Modeling Factors Associated With Diarrhea Caused by Cryptosporidium Species Using Machine Learning Methods.

Area of Science:

• Environmental Health
• Infectious Diseases
• Machine Learning in Healthcare

Background:

• Cryptosporidium spp. is a significant cause of severe diarrhea in children.
• Understanding transmission routes and risk factors is vital for public health interventions.

Purpose of the Study:

• To categorize Cryptosporidium spp. infection in children using machine learning.
• To identify key risk factors associated with Cryptosporidium infection.

Main Methods:

• Employed random forest and bagging CART algorithms for classification.
• Evaluated model performance using metrics like accuracy, sensitivity, and F1-score.
• Utilized 5-fold cross-validation and variable importance analysis.

Main Results:

• The bagged CART model demonstrated superior classification performance.
• High household density (≥5 people) and specific water sources (tap, pipeline) were identified as significant risk factors.
• Waterborne transmission was indicated as the primary route of infection.

Conclusions:

• Waterborne transmission is the predominant route for Cryptosporidium spp. infections.
• Interventions should focus on enhancing water quality and disinfection, especially in crowded households with poor sanitation.