OpenCost Health
Tool for patients that uses AI to extract billing codes from after visit summaries, provide cost estimate breakdowns, and search for cheapest facilities nearby to perform procedures.
Core trick
Extracts billing codes from after-visit summaries with AI, then maps them to cost estimates and ranks nearby facilities by price.
Description
Tool for patients that uses AI to extract billing codes from after visit summaries, provide cost estimate breakdowns, and search for cheapest facilities nearby to perform procedures.
README excerpt
# abridge-hackathon
## Python environment
Set up a local virtual environment (`.venv`, git-ignored) and install the pinned dependencies:
```bash
# from the repo root (abridge-hackathon/)
python3 -m venv .venv
source .venv/bin/activate # Windows: .venv\Scripts\activate
python -m pip install --upgrade pip
python -m pip install -r requirements.txt
```
Add your Anthropic key to a `.env` file at the repo root (also git-ignored):
```
ANTHROPIC_API_KEY=sk-ant-...
```
### Running the notebooks
The `data_processing/*.ipynb` notebooks expect this environment. Either:
- **In VS Code / Cursor:** open a notebook and pick the `.venv` interpreter as the kernel (top-right kernel picker → _Python Environments_ → `.venv`), or
- **From the CLI:**
```bash
source .venv/bin/activate
jupyter nbconvert --to notebook --execute --inplace \
data_processing/04_avs_line_to_billable_codes.ipynb
```
Frontend — in another terminal:
```bash
cd frontend
npm install
npm run dev
```
Then open http://localhost:5173. The Vite dev server proxies `/api` to the
backend on port 8000, so frontend code can use relative paths like
`fetch('/api/health')` without hardcoding a host.
## The patient flow
Three screens, in order:
1. **Visit summary** — two panels side by side. On the left, the after-visit
summary the care team uploaded (pick any of the 25 encounters from the
dropdown; ones with no priced follow-ups are marked). On the right, the
procedures _recommended as follow-up_ are extr