Data Cleaning Rules

The following transformations are applied to uploaded Excel files to ensure standardized, clean, and anonymized output:

General Cleaning

• Remove unnamed columns and empty rows.
• Remove duplicate rows after transformation.
• Normalize header names to ensure a consistent schema.

Sensitive Data Sanitization

• Replace the NHI column with an anonymized ID (persisted across sheets in the same upload).
• Convert DOB to a calculated Age, then drop DOB.
• Drop columns named Address or Contact.

Transformation Logic

• Multi-sheet alignment: the same ID maps across sheets for the same patient.
• Bias-detection rules for demographics (optional future extension).
• File-specific rules applied for:
  • Case-Mix
  • Holistic
  • Fare-up
  • Outpatient