Shang Zhao
Research Subjects
Shang Zhao has a total of 21 co-authors. Here are the top co-authors by number of shared publications:
- Ce Wang
School of Science, Sun Yat-Sen University, Shenzhen, China
2 shared publications - Haimiao Zhang
Beijing Information Science and Technology University
2 shared publications - Shaohua Kevin Zhou
University of Science and Technology of China, Suzhou, China
2 shared publications - Da An
China Satellite Network Digital Technology Co., Ltd., Xiong’an New Area, P. R. China
1 shared publication - Dong Liang
Shenzhen Institute of Advanced Technology, Chinese Academy of Sciences
1 shared publication - Dong Liang
Anhui University (China)
1 shared publication
Latest Publications
- 13 April 2026
- Published by World Scientific Pub Co Pte Ltd in Journal of Circuits, Systems and Computers
- Vol. 35 (16)
Multimodal Data Fusion-based Graph Neural Networks for Ship Trajectory Prediction
Global trade depends heavily on maritime transport, as ships carry more than [ ] of all international goods. The Automatic Identification System (AIS) is crucial for safe and efficient maritime navigation, providing real-time ship movement data. However, precise prediction of ship behavior is difficult because of complex environmental factors, human decisions and ship interactions. This paper introduces a novel method that uses multimodal data fusion with GNNs for improved trajectory prediction. Our model leverages AIS data in conjunction with oceanographic and meteorological data to create a ship interaction graph depicting spatial correlations. This graph-based methodology facilitates the modeling of ship dynamics, which are impacted by environmental variables and navigational regulations, by explicitly representing complex interdependencies. This proposed method yields a significant improvement in prediction accuracy, with consequential advantages for collision avoidance systems, route optimization strategies and maritime surveillance operations. Experimental findings reveal that our model achieves a higher level of accuracy than established time-series techniques in predicting ship behavior, thus setting a new benchmark.
- 13 April 2026
- Published by World Scientific Pub Co Pte Ltd in Journal of Circuits, Systems and Computers
- Vol. 35 (16)
Multimodal Data Fusion-based Graph Neural Networks for Ship Trajectory Prediction
Global trade depends heavily on maritime transport, as ships carry more than [ ] of all international goods. The Automatic Identification System (AIS) is crucial for safe and efficient maritime navigation, providing real-time ship movement data. However, precise prediction of ship behavior is difficult because of complex environmental factors, human decisions and ship interactions. This paper introduces a novel method that uses multimodal data fusion with GNNs for improved trajectory prediction. Our model leverages AIS data in conjunction with oceanographic and meteorological data to create a ship interaction graph depicting spatial correlations. This graph-based methodology facilitates the modeling of ship dynamics, which are impacted by environmental variables and navigational regulations, by explicitly representing complex interdependencies. This proposed method yields a significant improvement in prediction accuracy, with consequential advantages for collision avoidance systems, route optimization strategies and maritime surveillance operations. Experimental findings reveal that our model achieves a higher level of accuracy than established time-series techniques in predicting ship behavior, thus setting a new benchmark.
