Abstract: TH-PO015
Advancements in Predicting Pediatric AKI Using Artificial Intelligence: A Systematic Review
Session Information
- Augmented Intelligence for Prediction and Image Analysis
October 24, 2024 | Location: Exhibit Hall, Convention Center
Abstract Time: 10:00 AM - 12:00 PM
Category: Augmented Intelligence, Digital Health, and Data Science
- 300 Augmented Intelligence, Digital Health, and Data Science
Author
- Nada, Arwa, Loma Linda University, Loma Linda, California, United States
Background
The prediction and management of Acute Kidney Injury (AKI) in pediatric patients remain challenging due to its rapid onset and diverse etiologies. Recent advancements in artificial intelligence (AI) offer promising tools for early detection and improved clinical outcomes.
Methods
Search Strategy: Utilized PubMed with keywords and MeSH terms related to pediatric AKI and artificial intelligence.
Inclusion Criteria: Studies published from 2019-2023, involving AI models for pediatric AKI prediction or management.
Data Extraction: Study characteristics, AI methods, population, key findings, and references were extracted and tabulated.
Results
Table I shows summary of the studies (image of table uploaded)
The studies colectively showed: high Accuracy: GBM, ensemble learning, and RF models demonstrated high predictive accuracy.
Early Detection: AI models enabled early AKI detection, facilitating timely interventions.
Risk Stratification: Effective in identifying high-risk patients for personalized treatment.
Data Integration: AI successfully integrated diverse data sources for comprehensive analysis.
Conclusion
AI integration in clinical practice holds the potential for revolutionizing pediatric AKI management and improving patient outcomes through proactive and personalized care. Further research should focus on standardizing AI applications and ensuring equitable healthcare delivery.