gemeo-twin-stack / src /gemeo /datasus /__init__.py
timmers's picture
GEMEO world-model — initial release (module + NeuralSurv ckpt + RareBench v49 + KG embeddings)
089d665 verified
raw
history blame contribute delete
474 Bytes
"""DATASUS data pipeline — pulls real Brazilian public-health records.
Sources:
- SIM (Sistema de Informação sobre Mortalidade): death records with CID-10
- SIH-RD (Sistema de Informações Hospitalares): hospital admissions
- SIA-APAC: high-cost drug dispensations
- SINASC: birth records (congenital anomalies)
All pulls are read-only, FTP-based, with DBC→DBF conversion via pyreaddbc.
"""
from .sim_pull import pull_sim, parse_sim_record, RARE_CIDS_CID10