I participate various research projects since college.
Fast knee quantitative mri
2023.10 – present
This is an ongoing project focus on how to accelerate knee cartilage quantitative MRI.
Automated Post-Processing for Compositional Knee MRI
2023.07 – present
This work-in-progress pipeline gathers various advanced, deep-learning-based techniques for generating regional compositional reports for quantitative analysis of knee OA.
Related publications
- Zhong J, Yao Y, Xiao F, Ho KKW, Ong MTY, Griffith JF, Chen W. A systematic automated post-processing approach for quantitative analysis of 3D T1ρ knee MRI. In: Proceeding of the International Society for Magnetic Resonance in Medicine. Singapore; 2024.
- Yao Y, Zhong J, Zhang L, Khan S, Chen W. CartiMorph: A framework for automated knee articular cartilage morphometrics. Medical Image Analysis. 2024 Jan 1;91:103035.
Knee Osteoarthritis Phenotyping with Unsupervised Domain Adaptation
2021.08 – 2023.10
We developed a deep-learning method that transfers latent knowledge from a large public dataset to our locally collected small dataset. The proposed method was evaluated with a knee OA phenotyping task on 3D TSE MRI. Related papers were published in ISMRM and QIMS.
Related publications
- Zhong J, Yao Y, Cahill DG, Xiao F, Li S, Lee J, Ho KKW, Ong MTY, Griffith JF, Chen W. Unsupervised domain adaptation for automated knee osteoarthritis phenotype classification. Quantitative Imaging in Medicine and Surgery. 2023 Oct 17;13(11):7444458–7458.
- Zhong J, Yao Y, Cahill DG, Xiao F, Li S, Lee J, Ho KKW, Ong MTY, Griffith JF, Chen W. Unsupervised Domain Adaptation for Automated Knee Osteoarthritis Phenotype Classification. In: Proceeding of the International Society for Magnetic Resonance in Medicine. Toronto, ON, Canada; 2023.
- Zhong J, Yao Y, Khan S, Xiao F, Cahill DG, Griffith JF, Chen W. Knee Osteoarthritis: Automatic Grading with Deep Learning. In: Proceeding of the International Society for Magnetic Resonance in Medicine. London, England, UK; 2022.
Knee MRI Segmentation
2020.08 – 2021.08
We developed a novel deep learning structure on MRI to segment knee structures (femoral bone, femoral cartilage, tibial bone, and tibial cartilage). This novel deep learning structure improves segmentation performance at edge slices where ROI areas are small.
Related publications
- Li S, Khan S, Xiao F, Zhao S, Zhong J, Cahill DG, Griffith JF, Chen W. Unsupervised Domain Adaptation via CycleGAN for knee joint Segmentation in MR Images. In: Proceeding of the International Society for Magnetic Resonance in Medicine. London, England, UK; 2022.