Update README.md
Browse files
README.md
CHANGED
@@ -6,5 +6,32 @@ library_name: transformers
|
|
6 |
pipeline_tag: text-classification
|
7 |
tags:
|
8 |
- medical
|
|
|
9 |
---
|
10 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
6 |
pipeline_tag: text-classification
|
7 |
tags:
|
8 |
- medical
|
9 |
+
base_model: "emilyalsentzer/Bio_ClinicalBERT"
|
10 |
---
|
11 |
+
# BJH-perioperative-notes-bioClinicalBERT
|
12 |
+
This clinical foundational model is intended to predict post-operative surgical outcomes from clinical notes taken during perioperative care.
|
13 |
+
It was finetuned from the `emilyalsentzer/Bio_ClinicalBERT` model through a multi-task learning approach, spanning the following 6 outcomes:
|
14 |
+
|
15 |
+
- Death in 30 days
|
16 |
+
- Deep vein thrombosis (DVT)
|
17 |
+
- pulmonary embolism (PE)
|
18 |
+
- Pneumonia
|
19 |
+
- Acute Knee Injury
|
20 |
+
- delirium
|
21 |
+
|
22 |
+
## Dataset
|
23 |
+
|
24 |
+
We used 84,875 perioperative clinical notes from patients spanning the Barnes Jewish Hospital (BJH) system in St Louis, MO.
|
25 |
+
The following are the characteristics for the data:
|
26 |
+
|
27 |
+
- vocabulary size: 3203
|
28 |
+
- averaging words per clinical note: 8.9 words
|
29 |
+
- all single sentenced clinical notes
|
30 |
+
|
31 |
+
## How to use model
|
32 |
+
|
33 |
+
```
|
34 |
+
from transformers import AutoTokenizer, AutoModel
|
35 |
+
tokenizer = AutoTokenizer.from_pretrained("cja5553/BJH-perioperative-notes-bioClinicalBERT")
|
36 |
+
model = AutoModel.from_pretrained("cja5553/BJH-perioperative-notes-bioClinicalBERT")
|
37 |
+
```
|