Zhaoyang Huang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Zhaoyang Huang has a total of 41 co-authors. Here are the top co-authors by number of shared publications:
- Bian Ma
Beijing University of Technology
4 shared publications - Lijuan Duan
College of Computer Science, Beijing University of Technology, Beijing, China
4 shared publications - Yuanhua Qiao
School of Mathematics, Statistics and Mechanics, Beijing University of Technology, Beijing, China
4 shared publications Abdelkader Nasreddine Belkacem
Academy of Medical Engineering and Translational Medicine, Tianjin University, Tianjin, China
2 shared publicationsChao Chen
Academy of Medical Engineering and Translational Medicine, Tianjin University, Tianjin, China
2 shared publications- Dong Ming
Tianjin University
2 shared publications
Latest Publications
- 26 December 2024
- Published by Elsevier in Biomedical Signal Processing and Control
- Vol. 103
MMS-SleepNet: A knowledge-based multimodal and multiscale network for sleep staging
- 26 December 2024
- Published by Elsevier in Biomedical Signal Processing and Control
- Vol. 103
MMS-SleepNet: A knowledge-based multimodal and multiscale network for sleep staging
- 30 November 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Biomedical and Health Informatics
- Vol. 28 (3) , 1730-1741
Dual-Teacher Feature Distillation: A Transfer Learning Method for Insomniac PSG Staging
Insomnia is the most common sleep disorder linked with adverse long-term medical and psychiatric outcomes. Automatic sleep staging plays a crucial role in aiding doctors to diagnose insomnia disorder. Only a few studies have been conducted to develop automatic sleep staging methods for insomniacs, and most of them have utilized transfer learning methods, which involve pre-training models on healthy individuals and then fine-tuning them on insomniacs. Unfortunately, significant differences in feature distribution between the two subject groups impede the transfer performance, highlighting the need to effectively integrate the features of healthy subjects and insomniacs. In this paper, we propose a dual-teacher cross-domain knowledge transfer method based on the feature-based knowledge distillation to improve the performance of sleep staging for insomniacs. Specifically, the insomnia teacher directly learns from insomniacs and feeds the corresponding domain-specific features into the student network, while the health domain teacher guide the student network to learn domain-generic features. During the training process, we adopt the OFD (Overhaul of Feature Distillation) method to build the health domain teacher. We conducted the experiments to validate the proposed method, using the Sleep-EDF database as the source domain and the CAP-Database as the target domain. The results demonstrate that our method surpasses advanced techniques, achieving an average sleep staging accuracy of 80.56% on the CAP-Database. Furthermore, our method exhibits promising performance on the private dataset.
- 30 November 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Biomedical and Health Informatics
- Vol. 28 (3) , 1730-1741
Dual-Teacher Feature Distillation: A Transfer Learning Method for Insomniac PSG Staging
Insomnia is the most common sleep disorder linked with adverse long-term medical and psychiatric outcomes. Automatic sleep staging plays a crucial role in aiding doctors to diagnose insomnia disorder. Only a few studies have been conducted to develop automatic sleep staging methods for insomniacs, and most of them have utilized transfer learning methods, which involve pre-training models on healthy individuals and then fine-tuning them on insomniacs. Unfortunately, significant differences in feature distribution between the two subject groups impede the transfer performance, highlighting the need to effectively integrate the features of healthy subjects and insomniacs. In this paper, we propose a dual-teacher cross-domain knowledge transfer method based on the feature-based knowledge distillation to improve the performance of sleep staging for insomniacs. Specifically, the insomnia teacher directly learns from insomniacs and feeds the corresponding domain-specific features into the student network, while the health domain teacher guide the student network to learn domain-generic features. During the training process, we adopt the OFD (Overhaul of Feature Distillation) method to build the health domain teacher. We conducted the experiments to validate the proposed method, using the Sleep-EDF database as the source domain and the CAP-Database as the target domain. The results demonstrate that our method surpasses advanced techniques, achieving an average sleep staging accuracy of 80.56% on the CAP-Database. Furthermore, our method exhibits promising performance on the private dataset.
