Update README.md
Browse files
README.md
CHANGED
@@ -1,3 +1,32 @@
|
|
1 |
-
---
|
2 |
-
license: apache-2.0
|
3 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: apache-2.0
|
3 |
+
tags:
|
4 |
+
- vision
|
5 |
+
- image-classification
|
6 |
+
---
|
7 |
+
|
8 |
+
### (Lung) Lung Adenocarcinoma
|
9 |
+
|
10 |
+
This model can additionally be run on our [pathology reports platform](https://www.pathologyreports.ai/marketplace/browse/40e62b8e-f41b-4ff1-90d8-03940c11068a)
|
11 |
+
|
12 |
+
Credits: Dr. Assem Alrumeh
|
13 |
+
|
14 |
+
### Introduction
|
15 |
+
|
16 |
+
This H&E lung adenocarcinoma tissue classifier was developed using transfer learning on a histology optimized version of the VGG19 CNN [(DOI: 10.1038/s42256-019-0068-6)](https://doi.org/10.1038/s42256-019-0068-6) and trained to recognize lung adenocarcinoma and other surrounding tissue elements.
|
17 |
+
Annotations were carried out on batches of image tiles (dimensions: 256 x 256 um) grouped using image-based clustering [(HAVOC, DOI: 10.1126/sciadv.adg1894)](https://doi.org/10.1126/sciadv.adg1894) from 10 publicly available TCGA-LUAD H&E-stained whole slide images. Validation testing was carried out on non-overlapping cases from TCGA.
|
18 |
+
|
19 |
+
|
20 |
+
### Classes
|
21 |
+
1. Adenocarcinoma
|
22 |
+
2. Blank space
|
23 |
+
3. Fibroconnective and stromal elements
|
24 |
+
4. Lung parenchyma
|
25 |
+
5. Lymphoid tissue
|
26 |
+
6. Necrosis
|
27 |
+
|
28 |
+
This information can be found in the inference.json file
|
29 |
+
|
30 |
+
### Evaluation Metrics
|
31 |
+
|
32 |
+
Classifier validation can be found on the [pathology reports platform](https://www.pathologyreports.ai/marketplace/browse/40e62b8e-f41b-4ff1-90d8-03940c11068a)
|