[Radiomics in Practice and Its Basic Theory for Neurosurgeons].

Area of Science:

• Medical imaging analysis
• Quantitative imaging
• Machine learning in radiology

Background:

• Medical images are numerical data, suitable for machine learning and statistical analysis.
• Qualitative analysis of neurological images is increasingly insufficient; quantitative analysis is now standard.
• Radiomics, despite its name, has accessible foundational concepts and mathematical requirements.

Purpose of the Study:

• To provide a framework for integrating radiomics into analytical methodologies.
• To explain the accessible nature of radiomic feature calculations.
• To offer a deeper understanding of the image features underpinning radiomics.

Main Methods:

• Detailed procedural example of radiomic analysis.
• Overview of the historical development of radiomics.
• Exploration of gray level co-occurrence matrix and gray level run length matrix concepts.

Main Results:

• Radiomic feature calculations generally use high school-level mathematics.
• The foundational concepts of radiomics trace back to 1973.
• Provides a structured approach to understanding and applying radiomics.

Conclusions:

• Radiomics offers a quantitative approach to medical image analysis.
• The methodology is accessible to researchers with basic mathematical understanding.
• Understanding core concepts like gray level matrices is key to leveraging radiomics.