel method for EEG-based emotion recognition named Spatiotemporal Gated Graph Transformer (SGGT). The method takes advantage of graph structure to enrich spatial information and uses a two-tower transformer to encode spatial information and temporal information, respectively. For the spatial feature extraction, we adopt four encoding methods to compensate for the shortcomings of traditional transformers in graph learning tasks. Different from the previous processing methods, our method uses graph pooling to compress the graph, forcing the model to learn the relationship between different time stamps. In addition, a gating mechanism is used to merge the two-tower transformer to realize the fusion of temporal and spatial features. Our method is validated on the SEED, SEED-IV and SEED-V datasets, and it outperforms the baseline methods.
- 5 June 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Signal Processing Letters
- Vol. 31 (10709908) , 1630-1634
Spatiotemporal Gated Graph Transformer for EEG-Based Emotion Recognition
The availability of accurate and reliable electroencephalography (EEG) signal data makes emotion recognition feasible. In recent years, an increasing number of deep learning methods have been applied to emotion recognition tasks based on EEG signals, but they all have their own drawbacks, including the inability to simultaneously capture spatial and temporal information, the loss of spatial information, and even the improper selection of data scales. To address the above problems, this paper proposes a novel method for EEG-based emotion recognition named Spatiotemporal Gated Graph Transformer (SGGT). The method takes advantage of graph structure to enrich spatial information and uses a two-tower transformer to encode spatial information and temporal information, respectively. For the spatial feature extraction, we adopt four encoding methods to compensate for the shortcomings of traditional transformers in graph learning tasks. Different from the previous processing methods, our method uses graph pooling to compress the graph, forcing the model to learn the relationship between different time stamps. In addition, a gating mechanism is used to merge the two-tower transformer to realize the fusion of temporal and spatial features. Our method is validated on the SEED, SEED-IV and SEED-V datasets, and it outperforms the baseline methods.
- Open Access
- 2 June 2022
- Published by Wiley in Journal of Control Science and Engineering
- Vol. 2022, 1-13
Beamforming Optimization for Information Enhancement Transmission in MISO SWIPT with NOMA System
This paper mainly considers a communication network with multiple-input single-output (MISO), in which nonorthogonal multiple access (NOMA) decoding technology is adopted, and wireless power-carrying technology is used to enhance the transmission quality of user information in the communication network. Through joint optimization of power splitting ratio and beamforming vector, we aim at minimizing transmission power (base station transmit power and secondary transmit power) while satisfying the quality-of-service requirement of all users and minimum power for secondary transmission. The problem is a nonconvex optimization program, and it is difficult to get the optimal value directly. First, in order to solve the problem for two-user case, the original problem is transformed into a semidefinite programming (SDP) problem, and then, the iterative updating algorithm is used to approximate the optimal value. When the number of users is greater than two, the original problem is transformed into a second-order cone problem, in which we deal with a sequence of second-order cone programming (SOCP). Results verify that the optimal value of the sequence of SOCPs is not increasing and converges to a local optimal value. Detailed simulation experiments indicate that the algorithm improves the performance of NOMA downlink beamforming under the condition of enhancing information and eliminating network interference.
- 2 June 2022
- Published by Wiley in Journal of Control Science and Engineering
- Vol. 2022, 1-13
Open AccessBeamforming Optimization for Information Enhancement Transmission in MISO SWIPT with NOMA System
This paper mainly considers a communication network with multiple-input single-output (MISO), in which nonorthogonal multiple access (NOMA) decoding technology is adopted, and wireless power-carrying technology is used to enhance the transmission quality of user information in the communication network. Through joint optimization of power splitting ratio and beamforming vector, we aim at minimizing transmission power (base station transmit power and secondary transmit power) while satisfying the quality-of-service requirement of all users and minimum power for secondary transmission. The problem is a nonconvex optimization program, and it is difficult to get the optimal value directly. First, in order to solve the problem for two-user case, the original problem is transformed into a semidefinite programming (SDP) problem, and then, the iterative updating algorithm is used to approximate the optimal value. When the number of users is greater than two, the original problem is transformed into a second-order cone problem, in which we deal with a sequence of second-order cone programming (SOCP). Results verify that the optimal value of the sequence of SOCPs is not increasing and converges to a local optimal value. Detailed simulation experiments indicate that the algorithm improves the performance of NOMA downlink beamforming under the condition of enhancing information and eliminating network interference.