Baocan Zhang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Baocan Zhang has a total of 19 co-authors. Here are the top co-authors by number of shared publications:
- Wei Zhao
Chengyi College, Jimei University, Xiamen 361021, China
5 shared publications - Xiaolu Jiang
Chengyi College, Jimei University, Xiamen 361021, China
4 shared publications - Shixiao Xiao
Chengyi College, Jimei University, Xiamen 361021, China
2 shared publications - Gaowei Xu
College of Electronic and Information Engineering, Tongji University, Shanghai, China
1 shared publication - Jiajia Chen
Zhongshan Hospital Xiamen University, Xiamen, Fujian 361004, China
1 shared publication - Liting Chen
Big Data Analytics Laboratory, Chengyi College, Jimei University, Xiamen, Fujian, China
1 shared publication
Latest Publications
- Open Access
- 25 September 2026
- Published by MDPI AG in Information
- Vol. 17 (10) , 951
A Novel VMamba-Based Deep Learning Model for Accurate Nuclei and Cytoplasm Segmentation in Cervical Cytology Images
Cervical cancer continues to be a major global health challenge, and robust screening capabilities remain the most effective strategy for its prevention. Within this context, image segmentation plays a critical role in screening processes, as it enables a more precise analysis of cellular morphological features, thereby improving diagnostic accuracy. This paper presents a method for three-class cell segmentation into nucleus, cytoplasm and background by combining the backbone network VMamba and a transformer-based segmentation head. The model is fine-tuned on the Herlev dataset and outperforms state-of-the-art methods in cytoplasm segmentation by a moderate margin and achieves competitive above-average performance in nucleus segmentation. Across the tasks of nucleus and cytoplasm segmentation, the model yields average scores of 0.9372 for Dice, 0.8819 for IoU, 0.9217 for Precision, and 0.9551 for Recall. These results demonstrate the suitability of the proposed segmentation model to help experts effectively assess cervical cell lesions.
- 25 September 2026
- Published by MDPI AG in Information
- Vol. 17 (10) , 951
Open AccessA Novel VMamba-Based Deep Learning Model for Accurate Nuclei and Cytoplasm Segmentation in Cervical Cytology Images
Cervical cancer continues to be a major global health challenge, and robust screening capabilities remain the most effective strategy for its prevention. Within this context, image segmentation plays a critical role in screening processes, as it enables a more precise analysis of cellular morphological features, thereby improving diagnostic accuracy. This paper presents a method for three-class cell segmentation into nucleus, cytoplasm and background by combining the backbone network VMamba and a transformer-based segmentation head. The model is fine-tuned on the Herlev dataset and outperforms state-of-the-art methods in cytoplasm segmentation by a moderate margin and achieves competitive above-average performance in nucleus segmentation. Across the tasks of nucleus and cytoplasm segmentation, the model yields average scores of 0.9372 for Dice, 0.8819 for IoU, 0.9217 for Precision, and 0.9551 for Recall. These results demonstrate the suitability of the proposed segmentation model to help experts effectively assess cervical cell lesions.
- Open Access
A Framework for Nuclei and Overlapping Cytoplasm Segmentation with MaskDino and Hausdorff Distance
Accurate segmentation of nuclei and cytoplasm in cervical cytology images plays a pivotal role in characterizing cellular morphology. The primary challenge is to precisely delineate boundaries within densely clustered cells, which is complicated by low-contrast edges and irregular morphologies. This paper introduces a novel framework combining MaskDino architecture with Hausdorff distance loss, enhanced by a two-phase training strategy. The method begins by employing MaskDino for precise nucleus segmentation. Building on this foundation, the framework then enhances cytoplasmic boundary detection in cellular clusters by incorporating a Hausdorff distance loss, with weight transfer initialization ensuring feature consistency across tasks.. The symmetry between the nucleus and cytoplasm servers as a key morphological indicator for cell assessment, and our method provides a reliable basis for such analysis. Extensive experiments demonstrate that our method achieves state-of-the-art cytoplasm segmentation results on the ISBI2014 dataset, with absolute improvements of 2.9% in DSC, 1.6% in TPRp and 2.0% in FNRo. The performance of nucleus segmentation is better than the average level. These results validate the proposed framework’s effectiveness for improving cervical cancer screening through robust cellular segmentation.
