Automated machine learning with R: AutoML tools for beginners in clinical research.

Area of Science:

• Computer Science
• Biostatistics
• Clinical Research

Background:

• Machine learning (ML) adoption is rapidly growing across diverse scientific fields.
• Developing ML pipelines is resource-intensive, demanding significant time and human effort.
• Automated ML (AutoML) tools streamline ML pipeline creation by automating key tasks.

Purpose of the Study:

• To introduce AutoML tools for general research applications, with a specific focus on clinical studies.
• To provide a straightforward, beginner-accessible methodology for implementing AutoML using the R programming language.
• To equip clinical researchers with practical code and results for direct application in binary classification tasks.

Main Methods:

• A review of AutoML tools and their application in research.
• Demonstration of a simplified AutoML approach using the R programming language.
• Inclusion of practical code examples and output for binary classification.

Main Results:

• AutoML significantly reduces the time and resources required for ML pipeline development.
• The R-based approach provides an accessible entry point for clinical researchers new to AutoML.
• Provided code and results facilitate the direct implementation of binary classification models.

Conclusions:

• AutoML offers a powerful solution to accelerate ML adoption in clinical research.
• Beginner-friendly tools and methods, like the R-based approach presented, are crucial for broader uptake.
• This work empowers clinical researchers to leverage advanced ML techniques more efficiently.