Text Extraction and Standardization System Development for Pathological Records in the Korea Biobank Network

Area of Science:

• Biomedical Informatics
• Cancer Research
• Data Management

Background:

• The Korea BioBank Network (KBN) holds valuable pathological records for research in Korea.
• Efficient and accurate data extraction from these records is crucial for advancing biomedical studies.
• Existing methods may be time-consuming or prone to errors.

Purpose of the Study:

• To develop a time-efficient and error-reduced system for extracting data from KBN pathological records.
• To validate the system's performance using real-world cancer cohort data.
• To establish a scalable solution for multi-institutional data processing.

Main Methods:

• Development of a step-by-step data extraction protocol.
• System implementation and testing on lung cancer and breast cancer cohorts.
• Quantitative assessment of extraction accuracy and efficiency.

Main Results:

• The established system demonstrated significant improvements in time efficiency and error reduction.
• Accuracy of data extraction reached 91% across tested cohorts.
• The system proved effective for processing large-scale pathological records.

Conclusions:

• The developed system offers a robust solution for efficient pathological data extraction from the KBN.
• This methodology can enhance the usability of biobank data for cancer research.
• The system is adaptable for use with data from various institutions, including the KBN.