Automatic Text Extraction Using Optical Character Recognition (OCR) for Blood Test Report Management
Loading...
Date
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Universiti Malaysia Sarawak (UNIMAS)
Abstract
Description
The Automatic Medical Data Extraction System is to reduce the process of medical data entry by automating the extraction of data from physical blood test reports. This system utilizes Optical Character Recognition (OCR) and Natural Language Processing (NLP) techniques to efficiently extract, structure, and standardize medical data. Currently, medical staff manually input data from blood test reports into electronic systems, a process prone to human error and inefficiency. This project will eliminate the manual data entry process by developing a system to extract blood test report data from various laboratories. The system will extract relevant data, convert test units to a standardized format, and ensure the accuracy and consistency of the information stored in medical record systems. By automating the extraction and data structuring process, the system will significantly reduce human error, improve the efficiency of medical record management, and enhance the overall quality of patient care. The prototype’s performance will be evaluated through metrics such as Character Error Rate (CER), Word Error Rate (WER), and Named Entity Recognition (NER), ensuring the reliability of the extracted data.