# Kefir and healthy aging: revealing thematic gaps through AI-assisted screening and semantic evidence mapping

## TL;DR

This study uses AI and semantic tools to map research on kefir's anti-aging effects, revealing gaps in clinical aging outcomes.

## Contribution

A novel AI-assisted and semantic clustering method to identify thematic gaps in kefir and aging research.

## Key findings

- Most kefir studies focus on inflammation and oxidative stress, not neurocognitive or metabolic outcomes.
- Research on kefir and aging is thematically imbalanced, showing structural bias in current priorities.
- The method provides a reproducible strategy to uncover blind spots in aging-related research.

## Abstract

Kefir, fermented milk rich in probiotics, has attracted growing attention for its potential anti-aging effects. Yet, studies specifically addressing kefir in the context of aging remain limited and scattered across diverse biological fields. To overcome this fragmentation, we applied an integrative approach that combines a cutting-edge AI-assisted algorithm for evidence screening with a Python-based semantic clustering pipeline. This allowed us to systematically map and classify the existing literature into four functional domains of aging: changes in body composition, energy balance, homeostatic signaling networks, and neurodegeneration. The resulting evidence map revealed a marked thematic imbalance, with most studies concentrated in mechanistic pathways such as inflammation and oxidative stress, and far fewer addressing neurocognitive or metabolic outcomes. This asymmetry suggests a structural bias in current research priorities and highlights the need to expand kefir-related studies toward more clinically relevant aging endpoints. By merging AI with domain-specific linguistic tools, our study provides a reproducible and data-driven strategy to uncover thematic blind spots and guide future investigations into kefir’s anti-aging potential.