Brain MRI
Tumor Segmentation & Image Generation
Generating useful MR image information and segmenting lesion regions for brainstem glioma diagnosis under limited or incomplete imaging conditions.
Research Profile
I specialize in mathematics and applied mathematics, machine learning, computer vision, medical image processing, tumor segmentation, image fusion, and 3D cardiac reconstruction.
My current work connects segmentation, synthesis, reconstruction, and robust learning from incomplete medical imaging data.
- Self-supervised medical image segmentation
- Brainstem glioma image synthesis and segmentation
- Multi-contrast cardiac MRI reconstruction
- Missing-modality learning and robust representation
Selected Research Themes
Projects
Cardiac MRI
3D Cardiac Reconstruction
Developing efficient multi-contrast CMR reconstruction methods to reduce scan burden while preserving clinically meaningful structural detail.
Missing Modality
Diffusion-Guided Segmentation
Improving segmentation robustness when MRI modalities are missing or unavailable through diffusion-guided multi-modal learning.
View Streamlit demoResearch Output
Publications
A multi-task generative model for simultaneous post-contrast MR image synthesis and brainstem glioma segmentation
Generates post-contrast MR images while reducing exposure to gadolinium-based contrast agents and delineating brainstem glioma lesions.
Read DOIA Multi-Contrast Cardiac MRI Reconstruction Method Using an Advanced Unrolled Network Architecture
Accepted research on accelerated, high-quality multi-contrast CMR reconstruction.
Capabilities
Technical Stack
Python
PyTorch
TensorFlow
MONAI
OpenCV
Medical Imaging
Segmentation
Reconstruction
LaTeX
Git
Contact
Research Collaboration
For research discussion, academic collaboration, or project review, use the academic identity and GitHub links below.