Jiacheng Pan

Jiacheng Pan

Auto-generated profile, based on publicly available information

Jiacheng Pan

The Information and Intelligent Engineering Department, Ningbo City College of Vocational Technology, Ningbo, China

Research Focus & Metrics

Academic Career Length

5Years (2022-Present)

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

Representative Publications

View All Publications
6 July 2022 in Computers in Biology and Medicine
Verified
Journal Article
Closed access
JPJiacheng Pan
Jiacheng Pan
HLHaocai Lin
Haocai Lin
YDYihong Dong
Yihong Dong
YWYu Wang
Yu Wang
YJYunxin Ji
Yunxin Ji

13 August 2025 in IEEE Access
Verified
Journal Article
Open access
DJDaogen Jiang
Daogen Jiang
LWLongyang Wang
Longyang Wang
JLJiahao Li
Jiahao Li
JPJiacheng Pan
Jiacheng Pan
This paper proposes a novel event-triggered adaptive neural control scheme for pure-feedback stochastic nonlinear systems subject to input saturation and partial state constraints. By employing the implicit function theorem and mean value theorem, the pure-feedback system is systematically transformed into a strict-feedback form with bounded approximation errors. A key innovation is the introduction of a continuously differentiable asymmetric saturation model based on hyperbolic tangent functions, which accurately characterizes input constraints. The control design integrates backstepping with barrier Lyapunov functions (BLFs) and adaptive neural network, ensuring robust handling of state constraints and stochastic uncertainties at each design step. Theoretical analysis proves that all closed-loop signals are semi-globally uniformly ultimately bounded (SGUUB) in probability (fourth-moment sense), while the tracking error converges asymptotically to zero. An event-triggered mechanism further improves computational efficiency. Comparative simulations validate the superiority of the proposed method over conventional approaches.

1 January 2022 in SSRN Electronic Journal
Verified
Preprint
Open access
PDF file
JPJiacheng Pan
Jiacheng Pan
MJMin Jiang
Min Jiang
HLHaocai Lin
Haocai Lin
YDYihong Dong
Yihong Dong
Background and Objective: Benefiting from powerful graph representation capabilities, graph neural networks have shown impressive performance in various biomedical application tasks. Although GNN has achieved great success, the data noise and data scarcity problems commonly faced in real psychiatric disease prediction scenarios may affect the training and prediction of graph learning models. At present, there is no relevant work to obtain a reasonable solution. Data augmentation, which allows limited data to produce value equivalent to more data without substantially increasing the data, is considered a practical approach to addressing the problem of noisy data and data scarcity. Methods: In this work, we propose a method based on graph data augmentation for solving the problem of noisy data and data scarcity in mental illness prediction. To mitigate the negative effects of label noise, we use edge predictors to optimize the graph topology, enhance links to nodes with high similarity, remove erroneous noisy edges, and enhance the model robustness by adding adversarial perturbations in the feature space. In addition, a confident self-checking mechanism allows accurate pseudolabeling to be obtained, providing more supervision for the model training phase and further reducing the effect of label noise. Results: To demonstrate the effectiveness of our approach, we evaluate the performance of the proposed method on the Autism Brain Imaging Data Exchange (ABIDE) dataset and the Major Depressive Disorder (MDD) dataset. Sufficient ablation experimental studies were conducted to assess the effectiveness of each component. Our proposed method achieves an average accuracy of 70.86%/88.69% and an AUC of 74.28%/95.79 on ABIDE/MDD. Compared with the model of previous works, our approach improves approximately 8.75% 18.98% on the ABIDE dataset and approximately 7.38%23.46% on the MDD dataset. Conclusion: The comprehensive experiments on two multimodal real-life biomedical datasets demonstrate the effectiveness and generality of the proposed framework, which can use to address the data noise and data scarcity problems. Moreover, we find that our method can helps to build a more secure and reliable deep learning-assisted diagnosis system. More importantly, the promising results suggest that applying graph data augmentation in brain networks to improve the classification performance is worth further investigation.

1 December 2025 in Journal of Computational Biology
Verified
Journal Article
Closed access
JPJiacheng Pan
Jiacheng Pan
YDYihong Dong
Yihong Dong
DJDaogen Jiang
Daogen Jiang
LWLongyang Wang
Longyang Wang
Graph neural networks have shown impressive performance in a variety of biomedical application tasks due to their powerful graph representation capabilities. Although GNN has achieved great success, the data noise and data scarcity problems commonly faced in real psychiatric disease prediction scenarios may affect the training and prediction of graph learning models. At present, there is no relevant work to obtain a reasonable solution. Data augmentation, which allows limited data to produce value equivalent to more data without substantially increasing the data, is considered a practical approach to addressing the problem of noisy data and data scarcity. In this work, we propose a method based on graph data augmentation for solving the problem of noisy data and data scarcity in mental illness prediction. To mitigate the negative effects of label noise, we use edge predictors to optimize the graph topology, enhance links to nodes with high similarity, remove erroneous noisy edges, and enhance the model robustness by adding adversarial perturbations in the feature space. In addition, a confident self-checking mechanism allows accurate pseudolabeling to be obtained, providing more supervision for the model training phase and further reducing the effect of label noise. Extensive experiments on two multimodal real mental illness datasets show that the proposed approach has better performance. Sufficient ablation experimental studies were conducted to assess the effectiveness of each component. The experimental results validate the effectiveness and scalability of our framework for population-based disease prediction, even under challenging conditions of data noise and sparsity. The implementation code is publicly available at: https://github.com/jiachengpan98/GDA-GCN .