A Promising Approach: Artificial Intelligence Applied to Small Intestinal Bacterial Overgrowth (SIBO) Diagnosis Using...

Area of Science:

• Gastroenterology
• Medical Diagnostics
• Artificial Intelligence in Medicine

Background:

• Small intestinal bacterial overgrowth (SIBO) diagnosis relies heavily on breath testing, but lacks standardized criteria.
• The absence of a universal cut-off for breath tests complicates SIBO patient identification and gold standard validation.

Purpose of the Study:

• To apply artificial intelligence (AI) for analyzing breath test data to improve SIBO diagnosis.
• To explore AI-driven methods for establishing adaptive diagnostic criteria for SIBO.

Main Methods:

• Collected breath-testing data from 1071 patients.
• Utilized AI clustering algorithms (K-means, DBSCAN) after confirming clustering tendency with Hopkins Statistic.
• Evaluated clustering effectiveness using Silhouette scores and described identified patterns.

Main Results:

• AI clustering successfully identified distinct patterns within the breath-testing dataset.
• The study demonstrated the feasibility of using AI for SIBO diagnostic criteria development.
• AI offers advantages in high-dimensional analysis and data-driven insights for SIBO diagnosis.

Conclusions:

• AI presents a promising tool for enhancing SIBO diagnosis through breath testing.
• This approach can benefit both clinical practice and scientific research in SIBO.
• AI facilitates the development of more precise and data-driven diagnostic standards for SIBO.