Jiachen Ji
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
Research Subjects
Jiachen Ji has a total of 26 co-authors. Here are the top co-authors by number of shared publications:
Rui Li
Center for Biomedical Imaging Research, School of Biomedical Engineering, Tsinghua University Beijing China
3 shared publications- Changchun Liu
Department of Radiology, The Fifth Medical Center of Chinese PLA General Hospital, Beijing, China
2 shared publications - Di Zhao
Senior Department of Hepatology, The Fifth Medical Center of Chinese PLA General Hospital, Beijing 100039, China
2 shared publications - Hailong Yu
Department of Radiology, The Fifth Medical Center of Chinese PLA General Hospital, Beijing 100039, China
2 shared publications - Jianming Cai
Department of Radiology, The Fifth Medical Center of Chinese PLA General Hospital, Beijing, China
2 shared publications - Jinghui Dong
Department of Radiology, The Fifth Medical Center of Chinese PLA General Hospital, Beijing 100039, China
2 shared publications
Latest Publications
- 23 April 2026
- Published by Wiley in Magnetic Resonance in Medicine
Reconstruction of Under‐Sampled Images and Concurrent Optimization of Sampling Masks for 3D Carotid Simultaneous Non‐Contrast Angiography and intraPlaque Hemorrhage MRI With Model Based Deep Learning Architecture ( deepSNAP )
Purpose: To improve the imaging efficiency of 3D carotid simultaneous noncontrast angiography and intraplaque hemorrhage (SNAP) MRI by reconstruction of under‐sampled images and concurrent optimization of sampling masks for the two shots of SNAP respectively. Methods: A model‐based deep learning architecture (deepSNAP) was proposed to recover under‐sampled 3D carotid SNAP MRI. Sampling locations on the ky–kz plane were parameterized to enable respective optimization of the sampling masks for the two shots. A dataset of 100 3D carotid SNAP MRI scans was utilized (80 training, 20 test). Image recovery performance was compared with established techniques under different acceleration factors. Lumen area measurement accuracy and intraplaque hemorrhage (IPH) identification were evaluated at 6× acceleration. Prospective feasibility was assessed in 10 healthy volunteers with quantitative comparison against established methods. Results: deepSNAP exhibited superior image recovery performance on the test set, surpassing all comparison methods. Optimized masks generated by deepSNAP improved reconstruction performance across all comparison methods. High agreement between reconstructed images and original images was observed for lumen area measurement (ICC = 0.995, 95% CI: 0.993–0.996) and IPH detection (Cohen's κ = 0.976, 95% CI: 0.943–1.000). In the prospective experiment, deepSNAP achieved promising image quality and structural fidelity. Conclusion: The deepSNAP model achieved under‐sampled image reconstruction and simultaneous sampling mask optimization for SNAP, ensured the clinical practicability of the reconstructed images, and demonstrated preliminary technical feasibility in a prospective setting.
- 23 April 2026
- Published by Wiley in Magnetic Resonance in Medicine
Reconstruction of Under‐Sampled Images and Concurrent Optimization of Sampling Masks for 3D Carotid Simultaneous Non‐Contrast Angiography and intraPlaque Hemorrhage MRI With Model Based Deep Learning Architecture ( deepSNAP )
Purpose: To improve the imaging efficiency of 3D carotid simultaneous noncontrast angiography and intraplaque hemorrhage (SNAP) MRI by reconstruction of under‐sampled images and concurrent optimization of sampling masks for the two shots of SNAP respectively. Methods: A model‐based deep learning architecture (deepSNAP) was proposed to recover under‐sampled 3D carotid SNAP MRI. Sampling locations on the ky–kz plane were parameterized to enable respective optimization of the sampling masks for the two shots. A dataset of 100 3D carotid SNAP MRI scans was utilized (80 training, 20 test). Image recovery performance was compared with established techniques under different acceleration factors. Lumen area measurement accuracy and intraplaque hemorrhage (IPH) identification were evaluated at 6× acceleration. Prospective feasibility was assessed in 10 healthy volunteers with quantitative comparison against established methods. Results: deepSNAP exhibited superior image recovery performance on the test set, surpassing all comparison methods. Optimized masks generated by deepSNAP improved reconstruction performance across all comparison methods. High agreement between reconstructed images and original images was observed for lumen area measurement (ICC = 0.995, 95% CI: 0.993–0.996) and IPH detection (Cohen's κ = 0.976, 95% CI: 0.943–1.000). In the prospective experiment, deepSNAP achieved promising image quality and structural fidelity. Conclusion: The deepSNAP model achieved under‐sampled image reconstruction and simultaneous sampling mask optimization for SNAP, ensured the clinical practicability of the reconstructed images, and demonstrated preliminary technical feasibility in a prospective setting.