- 6 September 2023
- Published by Elsevier in Expert Systems with Applications
- Vol. 237
A federated semi-supervised automatic sleep staging method based on relationship knowledge sharing
- 6 September 2023
- Published by Elsevier in Expert Systems with Applications
- Vol. 237
A federated semi-supervised automatic sleep staging method based on relationship knowledge sharing
- Open Access
- 16 June 2023
- Published by Wiley in Alzheimer's & Dementia
- Vol. 19 (S2)
The value of heart rate variability in autonomic nervous function evaluation and diagnosis of fatal familial insomnia
Background: Fatal familial insomnia (FFI) is a rare autosomal dominant inherited prion disease characterized by prominent organic sleep‐related impairment accompanied by a series of symptoms with strong clinical heterogeneity in a rapid progression. As the most significant clinical feature and the potential diagnostic marker of FFI, progressive sympathetic symptoms were lack of a objective noninvasive quantitative method to detect.Method: From May 2013 to August 2020, nine patients with FFI, eight patients with Creutzfeldt‐jakob disease (CJD) and nine normal control (NC) with matched sex and age were recruited in this study. All the participants underwent a nocturnal video‐polysomnography included a lead II electrocardiography (ECG). Heart rate variability analysis with successive 5 min (in time domain methods, frequency domain methods and non‐linear measurement) was performed during quiet wake before sleep and in the following N2, N3 and REM of the first sleep cycle.Result: In FFI group, the continuous increased heart rate with the relatively decreased amplitude of the oscillations compared with NC and CJD group, revealed the axis of equilibrium shifts towards sympathetic activation. Furthermore, the RMSSD and HF during sleep stage in FFI were significantly lower than in NC group, which reflected the decreased activation of parasympathetic nervous system. The significant abnormal LF/HF ratio and SD1/SD2 ratio in combination with the lower values of correlation dimension D2 in FFI group supported its unbalanced autonomic nervous function.Conclusion: Dysfunctions of autonomic nerve system in patients with FFI may be affected by parasympathetic hypofunction partly except sympathetic activation, and HRV is expected to become a method in diagnosis for FFI.
- 16 June 2023
- Published by Wiley in Alzheimer's & Dementia
- Vol. 19 (S2)
Open AccessThe value of heart rate variability in autonomic nervous function evaluation and diagnosis of fatal familial insomnia
Background: Fatal familial insomnia (FFI) is a rare autosomal dominant inherited prion disease characterized by prominent organic sleep‐related impairment accompanied by a series of symptoms with strong clinical heterogeneity in a rapid progression. As the most significant clinical feature and the potential diagnostic marker of FFI, progressive sympathetic symptoms were lack of a objective noninvasive quantitative method to detect.Method: From May 2013 to August 2020, nine patients with FFI, eight patients with Creutzfeldt‐jakob disease (CJD) and nine normal control (NC) with matched sex and age were recruited in this study. All the participants underwent a nocturnal video‐polysomnography included a lead II electrocardiography (ECG). Heart rate variability analysis with successive 5 min (in time domain methods, frequency domain methods and non‐linear measurement) was performed during quiet wake before sleep and in the following N2, N3 and REM of the first sleep cycle.Result: In FFI group, the continuous increased heart rate with the relatively decreased amplitude of the oscillations compared with NC and CJD group, revealed the axis of equilibrium shifts towards sympathetic activation. Furthermore, the RMSSD and HF during sleep stage in FFI were significantly lower than in NC group, which reflected the decreased activation of parasympathetic nervous system. The significant abnormal LF/HF ratio and SD1/SD2 ratio in combination with the lower values of correlation dimension D2 in FFI group supported its unbalanced autonomic nervous function.Conclusion: Dysfunctions of autonomic nerve system in patients with FFI may be affected by parasympathetic hypofunction partly except sympathetic activation, and HRV is expected to become a method in diagnosis for FFI.