- 12 November 2024
- Published by Elsevier in Medical Image Analysis
- Vol. 99, 103387
LACOSTE: Exploiting stereo and temporal contexts for surgical instrument segmentation
- 12 November 2024
- Published by Elsevier in Medical Image Analysis
- Vol. 99, 103387
LACOSTE: Exploiting stereo and temporal contexts for surgical instrument segmentation
- Open Access
- 26 October 2023
- Published by Association for Computing Machinery (ACM)
- p. 7226-7235
Active CT Reconstruction with a Learned Sampling Policy
- 26 October 2023
- Published by Association for Computing Machinery (ACM)
- p. 7226-7235
Open AccessActive CT Reconstruction with a Learned Sampling Policy
- Open Access
Active CT Reconstruction with a Learned Sampling Policy
Computed tomography (CT) is a widely-used imaging technology that assists clinical decision-making with high-quality human body representations. To reduce the radiation dose posed by CT, sparse-view and limited-angle CT are developed with preserved image quality. However, these methods are still stuck with a fixed or uniform sampling strategy, which inhibits the possibility of acquiring a better image with an even reduced dose. In this paper, we explore this possibility via learning an active sampling policy that optimizes the sampling positions for patient-specific, high-quality reconstruction. To this end, we design an \textit{intelligent agent} for active recommendation of sampling positions based on on-the-fly reconstruction with obtained sinograms in a progressive fashion. With such a design, we achieve better performances on the NIH-AAPM dataset over popular uniform sampling, especially when the number of views is small. Finally, such a design also enables RoI-aware reconstruction with improved reconstruction quality within regions of interest (RoI's) that are clinically important. Experiments on the VerSe dataset demonstrate this ability of our sampling policy, which is difficult to achieve based on uniform sampling.
Open AccessActive CT Reconstruction with a Learned Sampling Policy
Computed tomography (CT) is a widely-used imaging technology that assists clinical decision-making with high-quality human body representations. To reduce the radiation dose posed by CT, sparse-view and limited-angle CT are developed with preserved image quality. However, these methods are still stuck with a fixed or uniform sampling strategy, which inhibits the possibility of acquiring a better image with an even reduced dose. In this paper, we explore this possibility via learning an active sampling policy that optimizes the sampling positions for patient-specific, high-quality reconstruction. To this end, we design an \textit{intelligent agent} for active recommendation of sampling positions based on on-the-fly reconstruction with obtained sinograms in a progressive fashion. With such a design, we achieve better performances on the NIH-AAPM dataset over popular uniform sampling, especially when the number of views is small. Finally, such a design also enables RoI-aware reconstruction with improved reconstruction quality within regions of interest (RoI's) that are clinically important. Experiments on the VerSe dataset demonstrate this ability of our sampling policy, which is difficult to achieve based on uniform sampling.
- Open Access
- 1 May 2022
- Published by College of Physicians and Surgeons Pakistan in Journal of the College of Physicians and Surgeons Pakistan
- Vol. 32 (05) , 632-638
Lox as the Potential Key Gene of Clear Cell Renal Carcinoma
Clear cell renal carcinoma (CCRC) is the most common type of renal carcinoma. We hope to find out the potential key genes playing important roles in CCRC genesis and progression by analysing the recent expression profiling by array from 2014 to 2016. In order to find out the differentially expressed genes (DEGs) between CCRC and normal renal tissue. Gene Ontology (GO) and Kyoto Encyclopedia of Genes, and Genomes (KEGG) pathway enrichment were carried out by the Database for Annotation, Visualization and Integrated Discovery (DAVID) v6.8. Protein-Protein Interaction (PPI) Networks Functional Enrichment Analysis of these DEGs was analyzed using the Search Tool for Recurring Instances of Neighbouring Genes (STRING). The results were then visualized by the software Cytoscape. The authors also used the online tool of Kaplan-Meier plotter survival analysis to assess the significance of the top ten genes in the prognosis of CCRC. A total of 192 DEGs were identified and the top ten key genes were picked out by the software Cytoscape. FN1, CXCR4, LOX, and PLG were then further screened out based on the overall survival analysis; SLC12A1 and LOX were screened out after the recurrence-free survival analysis. LOX was finally believed to be the most reliable prognostic factor since it has prognostic value for both overall survival and recurrence-free survival analysis. Our analysis suggests that LOX is the most reliable prognostic factor for CCRC patients. Key Words: Renal clear cell carcinoma, Microarray datasets, Bioinformatics approach, Prognostic factor, LOX.
