William Niemiec
Back to research
Medical Prescription Data Extraction Using OCR and LLM for Import into the Takere Platform

Regional School of Machine Learning and Artificial Intelligence (ERAMIA)

Medical Prescription Data Extraction Using OCR and LLM for Import into the Takere Platform

Cauã Miranda, Pedro Jacinto, William Niemiec, Érika Cota

Published in
Regional School of Machine Learning and Artificial Intelligence (ERAMIA)
Year
2026
DOI
10.5753/eramiars.2026.35757

Abstract

This paper presents an automated solution for extracting data from medical prescription images and importing it into the Takere care plan platform. The objective is to make the data entry process more agile and reliable, replacing the manual procedure. The methodology employs Image Processing and Optical Character Recognition for text extraction, combined with Large Language Models to correct and semantically structure the information. Preliminary results indicate promising preliminary performance, achieving an F1-Score of 96.8% in the OCR extraction stage and 99.5% in component structuring. We conclude that this approach is potentially viable for mitigating errors and enhancing usability in healthcare management systems.

How to cite

Cauã Miranda, Pedro Jacinto, William Niemiec, Érika Cota. Medical Prescription Data Extraction Using OCR and LLM for Import into the Takere Platform. In: Regional School of Machine Learning and Artificial Intelligence (ERAMIA), 2026. DOI: https://doi.org/10.5753/eramiars.2026.35757

Socials

William Niemiec © All rights reserved