- Open Access
- 30 March 2023
- Published by Frontiers Media SA in Frontiers in Neuroscience
- Vol. 17, 1110320
A comparative analysis of sleep spindle characteristics of sleep-disordered patients and normal subjects
Spindles differ in density, amplitude, and frequency, and these variations reflect different physiological processes. Sleep disorders are characterized by difficulty in falling asleep and maintaining sleep. In this study, we proposed a new spindle wave detection algorithm, which was more effective compared with traditional detection algorithms such as wavelet algorithm. Besides, we recorded EEG data from 20 subjects with sleep disorders and 10 normal subjects, and then we compared the spindle characteristics of sleep-disordered subjects and normal subjects (those without any sleep disorder) to assess the spindle activity during human sleep. Specifically, we scored 30 subjects on the Pittsburgh Sleep Quality Index and then analyzed the association between their sleep quality scores and spindle characteristics, reflecting the effect of sleep disorders on spindle characteristics. We found a significant correlation between the sleep quality score and spindle density (p = 1.84 × 10−8, p-value p-value of 0.667, suggesting that the spindle frequency and sleep quality score were not significantly correlated. The p-value between the sleep quality score and spindle amplitude was 1.33 × 10−4, indicating that the mean amplitude of the spindle decreases as the score increases, and the mean spindle amplitude is generally slightly higher in the normal population than in the sleep-disordered population. The normal and sleep-disordered groups did not show obvious differences in the number of spindles between symmetric channels C3/C4 and F3/F4. The difference in the density and amplitude of the spindles proposed in this paper can be a reference characteristic for the diagnosis of sleep disorders and provide valuable objective evidence for clinical diagnosis. In summary, our proposed detection method can effectively improve the accuracy of sleep spindle wave detection with stable performance. Meanwhile, our study shows that the spindle density, frequency and amplitude are different between the sleep-disordered and normal populations.
- 30 March 2023
- Published by Frontiers Media SA in Frontiers in Neuroscience
- Vol. 17, 1110320
Open AccessA comparative analysis of sleep spindle characteristics of sleep-disordered patients and normal subjects
Spindles differ in density, amplitude, and frequency, and these variations reflect different physiological processes. Sleep disorders are characterized by difficulty in falling asleep and maintaining sleep. In this study, we proposed a new spindle wave detection algorithm, which was more effective compared with traditional detection algorithms such as wavelet algorithm. Besides, we recorded EEG data from 20 subjects with sleep disorders and 10 normal subjects, and then we compared the spindle characteristics of sleep-disordered subjects and normal subjects (those without any sleep disorder) to assess the spindle activity during human sleep. Specifically, we scored 30 subjects on the Pittsburgh Sleep Quality Index and then analyzed the association between their sleep quality scores and spindle characteristics, reflecting the effect of sleep disorders on spindle characteristics. We found a significant correlation between the sleep quality score and spindle density (p = 1.84 × 10−8, p-value p-value of 0.667, suggesting that the spindle frequency and sleep quality score were not significantly correlated. The p-value between the sleep quality score and spindle amplitude was 1.33 × 10−4, indicating that the mean amplitude of the spindle decreases as the score increases, and the mean spindle amplitude is generally slightly higher in the normal population than in the sleep-disordered population. The normal and sleep-disordered groups did not show obvious differences in the number of spindles between symmetric channels C3/C4 and F3/F4. The difference in the density and amplitude of the spindles proposed in this paper can be a reference characteristic for the diagnosis of sleep disorders and provide valuable objective evidence for clinical diagnosis. In summary, our proposed detection method can effectively improve the accuracy of sleep spindle wave detection with stable performance. Meanwhile, our study shows that the spindle density, frequency and amplitude are different between the sleep-disordered and normal populations.
