YZYuchang Zhao

Yuchang Zhao

*This is an algorithmically generated author record based on publicly available information.

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


Yuchang Zhao has a total of 15 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Dataset distillation for efficient seizure prediction: tiny-scale data, comparable performance

    by Yuchang Zhao, Xun Chen, Chang Li, Yikai Gao, Ruobing Qian, Aiping Liu

    Objective. Deep learning has shown significant potential in electroencephalogram (EEG)-based seizure prediction. However, translating these advances into practical applications faces critical challenges: the prohibitive time costs of reviewing accumulated historical data during frequent model updates, the need for rapid adaptation to diverse model architectures across hardware platforms, and the substantial burden of massive data storage. To this end, we introduce the first dataset distillation study tailored for seizure prediction, which condenses the original dataset into a compact, information-dense synthetic dataset. This tiny-scale proxy allows models to match the performance of those trained on the full-scale dataset. Approach. Methodologically, we formulate the distillation as a min–max optimization task, introducing an EEG-aware alignment objective that jointly preserves latent feature distributions, temporal dynamics in the time-frequency domain, and inter-channel correlation structures. Through signal-compatible amplitude and phase decomposition, we optimize the frequency parameter to maximize the CF discrepancy via a sampling network, while simultaneously tuning the synthetic data to minimize this gap. Additionally, we integrate hard labels to refine soft labels to enhance the generalization of synthetic data. Main results. Extensive validation across 37 public and 35 clinical subjects demonstrate the method’s robustness across diverse scenarios and model architectures. With merely 1% of the original data volume, our approach retains roughly 95% of the original performance while reducing training duration by approximately 98%. Significance. This method addresses clinical translation bottlenecks by distilling massive EEG data into compact synthetic sets, offering a promising solution for efficient model maintenance, flexible model switching, and optimized storage.

  • Generalizable Seizure Prediction With LLMs: Converting EEG to Textual Representations

    by Yuchang Zhao, Aiping Liu, Chang Li, Lanlan Wang, Ruobing Qian, Xun Chen

    Seizure prediction through scalp electroencephalogram (EEG) holds considerable practical potential. The primary challenge faced by existing algorithms lies in the individual heterogeneity, which hinders the generalizability of models to new patients. Additionally, inconsistencies in channel settings across various epilepsy centers further limit the applicability of models to diverse datasets. To address these challenges, we incorporate large language models (LLMs) into EEG analysis and propose a novel seizure prediction method based on LLMs (SPLLM), significantly enhancing both model generalizability and applicability. Specifically, this approach reprograms LLMs by transforming EEG signals into textual representations compatible with LLMs via a single-channel pre-training strategy. The method integrates cross-domain knowledge from both text and EEG data through a cross-attention mechanism, utilizing autoregressive pretrained LLMs to capture the temporal dependencies inherent in EEG signals. Moreover, the cross-domain generalization ability of LLMs alleviates patient heterogeneity, while the single-channel pre-training strategy enables the model to adapt to diverse channel settings. On two public datasets and one private dataset, SPLLM increases the average AUC by 8.2%, and the average balanced accuracy by 8.4% compared to existing methods. Experimental results demonstrate that the proposed method not only enhances cross-patient prediction accuracy but also adapts to data from different datasets, offering a scalable solution for the clinical application of seizure prediction.


  • Data Augmentation for Seizure Prediction With Generative Diffusion Model

    by Kai Shu, Le Wu, Yuchang Zhao, Aiping Liu, Ruobing Qian, Xun Chen

    Data augmentation (DA) can significantly strengthen the electroencephalogram (EEG)-based seizure prediction methods. However, existing DA approaches are just the linear transformations of original data and cannot explore the feature space to increase diversity effectively. Therefore, we propose a novel diffusion-based DA method called DiffEEG. DiffEEG can fully explore data distribution and generate samples with high diversity, offering extra information to classifiers. It involves two processes: the diffusion process and the denoised process. In the diffusion process, the model incrementally adds noise with different scales to EEG input and converts it into random noise. In this way, the representation of data can be learned. In the denoised process, the model utilizes learned knowledge to sample synthetic data from random noise input by gradually removing noise. The randomness of input noise and the precise representation enable the synthetic samples to possess diversity while ensuring the consistency of feature space. We compared DiffEEG with original, down-sampling, sliding windows and recombination methods, and integrated them into five representative classifiers. The experiments demonstrate the effectiveness and generality of our method. With the contribution of DiffEEG, the Multi-scale CNN achieves state-of-the-art performance, with an average sensitivity, FPR, AUC of 95.4%, 0.051/h, 0.932 on the CHB-MIT database and 93.6%, 0.121/h, 0.822 on the Kaggle database.


  • EEG-Based Seizure Prediction Via GhostNet and Imbalanced Learning

    by Tingting Mao, Chang Li, Yuchang Zhao, Rencheng Song, Xun Chen

    Seizure prediction task based on electroencephalography (EEG) is a typical imbalanced classification problem since the preictal samples are much less than the interictal samples. Re-sampling is a common method to solve class-imbalance problems in the area of seizure prediction. However, up-sampling has the deficiency of overfitting minority class, and down-sampling tends to discard important information. In addition, many seizure prediction algorithms depend on artificially designed features, which is a time-consuming and laborious process. To this end, we propose an end-to-end ghost network and class re-balanced loss function (GhostNet-CRBLF). First, the ghostnet with one-dimensional convolution mainly uses linear operations instead of traditional convolution to generate feature maps, which improves model performance and reduces the amount of parameters. In addition, one-dimensional convolution is adopted to better reduce the data dimension. Then, cost-sensitive learning is utilized to solve the problem of data imbalance. Specifically, a simple but effective class re-balanced loss (CRB-loss) function is proposed to increase the misclassification cost of preictal data, which makes classifier to focus more on minority samples and classifies them correctly. Experimental performance indicates the proposed approach can achieve remarkable performance gains in seizure prediction task.