Open AccessA Framework for Nuclei and Overlapping Cytoplasm Segmentation with MaskDino and Hausdorff Distance
Accurate segmentation of nuclei and cytoplasm in cervical cytology images plays a pivotal role in characterizing cellular morphology. The primary challenge is to precisely delineate boundaries within densely clustered cells, which is complicated by low-contrast edges and irregular morphologies. This paper introduces a novel framework combining MaskDino architecture with Hausdorff distance loss, enhanced by a two-phase training strategy. The method begins by employing MaskDino for precise nucleus segmentation. Building on this foundation, the framework then enhances cytoplasmic boundary detection in cellular clusters by incorporating a Hausdorff distance loss, with weight transfer initialization ensuring feature consistency across tasks.. The symmetry between the nucleus and cytoplasm servers as a key morphological indicator for cell assessment, and our method provides a reliable basis for such analysis. Extensive experiments demonstrate that our method achieves state-of-the-art cytoplasm segmentation results on the ISBI2014 dataset, with absolute improvements of 2.9% in DSC, 1.6% in TPRp and 2.0% in FNRo. The performance of nucleus segmentation is better than the average level. These results validate the proposed framework’s effectiveness for improving cervical cancer screening through robust cellular segmentation.
- Open Access
- 24 October 2025
- Published by Association for Computing Machinery (ACM)
- p. 727-732
STRNet: A spatiotemporal residual network with a cascaded dilated convolutional pyramid for EEG-based motor imagery classification
- 24 October 2025
- Published by Association for Computing Machinery (ACM)
- p. 727-732
Open AccessSTRNet: A spatiotemporal residual network with a cascaded dilated convolutional pyramid for EEG-based motor imagery classification
- Open Access
- 9 October 2025
- Published by Springer Nature in Research Square
A novel Vmamba-based deep learning model for accurate nuclei and cytoplasm segmentation in cervical cytology images
Cervical cancer continues to be a major global health challenge, and robust screening capabilities remain the most effective strategy for its prevention. Within this context, image segmentation plays a critical role in screening processes,as it enables a more precise analysis of cellular morphological features, therebyimproving diagnostic accuracy. This paper presents a method for three-class cellsegmentation into nucleus, cytoplasm and background, by combining the backbonenetwork Vmamba and transformer-based segmentation head. The model is fine-tunedon the Herlev dataset and outperforms state-of-the-art methods by moderate margin.Across the tasks of nucleus and cytoplasm segmentation, the model yields average scores of 0.9372 for Dice, 0.8819 for IoU, 0.9217 for Precision, and 0.9551 for Recall.These results demonstrate the suitability of the proposed segmentation model to help experts effectively asset cervical cell lesions.
- 9 October 2025
- Published by Springer Nature in Research Square
Open AccessA novel Vmamba-based deep learning model for accurate nuclei and cytoplasm segmentation in cervical cytology images
Cervical cancer continues to be a major global health challenge, and robust screening capabilities remain the most effective strategy for its prevention. Within this context, image segmentation plays a critical role in screening processes,as it enables a more precise analysis of cellular morphological features, therebyimproving diagnostic accuracy. This paper presents a method for three-class cellsegmentation into nucleus, cytoplasm and background, by combining the backbonenetwork Vmamba and transformer-based segmentation head. The model is fine-tunedon the Herlev dataset and outperforms state-of-the-art methods by moderate margin.Across the tasks of nucleus and cytoplasm segmentation, the model yields average scores of 0.9372 for Dice, 0.8819 for IoU, 0.9217 for Precision, and 0.9551 for Recall.These results demonstrate the suitability of the proposed segmentation model to help experts effectively asset cervical cell lesions.
