Guodong Zhang
Research Subjects
Guodong Zhang has a total of 20 co-authors. Here are the top co-authors by number of shared publications:
- Guijin Wang
Department of Electronic Engineering, Tsinghua University, Beijing, China
2 shared publications - Qing Gao
School of Pharmacy, Guangdong Pharmaceutical University, Guangzhou, People’s Republic of China
2 shared publications - Shenglong Chen
College of Mathematics and System Science, Xinjiang University, Urumqi 830017, China
2 shared publications - Yu Chen
Cardiac Surgery Department, Peking University People’s Hospital, Beijing, China
2 shared publications - Feng Zheng
Department of Electrical Engineering, School of Electrical and Mechanical Engineering, Xidian University, Xian, China
1 shared publication Guoqiang Ren
Shandong University of Science and Technology
1 shared publication
Latest Publications
- 22 December 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-7
Efficient Scene Text Segmentation via Gate Mechanism and Multi-Scale Feature Fusion
In addressing the challenges posed by current text segmentation methods when dealing with multi-scale and irregular scene text, this paper presents an innovative approach to scene text segmentation. The proposed method leverages multiple Gated Fully Fusion Modules, facilitating adaptive weight adjustments for features at various levels and selective information aggregation. This ensures enhanced capture of multiscale text information. Additionally, a Multi-Scale Feature Fusion Module further improves information capture across different scales. The incorporation of a Text Prior Module and an Enhancement Module focuses on learning text prior information and edge features, respectively, refining the precise recognition and segmentation capabilities of multi-scale text targets. Experimental evaluations on the TextSegt dataset and the Bilingual Text Segmentation dataset validate the effectiveness of the proposed algorithm. Results demonstrate outstanding performance in handling text of various languages and sizes, highlighting robust adaptability and generalization capabilities. This work presents a compelling solution for scene text segmentation tasks.
- 22 December 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-7
Efficient Scene Text Segmentation via Gate Mechanism and Multi-Scale Feature Fusion
In addressing the challenges posed by current text segmentation methods when dealing with multi-scale and irregular scene text, this paper presents an innovative approach to scene text segmentation. The proposed method leverages multiple Gated Fully Fusion Modules, facilitating adaptive weight adjustments for features at various levels and selective information aggregation. This ensures enhanced capture of multiscale text information. Additionally, a Multi-Scale Feature Fusion Module further improves information capture across different scales. The incorporation of a Text Prior Module and an Enhancement Module focuses on learning text prior information and edge features, respectively, refining the precise recognition and segmentation capabilities of multi-scale text targets. Experimental evaluations on the TextSegt dataset and the Bilingual Text Segmentation dataset validate the effectiveness of the proposed algorithm. Results demonstrate outstanding performance in handling text of various languages and sizes, highlighting robust adaptability and generalization capabilities. This work presents a compelling solution for scene text segmentation tasks.
- 21 September 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Signal Processing Letters
- Vol. 28 (10709908) , 1993-1997
CLECG: A Novel Contrastive Learning Framework for Electrocardiogram Arrhythmia Classification
Deep learning-based intelligent electrocardiogram (ECG) diagnosis algorithms heavily rely on large annotated datasets. Unfortunately, in the context of ECG diagnosis, privacy issues and the high cost of data annotations lead to a shortage of ECG datasets which severely limits the performance of the state-of-the-art ECG diagnosis algorithms. In this paper, we propose a novel instance-level contrastive learning scheme for ECG signals, namely CLECG, to mine effective information from unlabeled data. During the pre-training, CLECG encourages the representations of different augmented views of the same signal (positive samples) to be similar and increases the distance between representations of augmented views from the different signals (negative samples). The whole pre-training process does not require any form of labeling. Experimental results show that the proposed CLECG strategy outperforms other self-supervised methods and supervised transfer learning strategies.
