Update README.md
Browse files
README.md
CHANGED
|
@@ -70,40 +70,28 @@ In 3D, the following applicable MLP training data files are:
|
|
| 70 |
|
| 71 |
### CNNs
|
| 72 |
|
|
|
|
| 73 |
|
|
|
|
|
|
|
| 74 |
|
| 75 |
-
|
| 76 |
-
|
| 77 |
-
## Dataset Creation
|
| 78 |
-
|
| 79 |
-
### Curation Rationale
|
| 80 |
|
| 81 |
-
|
| 82 |
|
| 83 |
-
|
| 84 |
|
| 85 |
#### Annotation process
|
| 86 |
|
| 87 |
Each simulation had its permeability calcualted using Darcy's Law. For the MLP networks, this became a column labeled `k`, and for each CNN network was embedded into each image's EXIF metadata as a user comment.
|
| 88 |
|
| 89 |
-
[More Information Needed]
|
| 90 |
-
|
| 91 |
-
|
| 92 |
#### Personal and Sensitive Information
|
| 93 |
|
| 94 |
This dataset does not include any personal and sensitive information.
|
| 95 |
|
| 96 |
## Bias, Risks, and Limitations
|
| 97 |
|
| 98 |
-
|
| 99 |
-
|
| 100 |
-
[More Information Needed]
|
| 101 |
-
|
| 102 |
-
### Recommendations
|
| 103 |
-
|
| 104 |
-
<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
|
| 105 |
-
|
| 106 |
-
Users should be made aware of the risks, biases and limitations of the dataset. More information needed for further recommendations.
|
| 107 |
|
| 108 |
## Citation [optional]
|
| 109 |
|
|
|
|
| 70 |
|
| 71 |
### CNNs
|
| 72 |
|
| 73 |
+
Each CNN data source is in two file formats:
|
| 74 |
|
| 75 |
+
- `.lmdb`: Stores the images in Lightning Memory-Mapped Database (LMDB) format for efficient network processing.
|
| 76 |
+
- `.tar.gz`: Regular storage of each raw image, with corresponding permeabilities embedded in each image's EXIF metadata as a user commment.
|
| 77 |
|
| 78 |
+
The directories ending in only `_crop` indicate that the image is rectangular (i.e., only white space has been removed). File names that end in `center_crop` indicate that a 512x512 pixel center crop has been taken from the original image (the original image being contained in the directories that end in `_crop`).
|
|
|
|
|
|
|
|
|
|
|
|
|
| 79 |
|
| 80 |
+
The directories that contain the word `test` in them correspond to the evaluation dataset and were used to benchmark the CNN models.
|
| 81 |
|
| 82 |
+
No 3D data was used for the CNNs.
|
| 83 |
|
| 84 |
#### Annotation process
|
| 85 |
|
| 86 |
Each simulation had its permeability calcualted using Darcy's Law. For the MLP networks, this became a column labeled `k`, and for each CNN network was embedded into each image's EXIF metadata as a user comment.
|
| 87 |
|
|
|
|
|
|
|
|
|
|
| 88 |
#### Personal and Sensitive Information
|
| 89 |
|
| 90 |
This dataset does not include any personal and sensitive information.
|
| 91 |
|
| 92 |
## Bias, Risks, and Limitations
|
| 93 |
|
| 94 |
+
Each simulation has flow entering with a fully-developed profile with a peak inlet velocity of 20 mm/s. While based on literature values for blood flow, it is recommended that ranges of values and angles of this blood flow be explored.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 95 |
|
| 96 |
## Citation [optional]
|
| 97 |
|