  • Online Test-Time Adaptation for Patient-Independent Seizure Prediction

    by Tingting Mao, Chang Li, Yuchang Zhao, Rencheng Song, Xun Chen

    Existing domain adaptation (DA) methods typically require access to source domain data, which raises privacy concerns due to the sensitive information contained in electroencephalogram (EEG) data. Moreover, test data generally arrive in a sequential or batch manner in real-world scenarios. To this end, we adopt online test-time adaptation (OTTA) paradigm, which adapts a pre-trained source model, instead of the source data, to the target data stream at test time. OTTA is very useful in the epilepsy prediction task, which can protect patient privacy and achieve online prediction for seizures. Specifically, we introduce a novel teacher-student online test-time adaptation (TSOTTA) approach based on knowledge distillation and exponential moving average (EMA) strategy. Experiments on two public seizure datasets demonstrated that TSOTTA can enhance the generalization ability of the pre-trained source model in cross-patient online seizure prediction.


  • Source-Free Domain Adaptation for Privacy-Preserving Seizure Prediction

    by Yuchang Zhao, Shuai Feng, Chang Li, Rencheng Song, Deng Liang, Xun Chen

    Domain adaptation (DA) techniques are frequently utilized to enhance seizure prediction accuracy by leveraging the labeled electroencephalogram data of existing patients on new patients. Traditional DA methods, however, require access to the source domain while training the adaptation model, which poses a threat to sensitive patient information and privacy. To address this issue, in this article, we propose a novel Gaussian mixture modeling (GMM)-based source-free domain adaptation (GSFDA). Our method leverages the GMM joint source model and target data structure for clustering, employs uncertainty learning to minimize DA uncertainty, and uses the mixup technique to increase model robustness while reducing the impact of noisy pseudolabels. Notably, GSFDA only requires access to the source model parameters, and not the source domain, effectively safeguarding the privacy of patient information. This has substantial clinical implications for seizure prediction.


  • Patient-Specific Seizure Prediction From Electroencephalogram Signal via Multichannel Feedback Capsule Network

    by Chang Li, Yuchang Zhao, Rencheng Song, Xiang Liu, Ruobing Qian, Xun Chen

    In recent years, the use of convolutional neural networks (CNNs) have been common in Electroencephalogram (EEG) based seizure prediction. However, CNNs lose local and global connections and spatial information due to local connections and pooling operations. In addition, existing seizure prediction methods often require the design of special feature pre-extraction steps. Therefore, an end-to-end patient-specific seizure predictor based on feedback capsule network (FB-CapsNet) is proposed in this study. It can characterize complex temporal information and precise spatial relationships, capture and integrate spatiotemporal properties directly from the raw EEG signal, and distinguish seizure states. The proposed FB-CapsNet first uses a feedback network to extract temporal information and one-dimensional convolution to reduce the data dimensionality. Then, it uses capsule network to capture spatial information and other instantiation properties and store them in capsule vectors. Finally, it performs information flow between low-level and high-level capsules by a dynamic routing mechanism to obtain superior classification performance. The proposed FB-CapsNet achieves 95.7% sensitivity, 0.087/h false prediction rate (FPR), and 0.948 area under curve (AUC) on the CHB-MIT scalp dataset and 88.6% sensitivity, 0.127/h FPR, and 0.837 AUC on Kaggle dataset.


  • Top Cited Papers
    Open Access

    Patient-Specific Seizure Prediction via Adder Network and Supervised Contrastive Learning

    by Yuchang Zhao, Chang Li, Xiang Liu, Ruobing Qian, Rencheng Song, Xun Chen

    Deep learning (DL) methods have been widely used in the field of seizure prediction from electroencephalogram (EEG) in recent years. However, DL methods usually have numerous multiplication operations resulting in high computational complexity. In addtion, most of the current approaches in this field focus on designing models with special architectures to learn representations, ignoring the use of intrinsic patterns in the data. In this study, we propose a simple and effective end-to-end adder network and supervised contrastive learning (AddNet-SCL). The method uses addition instead of the massive multiplication in the convolution process to reduce the computational cost. Besides, contrastive learning is employed to effectively use label information, points of the same class are clustered together in the projection space, and points of different class are pushed apart at the same time. Moreover, the proposed model is trained by combining the supervised contrastive loss from the projection layer and the cross-entropy loss from the classification layer. Since the adder networks uses the $\ell _{{1}}$ -norm distance as the similarity measure between the input feature and the filters, the gradient function of the network changes, an adaptive learning rate strategy is employed to ensure the convergence of AddNet-SCL. Experimental results show that the proposed method achieves 94.9% sensitivity, an area under curve (AUC) of 94.2%, and a false positive rate of (FPR) 0.077/h on 19 patients in the CHB-MIT database and 89.1% sensitivity, an AUC of 83.1%, and an FPR of 0.120/h in the Kaggle database. Competitive results show that this method has broad prospects in clinical practice.