Spaces:
Running
Running
Benjamin Consolvo
commited on
Commit
·
f9f7ead
1
Parent(s):
7d9f5c5
context for readme
Browse files
app.py
CHANGED
@@ -25,10 +25,14 @@ import streamlit.components.v1 as components # Add this import for custom HTML
|
|
25 |
st.set_page_config(page_title="Preventative Healthcare Outreach", layout="wide")
|
26 |
|
27 |
# Title at the top of the app
|
28 |
-
st.title("
|
29 |
st.markdown("""
|
30 |
-
Welcome to your preventative healthcare outreach agentic system, built using the open-source framework
|
31 |
-
|
|
|
|
|
|
|
|
|
32 |
2. Select default (synthetically generated) patient data, or upload your own CSV file.
|
33 |
3. Describe a medical screening task.
|
34 |
4. Click on "Generate Outreach Emails" to create draft emails to patients (.txt files with email drafts).
|
|
|
25 |
st.set_page_config(page_title="Preventative Healthcare Outreach", layout="wide")
|
26 |
|
27 |
# Title at the top of the app
|
28 |
+
st.title("Cloud Native Agentic Workflows in Healthcare")
|
29 |
st.markdown("""
|
30 |
+
Welcome to your preventative healthcare outreach agentic system, built using the open-source framework [Autogen](https://github.com/microsoft/autogen).
|
31 |
+
|
32 |
+
To improve patient health outcomes, healthcare providers are looking for ways to reach out to patients who may be eligible for preventative screenings. This system is designed to help you automate the process of identifying patients who meet specific screening criteria and generating personalized emails to encourage them to schedule their screenings.
|
33 |
+
|
34 |
+
The user provides a very broad screening criteria, and then the system uses AI agents to generate patient-specific criteria, filter patients from a given database, and ultimately write outreach emails to suggest to patients that they schedule a screening. To get the agents working, you can use the sidebar on the left of the UI to:
|
35 |
+
1. Customize the prompts for the agents. They use natural language understanding to execute on a workflow. You can use the default ones to get started, and modify to your more specific needs.
|
36 |
2. Select default (synthetically generated) patient data, or upload your own CSV file.
|
37 |
3. Describe a medical screening task.
|
38 |
4. Click on "Generate Outreach Emails" to create draft emails to patients (.txt files with email drafts).
|