Jiajun Xiang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Jiajun Xiang has a total of 16 co-authors. Here are the top co-authors by number of shared publications:
- Aihua Mao
School of Computer Science and Engineering, South China University of Technology, Guangzhou, China
2 shared publications - Lequan Yu
The University of Hong Kong
2 shared publications - Tianye Niu
School of Information Science and Technology, University of Science and Technology of China, Hefei, China
2 shared publications - Xianghong Wang
College of Food Science and Technology, Hebei Agricultural University, Baoding, 071000, China; Laboratory of Dietary Component Interactions and Precision Nutrition, Baoding, 071000, China
2 shared publications - Baiying Lei
Shenzhen University
1 shared publication - Hongbin Han
Peking University Health Science Center
1 shared publication
Latest Publications
- 20 January 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Computational Imaging
- Vol. 11 (25730436) , 349-361
Clip-Driven Universal Model for Multi-Material Decomposition in Dual-Energy CT
Dual-energy computed tomography (DECT) offers quantitative insights and facilitates material decomposition, aiding in precise diagnosis and treatment planning. However, existing methods for material decomposition, often tailored to specific material types, need more generalizability and increase computational load with each additional material. We propose a CLIP-Driven Universal Model for adaptive Multi-Material Decomposition (MMD) to tackle this challenge. This model utilizes the semantic capabilities of text embeddings from Contrastive Language-Image Pre-training (CLIP), allowing a single network to manage structured feature embedding for multiple materials. A novel Siamese encoder and differential map fusion technique have also been integrated to enhance the decomposition accuracy while maintaining robustness across various conditions. Experiments on the simulated and physical patient studies have evidenced our model's superiority over traditional methods. Notably, it has significantly improved the Dice Similarity Coefficient—4.1%. These results underscore the potential of our network in clinical MMD applications, suggesting a promising avenue for enhancing DECT imaging analysis.
- 20 January 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Computational Imaging
- Vol. 11 (25730436) , 349-361
Clip-Driven Universal Model for Multi-Material Decomposition in Dual-Energy CT
Dual-energy computed tomography (DECT) offers quantitative insights and facilitates material decomposition, aiding in precise diagnosis and treatment planning. However, existing methods for material decomposition, often tailored to specific material types, need more generalizability and increase computational load with each additional material. We propose a CLIP-Driven Universal Model for adaptive Multi-Material Decomposition (MMD) to tackle this challenge. This model utilizes the semantic capabilities of text embeddings from Contrastive Language-Image Pre-training (CLIP), allowing a single network to manage structured feature embedding for multiple materials. A novel Siamese encoder and differential map fusion technique have also been integrated to enhance the decomposition accuracy while maintaining robustness across various conditions. Experiments on the simulated and physical patient studies have evidenced our model's superiority over traditional methods. Notably, it has significantly improved the Dice Similarity Coefficient—4.1%. These results underscore the potential of our network in clinical MMD applications, suggesting a promising avenue for enhancing DECT imaging analysis.
- 29 December 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Computational Imaging
- Vol. 10 (25730436) , 28-42
DER-GAN: Dual-Energy Recovery GAN for Conebeam CT
Dual-energy cone-beam computed tomography (DE-CBCT) integrates dual-energy imaging seamlessly into the CBCT system, offering a practical solution for real-time clinical applications in treatment rooms. Traditional DE-CBCT systems often rely on intricate hardware or dual scanning, imposing significant constraints on the broader application of dual-energy CT (DECT) in CBCT machines. In this study, we introduce an innovative GAN-based single-scan dual-energy CBCT reconstruction strategy designed for DE-CBCT systems, effectively reducing acquisition time compared to conventional two-scan DE-CBCT approaches. Our approach leverages a strip-type modulator positioned in front of the detector, enabling the acquisition of spectra-mixed dual-energy projections in a single scan by modulating specific areas on the detector. The obtained incomplete dual-energy projections undergo precise recovery through our designed dual-energy recovery GAN (DER-GAN). DER-GAN adeptly extracts complementary spectra and ensures consistency in anatomical information between high and low-energy projections. Through qualitative and quantitative analyses, DER-GAN demonstrates commendable performance in terms of CT number accuracy and preservation of anatomical details. Furthermore, in the realm of DECT applications, particularly in multi-material decomposition, DER-GAN's reconstructed images exhibit promising potential for clinical CBCT applications. This pioneering approach represents a significant stride toward efficient and practical integration of dual-energy imaging into the CBCT paradigm.
- 29 December 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Computational Imaging
- Vol. 10 (25730436) , 28-42
DER-GAN: Dual-Energy Recovery GAN for Conebeam CT
Dual-energy cone-beam computed tomography (DE-CBCT) integrates dual-energy imaging seamlessly into the CBCT system, offering a practical solution for real-time clinical applications in treatment rooms. Traditional DE-CBCT systems often rely on intricate hardware or dual scanning, imposing significant constraints on the broader application of dual-energy CT (DECT) in CBCT machines. In this study, we introduce an innovative GAN-based single-scan dual-energy CBCT reconstruction strategy designed for DE-CBCT systems, effectively reducing acquisition time compared to conventional two-scan DE-CBCT approaches. Our approach leverages a strip-type modulator positioned in front of the detector, enabling the acquisition of spectra-mixed dual-energy projections in a single scan by modulating specific areas on the detector. The obtained incomplete dual-energy projections undergo precise recovery through our designed dual-energy recovery GAN (DER-GAN). DER-GAN adeptly extracts complementary spectra and ensures consistency in anatomical information between high and low-energy projections. Through qualitative and quantitative analyses, DER-GAN demonstrates commendable performance in terms of CT number accuracy and preservation of anatomical details. Furthermore, in the realm of DECT applications, particularly in multi-material decomposition, DER-GAN's reconstructed images exhibit promising potential for clinical CBCT applications. This pioneering approach represents a significant stride toward efficient and practical integration of dual-energy imaging into the CBCT paradigm.