Skip to content

ronenh24/mri_dictionary_learning

Repository files navigation

MRI Dictionary Learning

Author

Ronen Huang

Timeline

April 2025 to May 2025

Pipeline

  1. Resize 121 x 145 x 121 MRI images to 128 x 128.
  2. Concatenate 10 images side by side to create $\bm{Y} \in \mathbb{R}^{128 \times 1280}$ data matrix.
  3. Solve $\min_{\bm{q}} \frac{1}{m}|\bm{q}^T\bm{Y}|_1$ such that $|\bm{q}|_2 = 1$ with $R = 128 \log (128)$ rounds to create orthogonal dictionary $\bm{A} = {\bm{a}_1, \cdots, \bm{a}_n} \in O(n)$.
  4. Solve $\min_{\bm{A}, \bm{X}} \frac{1}{2}|\bm{Y} - \bm{A}\bm{X}|_F^2 + \lambda |\bm{X}|_1$ such that $\bm{A} \in O(n)$.

The implementation can be see in dictionary_learning.ipynb.

MRI Reconstruction Example

About

MRI Orthogonal Dictionary Learning.

Resources

License

Code of conduct

Stars

Watchers

Forks