Update README.md
Browse files
README.md
CHANGED
@@ -1,13 +1,11 @@
|
|
1 |
---
|
2 |
license: mit
|
3 |
-
---
|
4 |
-
Certainly! Below is a modified version of the text you provided, structured to serve as a model card for the "BioMed-VITAL" model. This format typically includes sections that outline the model's purpose, data used for training, and performance metrics:
|
5 |
-
|
6 |
-
---
|
7 |
|
8 |
#### Overview
|
9 |
BioMed-VITAL is a multimodal foundation model specifically tuned for biomedical applications. It leverages visual and textual data to improve understanding and reasoning within the biomedical domain.
|
10 |
|
|
|
|
|
11 |
#### Model Training
|
12 |
The training of BioMed-VITAL involved two key stages, both incorporating clinician preferences to ensure the relevance and quality of the training data:
|
13 |
|
@@ -19,4 +17,14 @@ The training of BioMed-VITAL involved two key stages, both incorporating clinici
|
|
19 |
The effectiveness of BioMed-VITAL was demonstrated through significant improvements in two key areas:
|
20 |
|
21 |
- **Open Visual Chat:** The model showed a relative improvement of 18.5%, indicating enhanced capabilities in engaging in visual dialogues pertinent to biomedical contexts.
|
22 |
-
- **Medical Visual Question Answering (VQA):** BioMed-VITAL achieved a win rate of up to 81.73% in this domain, showcasing its superior performance in interpreting and responding to complex medical imagery and queries.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
---
|
2 |
license: mit
|
|
|
|
|
|
|
|
|
3 |
|
4 |
#### Overview
|
5 |
BioMed-VITAL is a multimodal foundation model specifically tuned for biomedical applications. It leverages visual and textual data to improve understanding and reasoning within the biomedical domain.
|
6 |
|
7 |
+

|
8 |
+
|
9 |
#### Model Training
|
10 |
The training of BioMed-VITAL involved two key stages, both incorporating clinician preferences to ensure the relevance and quality of the training data:
|
11 |
|
|
|
17 |
The effectiveness of BioMed-VITAL was demonstrated through significant improvements in two key areas:
|
18 |
|
19 |
- **Open Visual Chat:** The model showed a relative improvement of 18.5%, indicating enhanced capabilities in engaging in visual dialogues pertinent to biomedical contexts.
|
20 |
+
- **Medical Visual Question Answering (VQA):** BioMed-VITAL achieved a win rate of up to 81.73% in this domain, showcasing its superior performance in interpreting and responding to complex medical imagery and queries.
|
21 |
+
|
22 |
+
## Case Study
|
23 |
+
### Biomedical Visual Instruction-Following Example
|
24 |
+

|
25 |
+
### Biomedical VQA Benchmark
|
26 |
+

|
27 |
+
### Clinician Annotation Examples
|
28 |
+

|
29 |
+
|
30 |
+
For more information, access to the dataset, and to contribute, please visit our [GitHub repository](https://github.com/yourrepo/biomed-vital).
|