- 21 September 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Signal Processing Letters
- Vol. 28 (10709908) , 1993-1997
CLECG: A Novel Contrastive Learning Framework for Electrocardiogram Arrhythmia Classification
Deep learning-based intelligent electrocardiogram (ECG) diagnosis algorithms heavily rely on large annotated datasets. Unfortunately, in the context of ECG diagnosis, privacy issues and the high cost of data annotations lead to a shortage of ECG datasets which severely limits the performance of the state-of-the-art ECG diagnosis algorithms. In this paper, we propose a novel instance-level contrastive learning scheme for ECG signals, namely CLECG, to mine effective information from unlabeled data. During the pre-training, CLECG encourages the representations of different augmented views of the same signal (positive samples) to be similar and increases the distance between representations of augmented views from the different signals (negative samples). The whole pre-training process does not require any form of labeling. Experimental results show that the proposed CLECG strategy outperforms other self-supervised methods and supervised transfer learning strategies.
- 17 March 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Signal Processing Letters
- Vol. 28 (10709908) , 658-662
Non-Local Aggregation for RGB-D Semantic Segmentation
Exploiting both RGB (2D appearance) and Depth (3D geometry) information can improve the performance of semantic segmentation. However, due to the inherent difference between the RGB and Depth information, it remains a challenging problem in how to integrate RGB-D features effectively. In this letter, to address this issue, we propose a Non-local Aggregation Network (NANet), with a well-designed Multi-modality Non-local Aggregation Module (MNAM), to better exploit the non-local context of RGB-D features at multi-stage. Compared with most existing RGB-D semantic segmentation schemes, which only exploit local RGB-D features, the MNAM enables the aggregation of non-local RGB-D information along both spatial and channel dimensions. The proposed NANet achieves comparable performances with state-of-the-art methods on popular RGB-D benchmarks, NYUDv2 and SUN-RGBD.
- 17 March 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Signal Processing Letters
- Vol. 28 (10709908) , 658-662
Non-Local Aggregation for RGB-D Semantic Segmentation
Exploiting both RGB (2D appearance) and Depth (3D geometry) information can improve the performance of semantic segmentation. However, due to the inherent difference between the RGB and Depth information, it remains a challenging problem in how to integrate RGB-D features effectively. In this letter, to address this issue, we propose a Non-local Aggregation Network (NANet), with a well-designed Multi-modality Non-local Aggregation Module (MNAM), to better exploit the non-local context of RGB-D features at multi-stage. Compared with most existing RGB-D semantic segmentation schemes, which only exploit local RGB-D features, the MNAM enables the aggregation of non-local RGB-D information along both spatial and channel dimensions. The proposed NANet achieves comparable performances with state-of-the-art methods on popular RGB-D benchmarks, NYUDv2 and SUN-RGBD.
- Open Access
- 1 August 2020
- Published by AME Publishing Company in Annals of Translational Medicine
- Vol. 8 (16) , 1012
Design, application, and evaluation of a novel method for determining optimal trajectory of thoracic pedicle screws
Background: This is an experimental study performed on 15 adult cadavers. In this cadaveric study, we designed and evaluated a novel methodology for determining the optimal trajectory for the placement of thoracic pedicle screws. The accuracy of thoracic pedicle screw placement is critical to the spinal surgery. The concept, implement method, and significance of the optimal thoracic pedicle trajectory have not been reported. Methods: The experimental study was performed on 15 adult cadavers. The Mimics software was used to design optimal trajectory through the pedicle central axis. Using three-dimensional (3D) printing, a navigation module with a locating facet and a stabilizing facet was developed. The thoracic pedicle screws were inserted with the help of the navigation module. The three-dimensional coordinates for the entry and the exit points of the screws were compared between the planned trajectories and the postoperative trajectories. The differences in coordinates were analyzed to evaluate the precision of the screw placement. Results: The trajectories through the pedicle central axis showed an excellent symmetry between the single segments and for all thoracic vertebrae. Out of a total of 358 screws that were inserted, 15 (4.2%) screws breached the pedicle cortex with a breach distance of 0.05). Conclusions: The optimal trajectory was obtained through the pedicle central axis, which significantly reduced the risk of cortex breach. A high degree of precision was obtained for the entry and the exit points of the screws when the postoperative trajectory was compared with the designed trajectory.
