diff --git a/Images/result_rationale.jpg b/Images/result_rationale.jpg
new file mode 100644
index 0000000..6a40386
Binary files /dev/null and b/Images/result_rationale.jpg differ
diff --git a/Images/result_report.jpg b/Images/result_report.jpg
new file mode 100644
index 0000000..eff871e
Binary files /dev/null and b/Images/result_report.jpg differ
diff --git a/Images/result_report_generation.jpg b/Images/result_report_generation.jpg
new file mode 100644
index 0000000..ed23ee8
Binary files /dev/null and b/Images/result_report_generation.jpg differ
diff --git a/Images/result_vqa.jpg b/Images/result_vqa.jpg
new file mode 100644
index 0000000..faa6993
Binary files /dev/null and b/Images/result_vqa.jpg differ
diff --git a/Images/result_vqa.pdf b/Images/result_vqa.pdf
new file mode 100644
index 0000000..101ed87
Binary files /dev/null and b/Images/result_vqa.pdf differ
diff --git a/README.md b/README.md
index 4c53d61..3b34969 100644
--- a/README.md
+++ b/README.md
@@ -5,7 +5,7 @@ The official code for "Towards Generalist Foundation Model for Radiology"
[Website](https://chaoyi-wu.github.io/RadFM/)
-
+
## Quick Start:
Download [Model checkpoint](https://huggingface.co/chaoyi-wu/RadFM) and check `./src/test.py` for how to generate text with our model.
@@ -17,9 +17,12 @@ Our pre-train code is given in ```./src/train.py```.
## Case Study:
-
-
-
+
+
+
+
+
+
## To-do List:
- Polish the code for easier usage.
- Update an easy sample for a quick start.