From: Automated extraction of Biomarker information from pathology reports
Rule | Example |
---|---|
1) Replace Roman numbers with the corresponding Arabic numbers | Factor VIII → Factor 8 |
2) Replace Greek symbol with the corresponding letters | CD79 α → CD79 alpha |
3) Remove laboratory related terms | CD10(repeat) → CD10 |
4) Generate new variants | GAL3 = GAL-3 = GAL 3 |
5) Convert to lowercase | DESMIN→ desmin |