- 15 August 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 473-477
Cross-Graph Fusion with Multi-Relational Syntactic and Semantic Branches for Multi-Aspect Sentiment Analysis
Aspect-based sentiment analysis aims to determine the sentiment polarity of specific aspects within a sentence. Existing methods often focus on a single aspect or a single type of relational information, limiting their ability to model complex dependencies. We propose a multi-branch cross-graph fusion network that uses a BERT encoder, followed by parallel multirelational syntactic and semantic branches. Each branch output is fused with BERT representations via cross-attention, and the resulting features are concatenated. Special encoding and masking strategies enable our model to classify the sentiment of all aspects simultaneously. Experiments on benchmark datasets confirm both its strong performance and interpretability.
- 15 August 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 473-477
Cross-Graph Fusion with Multi-Relational Syntactic and Semantic Branches for Multi-Aspect Sentiment Analysis
Aspect-based sentiment analysis aims to determine the sentiment polarity of specific aspects within a sentence. Existing methods often focus on a single aspect or a single type of relational information, limiting their ability to model complex dependencies. We propose a multi-branch cross-graph fusion network that uses a BERT encoder, followed by parallel multirelational syntactic and semantic branches. Each branch output is fused with BERT representations via cross-attention, and the resulting features are concatenated. Special encoding and masking strategies enable our model to classify the sentiment of all aspects simultaneously. Experiments on benchmark datasets confirm both its strong performance and interpretability.
- Open Access
- 25 November 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 12, 176586-176597
An Enhanced Mask Transformer for Overlapping Cervical Cell Segmentation Based on DETR
Automated cell segmentation in cervical cytology images is an essential task because it can present a deep understanding of the characteristics of cervical cells. The main challenge is that cells overlap at a high rate, making the cell boundaries extremely blurred. While the transformer-based models have been proven to be effective in vision tasks, this paper proposes an enhanced mask transformer based on DETR to segment overlapping cervical cells. Dynamic anchor box initialization and noised ground truth box embedding are introduced, to improve segmentation performance and accelerate model convergence. The proposed model achieves a 0.974 DSC, 0.971 TPRp, 0.0005 FPRp and 0.0012 FNRo on the ISBI2014 dataset. Specially, the proposed method outperforms state-of-the-art result by about 3.4% on DSC, 2% on TPRp and 1.88% on FNRo, respectively. The metrics of our model on the ISBI2015 dataset are a little better than the averaged metrics of other impressive methods. These findings present strong support for the transformer-based neural networks for effective segmentation of cells in cervical cytology images.
- 25 November 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 12, 176586-176597
Open AccessAn Enhanced Mask Transformer for Overlapping Cervical Cell Segmentation Based on DETR
Automated cell segmentation in cervical cytology images is an essential task because it can present a deep understanding of the characteristics of cervical cells. The main challenge is that cells overlap at a high rate, making the cell boundaries extremely blurred. While the transformer-based models have been proven to be effective in vision tasks, this paper proposes an enhanced mask transformer based on DETR to segment overlapping cervical cells. Dynamic anchor box initialization and noised ground truth box embedding are introduced, to improve segmentation performance and accelerate model convergence. The proposed model achieves a 0.974 DSC, 0.971 TPRp, 0.0005 FPRp and 0.0012 FNRo on the ISBI2014 dataset. Specially, the proposed method outperforms state-of-the-art result by about 3.4% on DSC, 2% on TPRp and 1.88% on FNRo, respectively. The metrics of our model on the ISBI2015 dataset are a little better than the averaged metrics of other impressive methods. These findings present strong support for the transformer-based neural networks for effective segmentation of cells in cervical cytology images.
- Open Access
- 4 October 2021
- Published by Wiley in Computational and Mathematical Methods in Medicine
- Vol. 2021, 1-10
Segmentation of Overlapping Cervical Cells with Mask Region Convolutional Neural Network
The task of segmenting cytoplasm in cytology images is one of the most challenging tasks in cervix cytological analysis due to the presence of fuzzy and highly overlapping cells. Deep learning-based diagnostic technology has proven to be effective in segmenting complex medical images. We present a two-stage framework based on Mask RCNN to automatically segment overlapping cells. In stage one, candidate cytoplasm bounding boxes are proposed. In stage two, pixel-to-pixel alignment is used to refine the boundary and category classification is also presented. The performance of the proposed method is evaluated on publicly available datasets from ISBI 2014 and 2015. The experimental results demonstrate that our method outperforms other state-of-the-art approaches with DSC 0.92 and FPRp 0.0008 at the DSC threshold of 0.8. Those results indicate that our Mask RCNN-based segmentation method could be effective in cytological analysis.