- Open Access
- 12 August 2024
- Published by Elsevier in Journal of Clinical and Experimental Hepatology
- Vol. 15 (1) , 102403
Prediction Modelling for Gastroesophageal Variceal Bleeding in Patients With Chronic Hepatitis B Using Four-dimensional Flow MRI
- 12 August 2024
- Published by Elsevier in Journal of Clinical and Experimental Hepatology
- Vol. 15 (1) , 102403
Open AccessPrediction Modelling for Gastroesophageal Variceal Bleeding in Patients With Chronic Hepatitis B Using Four-dimensional Flow MRI
- Open Access
- 26 April 2024
- Published by Bentham Science Publishers Ltd. in Current Medical Imaging
- Vol. 20, 1-7
4D Flow MRI of Portal Vein Hemodynamics in Healthy Volunteers and Patients with Chronic Liver Disease
Aim:: To identify age-matched healthy volunteers, non-cirrhotic chronic liver disease (CLD) and cirrhotic patients based on portal hemodynamic parameters using 4D flow MRI. Methods:: A total of 10 age-matched healthy volunteers and 69 CLD patients were enrolled and underwent 4D flow MRI prospectively. 4D flow MR images were processed by an MD in biomedical engineering working on the GTFlow platform. Portal hemodynamic parameters include net flow (mL/cycle), flow volume per second through the lumen (mL/sec), average flow velocity (cm/sec), and maximum flow velocity (cm/sec). The difference in portal hemodynamic parameters of 4D flow MRI was compared among healthy volunteers, non-cirrhotic CLD patients and patients with cirrhosis by one-way ANOVA or Kruskal-Wallis nonparametric test and post hoc tests. Results:: 10 CLD patients without cirrhosis and 56 patients with cirrhosis were eventually included, along with 10 healthy volunteers who were divided into three groups. 3 patients with cirrhosis whose image quality did not meet the requirements were excluded. There were no significant differences in portal hemodynamic parameters among the three groups except portal average velocity (P > 0.05). There was no statistical difference in all portal hemodynamic parameters of 4D flow MRI between healthy volunteers and patients with cirrhosis (P > 0.05). There were significant differences in portal average velocity between non-cirrhotic CLD patients, healthy volunteers and patients with cirrhosis, respectively (11.44±3.93 vs 8.10±2.66, P=0.013; 11.44±3.93 vs 8.60±2.22, P=0.007). Conclusion:: Portal average velocity obtained by 4D flow MRI can be an auxiliary means to identify cirrhosis in patients with CLD.
- 26 April 2024
- Published by Bentham Science Publishers Ltd. in Current Medical Imaging
- Vol. 20, 1-7
Open Access4D Flow MRI of Portal Vein Hemodynamics in Healthy Volunteers and Patients with Chronic Liver Disease
Aim:: To identify age-matched healthy volunteers, non-cirrhotic chronic liver disease (CLD) and cirrhotic patients based on portal hemodynamic parameters using 4D flow MRI. Methods:: A total of 10 age-matched healthy volunteers and 69 CLD patients were enrolled and underwent 4D flow MRI prospectively. 4D flow MR images were processed by an MD in biomedical engineering working on the GTFlow platform. Portal hemodynamic parameters include net flow (mL/cycle), flow volume per second through the lumen (mL/sec), average flow velocity (cm/sec), and maximum flow velocity (cm/sec). The difference in portal hemodynamic parameters of 4D flow MRI was compared among healthy volunteers, non-cirrhotic CLD patients and patients with cirrhosis by one-way ANOVA or Kruskal-Wallis nonparametric test and post hoc tests. Results:: 10 CLD patients without cirrhosis and 56 patients with cirrhosis were eventually included, along with 10 healthy volunteers who were divided into three groups. 3 patients with cirrhosis whose image quality did not meet the requirements were excluded. There were no significant differences in portal hemodynamic parameters among the three groups except portal average velocity (P > 0.05). There was no statistical difference in all portal hemodynamic parameters of 4D flow MRI between healthy volunteers and patients with cirrhosis (P > 0.05). There were significant differences in portal average velocity between non-cirrhotic CLD patients, healthy volunteers and patients with cirrhosis, respectively (11.44±3.93 vs 8.10±2.66, P=0.013; 11.44±3.93 vs 8.60±2.22, P=0.007). Conclusion:: Portal average velocity obtained by 4D flow MRI can be an auxiliary means to identify cirrhosis in patients with CLD.
- Open Access
- 13 November 2023
- Published by Elsevier in Science of The Total Environment
- Vol. 908, 168531
Soil bacterial succession with different land uses along a millennial chronosequence derived from the Yangtze River flood plain
- 13 November 2023
- Published by Elsevier in Science of The Total Environment
- Vol. 908, 168531
Open AccessSoil bacterial succession with different land uses along a millennial chronosequence derived from the Yangtze River flood plain