Files

Gitead 5cdd3452f2 fix: use Google Vision OCR instead of AI-vision estimates for Windows Type Agent clicks

Claude vision's self-reported text coordinates had a small but consistent bias,
causing the patient-row double-click to land above the name and the Exam-column
divider count to occasionally miscount and hard-fail. Swap patient-row, Exam,
and Save button targeting to exact OCR bounding boxes (Google Vision via the
existing PaymentOCRService) instead of AI-estimated ratios.

2026-07-30 23:14:02 -04:00

.turbo

initial commit

2026-04-04 22:13:55 -04:00

.env

initial commit

2026-04-04 22:13:55 -04:00

.env.example

initial commit

2026-04-04 22:13:55 -04:00

.gitignore

initial commit

2026-04-04 22:13:55 -04:00

complete_pipeline_adapter.py

initial commit

2026-04-04 22:13:55 -04:00

complete_pipeline.py

initial commit

2026-04-04 22:13:55 -04:00

main.py

fix: use Google Vision OCR instead of AI-vision estimates for Windows Type Agent clicks

2026-07-30 23:14:02 -04:00

package.json

fix: use venv for Python services to support Debian 13 managed environment

2026-04-18 20:50:47 -04:00

pdf_extractor.py

feat: payment PDF extraction, import, and remittance tracking

2026-05-07 12:53:50 -04:00

README.md

initial commit

2026-04-04 22:13:55 -04:00

requirements.txt

feat: payment PDF extraction, import, and remittance tracking

2026-05-07 12:53:50 -04:00

README.md

Medical Billing OCR API (FastAPI)

1) Prereqs

Google Cloud Vision service-account JSON.
GOOGLE_APPLICATION_CREDENTIALS env var pointing to that JSON.
Tesseract installed (for fallback OCR), and on PATH.

2) Install & run (local)

python -m venv .venv && source .venv/bin/activate
pip install -r requirements.txt
export GOOGLE_APPLICATION_CREDENTIALS=/absolute/path/to/service-account.json
uvicorn app.main:app --reload --port 8080