minhanhto09 commited on
Commit
38e5f8e
1 Parent(s): 28da202

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +5 -4
README.md CHANGED
@@ -32,7 +32,7 @@ This [repository](https://github.com/PathologyDataScience/BCSS) contains the nec
32
 
33
  - Amgad, Mohamed, et al. "NuCLS: A Scalable Crowdsourcing Approach and Dataset for Nucleus Classification and Segmentation in Breast Cancer." GigaScience, vol. 11, 2022, https://doi.org/10.1093/gigascience/giac037. Accessed 18 Mar. 2024.
34
 
35
- ![](https://huggingface.co/datasets/minhanhto09/NuCLS_dataset/resolve/main/fig1.PNG)
36
 
37
 
38
  ## Accessing the Data
@@ -71,7 +71,7 @@ A single dataset entry contains the following details:
71
 
72
  - `mask_image`: A mask image with each nucleus labeled. Class labels are encoded in the first channel. The second and third channels are used to create a unique identifier for each nucleus. The field of view (gray area) is marked to delineate the annotated region.
73
 
74
- [This file](https://drive.google.com/file/d/1vT6ZG1s3IQkB9suI21qgzF2N5zM8z0qd/view?usp=sharing) contains the nucleus label encoding, including a special 'fov' code encoding the intended annotation region.
75
 
76
  - `visualization_image`: A visualization image that overlays the RGB and mask images to assist in interpretability.
77
 
@@ -89,7 +89,7 @@ A single dataset entry contains the following details:
89
 
90
  - `coords_x`, `coords_y`: The specific boundary coordinates of the nucleus.
91
 
92
- ![](https://huggingface.co/datasets/minhanhto09/NuCLS_dataset/resolve/main/Image%2017-03-2024%20at%2023.54.jpeg)
93
 
94
  ### Data Split
95
 
@@ -115,7 +115,8 @@ The dataset is divided into the following folds:
115
 
116
  - `train_fold_999`: 21 examples
117
  - `test_fold_999`: 7 examples
118
- Note that the debug configuration utilizes these particular folds `train_fold_999` and `test_fold_999` due to their smaller numbers of examples.
 
119
 
120
  ## Usage Example
121
 
 
32
 
33
  - Amgad, Mohamed, et al. "NuCLS: A Scalable Crowdsourcing Approach and Dataset for Nucleus Classification and Segmentation in Breast Cancer." GigaScience, vol. 11, 2022, https://doi.org/10.1093/gigascience/giac037. Accessed 18 Mar. 2024.
34
 
35
+ ![](https://huggingface.co/datasets/minhanhto09/NuCLS_dataset/resolve/main/Images/fig1.PNG)
36
 
37
 
38
  ## Accessing the Data
 
71
 
72
  - `mask_image`: A mask image with each nucleus labeled. Class labels are encoded in the first channel. The second and third channels are used to create a unique identifier for each nucleus. The field of view (gray area) is marked to delineate the annotated region.
73
 
74
+ [This file](https://huggingface.co/datasets/minhanhto09/NuCLS_dataset/resolve/main/Images/fig2.png) contains the nucleus label encoding, including a special 'fov' code encoding the intended annotation region.
75
 
76
  - `visualization_image`: A visualization image that overlays the RGB and mask images to assist in interpretability.
77
 
 
89
 
90
  - `coords_x`, `coords_y`: The specific boundary coordinates of the nucleus.
91
 
92
+ ![](https://huggingface.co/datasets/minhanhto09/NuCLS_dataset/resolve/main/Images/fig3.jpeg)
93
 
94
  ### Data Split
95
 
 
115
 
116
  - `train_fold_999`: 21 examples
117
  - `test_fold_999`: 7 examples
118
+
119
+ Note that the debug configuration utilizes these particular folds `train_fold_999` and `test_fold_999` due to their smaller numbers of examples.
120
 
121
  ## Usage Example
122