English
medical
brain-data
mri
jesseab commited on
Commit
5a297a9
·
1 Parent(s): 8c84e52

License change to apache-2.0

Browse files
Files changed (1) hide show
  1. README.md +8 -7
README.md CHANGED
@@ -62,9 +62,10 @@ python inference_brain2vec.py \
62
  ```
63
 
64
  # Methods
65
- transform:
66
- (80, 96, 80)
67
- pixdim=2
 
68
  10 epochs
69
  max_batch_size: int = 2,
70
  batch_size: int = 16,
@@ -76,13 +77,13 @@ Pinaya
76
 
77
  # Citation
78
  ```
79
- @dataset{Radiata-Brain-Structure,
80
  author = {Jesse Brown and Clayton Young},
81
- title = {Brain-Structure: Processed Structural MRI Brain Scans Across the Lifespan},
82
  year = {2025},
83
- url = {https://huggingface.co/datasets/radiata-ai/brain-structure},
84
  note = {Version 1.0},
85
- publisher = {Hugging Face}
86
  }
87
  ```
88
 
 
62
  ```
63
 
64
  # Methods
65
+ Input scan image dimensions are 113x137x113, 1.5mm^3 resolution, aligned to MNI152 space (see [radiata-ai/brain-structure](https://huggingface.co/datasets/radiata-ai/brain-structure)).
66
+
67
+ The image transform crops to 80 x 96 x 80, 2mm^3 resolution, and scales image intensity to range [0,1]. Images are flattened to 614400-length 1D vectors.
68
+
69
  10 epochs
70
  max_batch_size: int = 2,
71
  batch_size: int = 16,
 
77
 
78
  # Citation
79
  ```
80
+ @misc{Radiata-Brain2Vec,
81
  author = {Jesse Brown and Clayton Young},
82
+ title = {brain2vec_PCA: A VAE Model for Brain Structure T1 MRIs},
83
  year = {2025},
84
+ url = {https://huggingface.co/radiata-ai/brain2vec},
85
  note = {Version 1.0},
86
+ publisher = {Hugging Face}
87
  }
88
  ```
89