- 1 August 2020
- Published by AME Publishing Company in Annals of Translational Medicine
- Vol. 8 (16) , 1012
Open AccessDesign, application, and evaluation of a novel method for determining optimal trajectory of thoracic pedicle screws
Background: This is an experimental study performed on 15 adult cadavers. In this cadaveric study, we designed and evaluated a novel methodology for determining the optimal trajectory for the placement of thoracic pedicle screws. The accuracy of thoracic pedicle screw placement is critical to the spinal surgery. The concept, implement method, and significance of the optimal thoracic pedicle trajectory have not been reported. Methods: The experimental study was performed on 15 adult cadavers. The Mimics software was used to design optimal trajectory through the pedicle central axis. Using three-dimensional (3D) printing, a navigation module with a locating facet and a stabilizing facet was developed. The thoracic pedicle screws were inserted with the help of the navigation module. The three-dimensional coordinates for the entry and the exit points of the screws were compared between the planned trajectories and the postoperative trajectories. The differences in coordinates were analyzed to evaluate the precision of the screw placement. Results: The trajectories through the pedicle central axis showed an excellent symmetry between the single segments and for all thoracic vertebrae. Out of a total of 358 screws that were inserted, 15 (4.2%) screws breached the pedicle cortex with a breach distance of 0.05). Conclusions: The optimal trajectory was obtained through the pedicle central axis, which significantly reduced the risk of cortex breach. A high degree of precision was obtained for the entry and the exit points of the screws when the postoperative trajectory was compared with the designed trajectory.
- 1 August 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 331-335
Course Knowledge Graph Review
In recent years, researches based on knowledge graph have attracted widespread attention from various fields. Taking the application of knowledge graph in pedagogy as an example, this study firstly introduces the significance, value and status quo of the course knowledge graph research, and then expounds and analyzes the application of knowledge graph in different courses in China.
- 1 August 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 331-335
Course Knowledge Graph Review
In recent years, researches based on knowledge graph have attracted widespread attention from various fields. Taking the application of knowledge graph in pedagogy as an example, this study firstly introduces the significance, value and status quo of the course knowledge graph research, and then expounds and analyzes the application of knowledge graph in different courses in China.
- Open Access
- 13 April 2019
- Published by Wiley in Journal of Cardiac Surgery
- Vol. 34 (6) , 506-510
Intimal sarcoma of the inferior vena cava with extension to the right atrium, bilateral renal vasculature, and bilateral venae iliaca communis, accompanied by severe thrombocytopenia
A 66‐year‐old Asian female diagnosed with an occupied mass of the inferior vena cava (IVC) case with extension to the right atrium, bilateral renal vasculature, and venae iliaca communis, which caused severe thrombocytopenia and severe liver and kidney dysfunction. The postoperative pathological diagnosis was high‐grade undifferentiated intimal sarcoma. There have been less than 10 cases of venous undifferentiated intimal sarcomas reported in the literature, three of which occurred in the IVC. Unfortunately, the patient died of multiple organ failure in the intensive care unit postoperatively.
- 13 April 2019
- Published by Wiley in Journal of Cardiac Surgery
- Vol. 34 (6) , 506-510
Open AccessIntimal sarcoma of the inferior vena cava with extension to the right atrium, bilateral renal vasculature, and bilateral venae iliaca communis, accompanied by severe thrombocytopenia
A 66‐year‐old Asian female diagnosed with an occupied mass of the inferior vena cava (IVC) case with extension to the right atrium, bilateral renal vasculature, and venae iliaca communis, which caused severe thrombocytopenia and severe liver and kidney dysfunction. The postoperative pathological diagnosis was high‐grade undifferentiated intimal sarcoma. There have been less than 10 cases of venous undifferentiated intimal sarcomas reported in the literature, three of which occurred in the IVC. Unfortunately, the patient died of multiple organ failure in the intensive care unit postoperatively.