- Open Access
- 9 March 2023
- Published by Frontiers Media SA in Frontiers in Neuroscience
- Vol. 17, 1105696
Hierarchical fusion detection algorithm for sleep spindle detection
Background: Sleep spindles are a vital sign implying that human beings have entered the second stage of sleep. In addition, they can effectively reflect a person’s learning and memory ability, and clinical research has shown that their quantity and density are crucial markers of brain function. The “gold standard” of spindle detection is based on expert experience; however, the detection cost is high, and the detection time is long. Additionally, the accuracy of detection is influenced by subjectivity.Methods: To improve detection accuracy and speed, reduce the cost, and improve efficiency, this paper proposes a layered spindle detection algorithm. The first layer used the Morlet wavelet and RMS method to detect spindles, and the second layer employed an improved k-means algorithm to improve spindle detection efficiency. The fusion algorithm was compared with other spindle detection algorithms to prove its effectiveness.Results: The hierarchical fusion spindle detection algorithm showed good performance stability, and the fluctuation range of detection accuracy was minimal. The average value of precision was 91.6%, at least five percentage points higher than other methods. The average value of recall could reach 89.1%, and the average value of specificity was close to 95%. The mean values of accuracy and F1-score in the subject sample data were 90.4 and 90.3%, respectively. Compared with other methods, the method proposed in this paper achieved significant improvement in terms of precision, recall, specificity, accuracy, and F1-score.Conclusion: A spindle detection method with high steady-state accuracy and fast detection speed is proposed, which combines the Morlet wavelet with window RMS and an improved k-means algorithm. This method provides a powerful tool for the automatic detection of spindles and improves the efficiency of spindle detection. Through simulation experiments, the sampled data were analyzed and verified to prove the feasibility and effectiveness of this method.
- 9 March 2023
- Published by Frontiers Media SA in Frontiers in Neuroscience
- Vol. 17, 1105696
Open AccessHierarchical fusion detection algorithm for sleep spindle detection
Background: Sleep spindles are a vital sign implying that human beings have entered the second stage of sleep. In addition, they can effectively reflect a person’s learning and memory ability, and clinical research has shown that their quantity and density are crucial markers of brain function. The “gold standard” of spindle detection is based on expert experience; however, the detection cost is high, and the detection time is long. Additionally, the accuracy of detection is influenced by subjectivity.Methods: To improve detection accuracy and speed, reduce the cost, and improve efficiency, this paper proposes a layered spindle detection algorithm. The first layer used the Morlet wavelet and RMS method to detect spindles, and the second layer employed an improved k-means algorithm to improve spindle detection efficiency. The fusion algorithm was compared with other spindle detection algorithms to prove its effectiveness.Results: The hierarchical fusion spindle detection algorithm showed good performance stability, and the fluctuation range of detection accuracy was minimal. The average value of precision was 91.6%, at least five percentage points higher than other methods. The average value of recall could reach 89.1%, and the average value of specificity was close to 95%. The mean values of accuracy and F1-score in the subject sample data were 90.4 and 90.3%, respectively. Compared with other methods, the method proposed in this paper achieved significant improvement in terms of precision, recall, specificity, accuracy, and F1-score.Conclusion: A spindle detection method with high steady-state accuracy and fast detection speed is proposed, which combines the Morlet wavelet with window RMS and an improved k-means algorithm. This method provides a powerful tool for the automatic detection of spindles and improves the efficiency of spindle detection. Through simulation experiments, the sampled data were analyzed and verified to prove the feasibility and effectiveness of this method.