- 1 May 2022
- Published by College of Physicians and Surgeons Pakistan in Journal of the College of Physicians and Surgeons Pakistan
- Vol. 32 (05) , 632-638
Open AccessLox as the Potential Key Gene of Clear Cell Renal Carcinoma
Clear cell renal carcinoma (CCRC) is the most common type of renal carcinoma. We hope to find out the potential key genes playing important roles in CCRC genesis and progression by analysing the recent expression profiling by array from 2014 to 2016. In order to find out the differentially expressed genes (DEGs) between CCRC and normal renal tissue. Gene Ontology (GO) and Kyoto Encyclopedia of Genes, and Genomes (KEGG) pathway enrichment were carried out by the Database for Annotation, Visualization and Integrated Discovery (DAVID) v6.8. Protein-Protein Interaction (PPI) Networks Functional Enrichment Analysis of these DEGs was analyzed using the Search Tool for Recurring Instances of Neighbouring Genes (STRING). The results were then visualized by the software Cytoscape. The authors also used the online tool of Kaplan-Meier plotter survival analysis to assess the significance of the top ten genes in the prognosis of CCRC. A total of 192 DEGs were identified and the top ten key genes were picked out by the software Cytoscape. FN1, CXCR4, LOX, and PLG were then further screened out based on the overall survival analysis; SLC12A1 and LOX were screened out after the recurrence-free survival analysis. LOX was finally believed to be the most reliable prognostic factor since it has prognostic value for both overall survival and recurrence-free survival analysis. Our analysis suggests that LOX is the most reliable prognostic factor for CCRC patients. Key Words: Renal clear cell carcinoma, Microarray datasets, Bioinformatics approach, Prognostic factor, LOX.
- 10 May 2018
- Published by Wiley in Computer Animation and Virtual Worlds
- Vol. 29 (5)
Accurate nonrigid 3D human body surface reconstruction using commodity depth sensors
In the last decade, 3D modeling techniques enjoyed a booming development in both hardware and software. High‐end hardware generates high fidelity results, but the cost is prohibitive, whereas consumer‐level devices generate plausible results for entertainment purposes but are not appropriate for medical uses. We present a cost‐effective and easy‐to‐use 3D body reconstruction system using consumer‐grade depth sensors, which provides reconstructed body shapes with a high degree of accuracy and reliability appropriate for medical applications. Our surface registration framework integrates the articulated motion assumption, global loop closure constraint, and a general as‐rigid‐as‐possible deformation model. To enhance the reconstruction quality, we propose a novel approach to accurately infer skeletal joints from anatomical data using multimodality registration. We further propose a supervised predictive model to infer the skeletal joints for arbitrary subjects independent from anatomical data reference. A rigorous validation test has been conducted on real subjects to evaluate the reconstruction accuracy and repeatability. Our system has the potential to make accurate body surface scanning systems readily available for medical professionals and the general public. The system can be used to obtain additional health data derived from 3D body shapes, such as the percentage of body fat.
- 10 May 2018
- Published by Wiley in Computer Animation and Virtual Worlds
- Vol. 29 (5)
Accurate nonrigid 3D human body surface reconstruction using commodity depth sensors
In the last decade, 3D modeling techniques enjoyed a booming development in both hardware and software. High‐end hardware generates high fidelity results, but the cost is prohibitive, whereas consumer‐level devices generate plausible results for entertainment purposes but are not appropriate for medical uses. We present a cost‐effective and easy‐to‐use 3D body reconstruction system using consumer‐grade depth sensors, which provides reconstructed body shapes with a high degree of accuracy and reliability appropriate for medical applications. Our surface registration framework integrates the articulated motion assumption, global loop closure constraint, and a general as‐rigid‐as‐possible deformation model. To enhance the reconstruction quality, we propose a novel approach to accurately infer skeletal joints from anatomical data using multimodality registration. We further propose a supervised predictive model to infer the skeletal joints for arbitrary subjects independent from anatomical data reference. A rigorous validation test has been conducted on real subjects to evaluate the reconstruction accuracy and repeatability. Our system has the potential to make accurate body surface scanning systems readily available for medical professionals and the general public. The system can be used to obtain additional health data derived from 3D body shapes, such as the percentage of body fat.