Application of Machine Learning Based on Structured Medical Data in Gastroenterology

Area of Science:

• Medical Informatics
• Artificial Intelligence in Medicine
• Clinical Data Analysis

Background:

• The increasing volume of clinical data necessitates advanced analytical tools like artificial intelligence (AI).
• Machine learning (ML), a subset of AI, is vital for processing large datasets, with deep learning excelling in unstructured data analysis.
• Traditional ML models offer significant potential for structured clinical data, improving healthcare efficiency.

Purpose of the Study:

• To provide a comprehensive overview of the current adoption status of ML in gastroenterology.
• To discuss the future potential and directions for ML applications within gastroenterology.
• To briefly summarize recent advancements in large language models (LLMs) relevant to the field.

Main Methods:

• This study employed a narrative review methodology.
• Literature search focused on the application of ML in gastroenterology, contrasting it with deep learning and statistical models.
• Recent developments in LLMs were also synthesized.

Main Results:

• ML adoption in gastroenterology lags behind traditional statistical models and deep learning approaches.
• Despite limited current use, ML holds substantial promise for enhancing diagnostic and prognostic predictions in gastroenterology.
• Recent advances in LLMs present new opportunities for clinical data analysis.

Conclusions:

• There is a need to increase the adoption and exploration of ML techniques in gastroenterology.
• Future research should focus on leveraging ML, including LLMs, for structured and unstructured data in gastroenterology.
• Integrating ML can significantly improve healthcare efficiency and patient outcomes in gastroenterology.