- Open Access
- 30 November 2022
- Published by Elsevier in Brain Stimulation
- Vol. 15 (6) , 1530-1537
Efficacy and safety of simultaneous rTMS–tDCS over bilateral angular gyrus on neuropsychiatric symptoms in patients with moderate Alzheimer's disease: A prospective, randomized, sham-controlled pilot study
- 30 November 2022
- Published by Elsevier in Brain Stimulation
- Vol. 15 (6) , 1530-1537
Open AccessEfficacy and safety of simultaneous rTMS–tDCS over bilateral angular gyrus on neuropsychiatric symptoms in patients with moderate Alzheimer's disease: A prospective, randomized, sham-controlled pilot study
- Open Access
- 16 March 2022
- Published by Springer Nature in BMC Medical Genomics
- Vol. 15 (1) , 1-9
Novel compound heterozygous variants in the PCCB gene causing adult-onset propionic acidemia presenting with neuropsychiatric symptoms: a case report and literature review
Background: Propionic acidemia (PA) is a rare autosomal recessive disorder of metabolism caused by mutations in the PCCA or PCCB gene, leading to propionyl CoA carboxylase (PCC) enzyme deficiencies. Most PA patients present variable clinical phenotypes and severity in the neonatal or infant period, with only a few developing symptoms after infancy. This report describes a PA patient with an adult-onset phenotype and a novel compound heterozygous mutation in the PCCB gene. To further explore the genotype–phenotype correlations in late-onset PA, we performed a literature review focusing on and summarizing 11 patients with PCC gene mutations who had the first onset and/or the definite diagnosis after infancy. Case presentation: A 21-year-old PA patient presented with weakness of four limbs, gait abnormalities, two episodes of seizures, mental and behavior disorders after severe vomiting. Magnetic Resonance Imaging (MRI) demonstrated sustained bilateral caudate head and putamen symmetrical hyperintensity. Biochemical investigations revealed plasma amino and urine values correlating with a PA profile. Genetic analysis confirmed novel compound heterozygous variants in PCCB, with a newly-found pathogenic mutation (c.467T>C) and the c.1316A>G mutation associated with pathogenicity. Conclusion: We identified a novel compound heterozygous mutation in the PCCB gene causing late-onset PA. Patients carrying mutations in the PCCB gene tend to develop late-onset PA and present neuropsychiatric symptoms and/or signs. Further molecular biological research is needed to explore the genotype–phenotype correlations of PA.
- 16 March 2022
- Published by Springer Nature in BMC Medical Genomics
- Vol. 15 (1) , 1-9
Open AccessNovel compound heterozygous variants in the PCCB gene causing adult-onset propionic acidemia presenting with neuropsychiatric symptoms: a case report and literature review
Background: Propionic acidemia (PA) is a rare autosomal recessive disorder of metabolism caused by mutations in the PCCA or PCCB gene, leading to propionyl CoA carboxylase (PCC) enzyme deficiencies. Most PA patients present variable clinical phenotypes and severity in the neonatal or infant period, with only a few developing symptoms after infancy. This report describes a PA patient with an adult-onset phenotype and a novel compound heterozygous mutation in the PCCB gene. To further explore the genotype–phenotype correlations in late-onset PA, we performed a literature review focusing on and summarizing 11 patients with PCC gene mutations who had the first onset and/or the definite diagnosis after infancy. Case presentation: A 21-year-old PA patient presented with weakness of four limbs, gait abnormalities, two episodes of seizures, mental and behavior disorders after severe vomiting. Magnetic Resonance Imaging (MRI) demonstrated sustained bilateral caudate head and putamen symmetrical hyperintensity. Biochemical investigations revealed plasma amino and urine values correlating with a PA profile. Genetic analysis confirmed novel compound heterozygous variants in PCCB, with a newly-found pathogenic mutation (c.467T>C) and the c.1316A>G mutation associated with pathogenicity. Conclusion: We identified a novel compound heterozygous mutation in the PCCB gene causing late-onset PA. Patients carrying mutations in the PCCB gene tend to develop late-onset PA and present neuropsychiatric symptoms and/or signs. Further molecular biological research is needed to explore the genotype–phenotype correlations of PA.
- 1 August 2019
- Published by Elsevier in Epilepsy & Behavior
- Vol. 97, 291-295
The localizing and lateralizing value of palpitation aura in patients with focal epilepsy
- 1 August 2019
- Published by Elsevier in Epilepsy & Behavior
- Vol. 97, 291-295
The localizing and lateralizing value of palpitation aura in patients with focal epilepsy