Effective data quality management for electronic medical record data using SMART DATA.

Area of Science:

• Medical Informatics
• Data Science
• Health Data Management

Background:

• Medical data management faces challenges like high costs, standardization issues, and complex preprocessing.
• A robust data quality management system is crucial for reliable data, risk mitigation, cost reduction, and increased data utilization.

Purpose of the Study:

• To introduce the concept of SMART data within a medical data quality management system.
• To conduct a case study using real-world colorectal cancer data to test the SMART data concept.

Main Methods:

• Defined a data quality management system based on the medical data lifecycle (Construction - Operation - Utilization).
• Proposed and applied the "SMART DATA" concept to colorectal cancer patient data from a single institution.

Main Results:

• Defined SMART DATA as systematized, high-quality data managed through its lifecycle via quality control.
• Curated 1,724 learning datasets and 27 Clinically Critical Set (CCS) datasets for colorectal cancer prediction.
• Identified unique patterns in CCS cases requiring further clinical review for improved prediction models.

Conclusions:

• Primary research was conducted to develop a medical data quality management system.
• The study aims to standardize medical data extraction and quality control, enhancing data utilization.
• This work contributes to developing a methodology and establishing a comprehensive medical data quality management system.