Artificial Intelligence in Pediatric Surgery: From Diagnostics and Preoperative Planning to Risk Stratification: A...

Area of Science:

• Pediatric Surgery
• Medical Artificial Intelligence
• Surgical Innovation

Background:

• Artificial intelligence (AI) applications in pediatric surgical care are growing but lag behind adult surgery.
• Existing reviews lack a comprehensive synthesis across key pediatric surgical domains.

Purpose of the Study:

• To provide a comprehensive review of AI in pediatric surgery across diagnostics, preoperative planning, risk stratification, and error prevention.
• To highlight recent advancements and identify unmet challenges in the clinical translation of AI.

Main Methods:

• A narrative review of PubMed/MEDLINE (2020-2025) for peer-reviewed studies on AI in pediatric surgery.
• Studies were assessed based on methodology, clinical applicability, and relevance to pediatric surgical patients across four defined domains.

Main Results:

• Diagnostic imaging shows high accuracy (up to 95%) with deep learning, but external validation is limited.
• AI aids preoperative planning, altering strategy in 8% of oncology cases, yet outcome benefits need more evidence.
• Risk models show potential, but few have external validation; error prevention tools are in early stages.
• Most studies are retrospective, single-center, and heterogeneous.

Conclusions:

• AI holds significant potential to enhance pediatric surgical diagnostics, planning, and safety.
• Clinical translation requires robust multicenter pediatric datasets, prospective validation, and interpretable AI models.
• Addressing identified gaps is essential for the safe and effective adoption of AI in pediatric surgery.