Quantitative Morphometry and Machine Learning Model to Explore Duodenal and Rectal Mucosal Tissue of Children with...

Area of Science:

• Gastroenterology
• Pediatric Pathology
• Immunology

Background:

• Environmental enteric dysfunction (EED) is a subclinical enteropathy linked to malnutrition and impaired development in resource-limited settings.
• Chronic exposure to environmental enteropathogens is hypothesized to cause EED, leading to growth failure and vaccine failure.
• Understanding EED's pathology is crucial for developing effective interventions.

Purpose of the Study:

• To investigate the duodenal and colonic histopathology of children with EED, celiac disease, and other enteropathies.
• To compare EED and celiac disease tissue characteristics using quantitative morphometry and machine learning.
• To explore potential biomarkers and anatomical regions for EED diagnosis and management.

Main Methods:

• Quantitative mucosal morphometry and histopathologic scoring (Marsh classification) were applied to duodenal and rectal tissues.
• Archival and prospective cohorts from Pakistan and the United States were analyzed.
• Machine learning-based image analysis was employed to assess tissue characteristics.

Main Results:

• Villus blunting was more severe in celiac disease than EED, with shorter villi observed in Pakistani celiac patients.
• Goblet cell depletion and increased intraepithelial lymphocytes were noted in both EED and celiac disease.
• Rectal crypts in EED showed increased inflammatory cells, and rectal neutrophils correlated with duodenal EED severity.

Conclusions:

• EED presents a spectrum of inflammation involving both the duodenum and rectal mucosa.
• Histopathological findings in EED overlap with other enteropathies, necessitating advanced analysis like machine learning.
• Examining both duodenal and rectal tissues is recommended for a comprehensive understanding and management of EED.