- Open Access
- 4 February 2019
- Published by Springer Nature in Journal of Cardiothoracic Surgery
- Vol. 14 (1) , 1-6
Primary cardiac synovial sarcoma that was continuous with the mitral valve caused severe thrombocytopenia: a case report
Primary cardiac sarcomas are exceedingly rare, and they commonly result in nonspecific constitutional symptoms such as shortness of breath, weight loss, and anaemia-related fatigue and malaise. However, thrombocytopenia has very rarely been reported in association with cardiac tumours, either benign or malignant. We report one case of primary cardiac synovial sarcoma continuous with the mitral valve, which was accompanied by severe thrombocytopenia, and the platelet counts returned rapidly to a normal range early after tumour excision and without any special therapies. A 52-year-old male diagnosed with atrial myxoma with severe thrombocytopenia was admitted to our hospital. Blood analysis showed severe thrombocytopenia, whereas erythrocyte and leucocyte counts were within the normal range. A 50 × 35 mm mobile mass continuous with the mitral valve was found to be present in the left atrium upon echocardiography. Bone marrow aspiration and related examinations excluded thrombocytopenia caused by haematologic malignancies. The patient received a platelet transfusion, but platelet counts decreased quickly. Glucocorticoid therapy and immunoglobulin transfusion were also used, but were ineffective. Although the operation risk was high, tumour resection was performed via a median sternotomy with a cardiopulmonary bypass system. The postoperative pathological diagnosis was biphasic cardiac synovial sarcoma. Surprisingly, the platelet counts returned rapidly to a normal range early after tumour excision without any special therapies. The disappearance of the tumour from the annular region was confirmed on transthoracic echocardiography 6 days after surgery, and an FDG-PET scan performed 8 days after surgery showed no abnormal accumulation. Unfortunately, the patient died suddenly 6 months later without unknown cause. We report that a rare primary cardiac synovial sarcoma case continuous with the mitral valve caused severe thrombocytopenia; this provides further support for the awareness and diagnosis of primary cardiac synovial sarcoma. We also highlight that thrombocytopenia might be one rare symptom of a solid cardiac tumour but need more cases for support.
- 4 February 2019
- Published by Springer Nature in Journal of Cardiothoracic Surgery
- Vol. 14 (1) , 1-6
Open AccessPrimary cardiac synovial sarcoma that was continuous with the mitral valve caused severe thrombocytopenia: a case report
Primary cardiac sarcomas are exceedingly rare, and they commonly result in nonspecific constitutional symptoms such as shortness of breath, weight loss, and anaemia-related fatigue and malaise. However, thrombocytopenia has very rarely been reported in association with cardiac tumours, either benign or malignant. We report one case of primary cardiac synovial sarcoma continuous with the mitral valve, which was accompanied by severe thrombocytopenia, and the platelet counts returned rapidly to a normal range early after tumour excision and without any special therapies. A 52-year-old male diagnosed with atrial myxoma with severe thrombocytopenia was admitted to our hospital. Blood analysis showed severe thrombocytopenia, whereas erythrocyte and leucocyte counts were within the normal range. A 50 × 35 mm mobile mass continuous with the mitral valve was found to be present in the left atrium upon echocardiography. Bone marrow aspiration and related examinations excluded thrombocytopenia caused by haematologic malignancies. The patient received a platelet transfusion, but platelet counts decreased quickly. Glucocorticoid therapy and immunoglobulin transfusion were also used, but were ineffective. Although the operation risk was high, tumour resection was performed via a median sternotomy with a cardiopulmonary bypass system. The postoperative pathological diagnosis was biphasic cardiac synovial sarcoma. Surprisingly, the platelet counts returned rapidly to a normal range early after tumour excision without any special therapies. The disappearance of the tumour from the annular region was confirmed on transthoracic echocardiography 6 days after surgery, and an FDG-PET scan performed 8 days after surgery showed no abnormal accumulation. Unfortunately, the patient died suddenly 6 months later without unknown cause. We report that a rare primary cardiac synovial sarcoma case continuous with the mitral valve caused severe thrombocytopenia; this provides further support for the awareness and diagnosis of primary cardiac synovial sarcoma. We also highlight that thrombocytopenia might be one rare symptom of a solid cardiac tumour but need more cases for support.