The big data effort in radiation oncology: Data mining or data farming?

Area of Science:

• Oncology
• Health Informatics
• Data Science

Background:

• Electronic health records and radiation oncology systems generate vast amounts of daily data.
• The extraction and utilization of this big data for clinical improvement has been slow.
• Current data strategies may not be optimal for realizing the potential of this information.

Purpose of the Study:

• To propose a conceptual framework for effectively utilizing radiation oncology big data.
• To identify key elements, process changes, and technological solutions needed for data extraction and use.
• To outline the role of professional societies in advancing data-driven strategies.

Main Methods:

• Conceptualization of data farming as an alternative to data mining for radiation oncology data.
• Analysis of critical data elements, clinical processes, and technological challenges.
• Consideration of standardization and the role of professional organizations.

Main Results:

• A data farming model provides a clearer vision for extracting and using big data.
• Identification of essential data elements and necessary clinical process adjustments.
• Highlights technology, process, and standardization as crucial factors for success.

Conclusions:

• Adopting a data farming approach can accelerate the use of big data in radiation oncology.
• Strategic focus on technology, process, and standardization is essential for effective data utilization.
• Prioritizing efforts based on these factors will enhance the impact of data in clinical practice.