# Role of artificial intelligence in determining nutritional risk factors among post-periodontal surgical patients. A scoping review

## TL;DR

This scoping review explores how artificial intelligence can help manage nutrition and identify risk factors for patients recovering from periodontal surgery.

## Contribution

The paper provides a comprehensive overview of AI applications in post-periodontal nutritional care and identifies key risk factors using recent literature.

## Key findings

- AI models predict healing and nutritional risks by analyzing factors like age, diabetes, and dietary compliance.
- Poor pre-surgery nutrition and chewing impairment are linked to lower nutrient intake and delayed healing.
- Baseline pocket depth and antibiotic use are strong predictors of post-surgical outcomes.

## Abstract

This scoping review examines recent peer-reviewed literature (2019–2025) on the role of artificial intelligence (AI) in managing nutrition care for post-periodontal surgical patients, and identifies key risk factors influencing nutritional outcomes after periodontal surgery. AI modalities considered include machine learning, expert systems, clinical decision support, and predictive analytics.

A systematic search of databases (e.g., PubMed, Scopus) identified studies on AI applications in periodontology, nutrition, or wound healing. The inclusion criteria were English-language, peer-reviewed publications from 2019 onwards that focused on AI in periodontal care or nutritional management, and studies addressing risk factors (such as age, comorbidities, dietary compliance, oral function, socioeconomic status, etc.) that affect post-surgical nutrition or healing. Data were charted on study characteristics, AI type, nutritional outcomes, and reported risk factors. 28 publications were included (10 original studies, eight reviews, five clinical reports, five conceptual papers). AI has been used in periodontal care for diagnostics, prognostics, and decision support.

Machine learning models can predict healing and nutritional risks by analyzing patient data, with key risk factors including age, comorbidities such as diabetes, poor nutrition, low dietary compliance, oral function, and socioeconomic status. Older, chewing-impaired patients have lower nutrient intake and a higher risk of malnutrition. Poor pre-surgery nutrition delays healing. AI models forecast outcomes, identifying baseline pocket depth and antibiotic use as strong predictors. Emerging AI tools in periodontology can enhance nutrition management through early risk detection and personalized diets.

Factors like age, health, oral function, and socioeconomic status affect recovery. Using AI risk assessments with nutritional plans may improve healing. More research is needed to realize AI's full potential. While direct studies are limited, emerging evidence indicates strong potential for personalized, AI-supported nutritional care.