WOFAPS 2025 8th World Congress of Pediatric Surgery

Krasimira Kalinova ¹, Kaloyan Georgiev ¹, Jivko Jelyazkov ²

¹ Faculty of Medicine, Trakian University, Bulgaria
² Trakian University, Faculty of Innovation and trhnology,Bulgaria

Background

Artificial intelligence (AI) has been recently shown to improve clinical workflows, yet its potential in pediatric surgery remains largely unexplored. This systematic review details the use of in pediatric surgery.

Methods

Ten medical databases were searched from inception until January 2025, identifying articles focused on Artificial intelligence in pediatric surgery. The authors reviewed full texts of all eligible articles. Studies were included if they were original investigations or clinical application of models for pediatric health conditions managed surgically and possibilyty to teach the students and specializants . Studies were excluded if they were not peer-reviewed, commentaries, or case reports, did not focus on pediatric surgical conditions. Extracted data included study characteristics, place and time for teach for the clinical specialty, AI method and algorithm type, and performance metrics, key results, validation, and risk of bias using PROBAST and QUADAS-2 and others.

Results

Authors screened 6179 articles and included 202. Half of the studies (50%) reported predictive models (for adverse events [30%], surgical outcomes [26%] and survival [10%]), followed by diagnostic (30%) and decision support models (42%). The main pediatric surgical subspecialties represented across all models were general surgery (45%), pediatric urology (42%), thoracic and neurosurgery (36%). Fifty-three percent of models were interpretable, and 10% were both interpretable and externally validated. Concerns over applicability were identified in 10%.

Conclusions

While AI has wide potential clinical applications in pediatric surgery, very few published AI algorithms were externally validated, interpretable, and unbiased. Future research needs to focus on developing AI models which are prospectively validated and ultimately integrated into clinical workflows.