- 4 October 2021
- Published by Wiley in Computational and Mathematical Methods in Medicine
- Vol. 2021, 1-10
Open AccessSegmentation of Overlapping Cervical Cells with Mask Region Convolutional Neural Network
The task of segmenting cytoplasm in cytology images is one of the most challenging tasks in cervix cytological analysis due to the presence of fuzzy and highly overlapping cells. Deep learning-based diagnostic technology has proven to be effective in segmenting complex medical images. We present a two-stage framework based on Mask RCNN to automatically segment overlapping cells. In stage one, candidate cytoplasm bounding boxes are proposed. In stage two, pixel-to-pixel alignment is used to refine the boundary and category classification is also presented. The performance of the proposed method is evaluated on publicly available datasets from ISBI 2014 and 2015. The experimental results demonstrate that our method outperforms other state-of-the-art approaches with DSC 0.92 and FPRp 0.0008 at the DSC threshold of 0.8. Those results indicate that our Mask RCNN-based segmentation method could be effective in cytological analysis.
- Open Access
- 8 May 2020
- Published by Wiley in Computational and Mathematical Methods in Medicine
- Vol. 2020, 1-8
Cross-Subject Seizure Detection in EEGs Using Deep Transfer Learning
Electroencephalography (EEG) plays an import role in monitoring the brain activities of patients with epilepsy and has been extensively used to diagnose epilepsy. Clinically reading tens or even hundreds of hours of EEG recordings is very time consuming. Therefore, automatic detection of seizure is of great importance. But the huge diversity of EEG signals belonging to different patients makes the task of seizure detection much challenging, for both human experts and automation methods. We propose three deep transfer convolutional neural networks (CNN) for automatic cross-subject seizure detection, based on VGG16, VGG19, and ResNet50, respectively. The original dataset is the CHB-MIT scalp EEG dataset. We use short time Fourier transform to generate time-frequency spectrum images as the input dataset, while positive samples are augmented due to the infrequent nature of seizure. The model parameters pretrained on ImageNet are transferred to our models. And the fine-tuned top layers, with an output layer of two neurons for binary classification (seizure or nonseizure), are trained from scratch. Then, the input dataset are randomly shuffled and divided into three partitions for training, validating, and testing the deep transfer CNNs, respectively. The average accuracies achieved by the deep transfer CNNs based on VGG16, VGG19, and ResNet50 are 97.75%, 98.26%, and 96.17% correspondingly. On those results of experiments, our method could prove to be an effective method for cross-subject seizure detection.
- 8 May 2020
- Published by Wiley in Computational and Mathematical Methods in Medicine
- Vol. 2020, 1-8
Open AccessCross-Subject Seizure Detection in EEGs Using Deep Transfer Learning
Electroencephalography (EEG) plays an import role in monitoring the brain activities of patients with epilepsy and has been extensively used to diagnose epilepsy. Clinically reading tens or even hundreds of hours of EEG recordings is very time consuming. Therefore, automatic detection of seizure is of great importance. But the huge diversity of EEG signals belonging to different patients makes the task of seizure detection much challenging, for both human experts and automation methods. We propose three deep transfer convolutional neural networks (CNN) for automatic cross-subject seizure detection, based on VGG16, VGG19, and ResNet50, respectively. The original dataset is the CHB-MIT scalp EEG dataset. We use short time Fourier transform to generate time-frequency spectrum images as the input dataset, while positive samples are augmented due to the infrequent nature of seizure. The model parameters pretrained on ImageNet are transferred to our models. And the fine-tuned top layers, with an output layer of two neurons for binary classification (seizure or nonseizure), are trained from scratch. Then, the input dataset are randomly shuffled and divided into three partitions for training, validating, and testing the deep transfer CNNs, respectively. The average accuracies achieved by the deep transfer CNNs based on VGG16, VGG19, and ResNet50 are 97.75%, 98.26%, and 96.17% correspondingly. On those results of experiments, our method could prove to be an effective method for cross-subject seizure detection.