OCR (Optical Character Recognition)

OCR (Optical Character Recognition) is technology that converts an image of text into machine-readable characters. In pharmacy automation, OCR is used to extract text from fax images, reading a prescription’s drug name, patient information, and dosage instructions as plain text.

OCR is the starting point for fax prescription automation, but it is not the full solution. OCR produces a text string; it does not interpret that string against clinical context. A prescription for “Lipitor 20mg” becomes the text “Lipitor 20mg”, OCR cannot tell you which DIN to select in Kroll, whether you carry a generic alternative, or how to map the sig line to Kroll’s format.

OCR-only pharmacy tools require manual DIN selection after the text has been extracted. Staff review the OCR output and look up the correct DIN themselves, reducing the physical typing but not the decision-making burden.

Multi-agent AI builds on OCR by adding a comprehension layer: reading the extracted text against your Kroll drug catalog, patient history, and other context before generating a write. This is how AutoRx achieves a DIN match rate above 95% on first attempt, not from better OCR alone, but from using the extracted text as input to a broader reasoning process.