Radiology societies urge caution when implementing AI in pediatric imaging
Radiology societies are advocating for caution when implementing artificial intelligence in pediatric care.
Six professional organizations including the Society for Pediatric Radiology made their case in a joint statement published across multiple medical journals Tuesday. They noted that AI has the potential to revolutionize the specialty. However, current solutions are “predominantly focused on adult populations, often overlooking the specific requirements of children.”
“This is important because children differ significantly from adults in terms of physiology, developmental stages and clinical needs, necessitating tailored approaches for the safe and effective integration of AI tools,” SPR and others including lead author Susan C. Shelmerdine, MBBS, PhD, a radiologist with Great Ormond Street Hospital, London, wrote Nov. 25 in JACR and other journals.
The multisociety position statement focuses on four pillars of AI adoption including (1) regulation and purchasing, (2) implementation and integration, (3) interpretation and post-market surveillance and (4) education. Others involved in producing the paper include the pediatric subcommittee of the American College of Radiology, the European Society of Pediatric Radiology, the Asian Oceanic Society of Pediatric Radiology, the Sociedad Latinoamericana de Radiología Pediátrica, and the Society of Pediatric Neuroimaging.
Experts are proposing pediatric-specific safety ratings, inclusion of datasets from diverse pediatric populations, “quantifiable transparency metrics,” and explainability in AI models. The latter seeks to help “mitigate biases and ensure AI systems are appropriate for use in children,” Shelmerdine and colleagues noted.
“Risk assessment, dataset diversity, transparency, and cybersecurity are important steps in regulation and purchasing,” they wrote. “For successful implementation, a phased strategy is recommended, involving early pilot testing, stakeholder engagement, and comprehensive post-market surveillance with continuous monitoring of defined performance benchmarks.”
The Society for Pediatric Radiology et al. also are urging for “clear protocols” when managing discrepancies, along with adverse incident reporting to help “maintain trust and safety.” They’re additionally emphasizing the need for “foundational AI literacy courses” for all health professionals, focusing on pediatric safety considerations, along with other specialized training for those directly involved in imaging.
“Public and patient engagement is crucial to foster understanding and acceptance of AI in pediatric radiology,” the authors added. “Ultimately, we advocate for a child-centered framework for AI integration, ensuring that the distinct needs of children are prioritized and that their safety, accuracy, and overall well-being are safeguarded.”
Radiologists are calling for collaboration between pediatric rads, AI vendors, and healthcare institutions to help reach their goals.
“AI in pediatric radiology holds immense potential to improve diagnostic accuracy, efficiency, and personalized care but requires cautious implementation,” the authors concluded. “Transparency, ethical practices, and education are essential to ensure safe and effective use, minimizing bias and harm,” they added later.
Read much more in the Journal of the American College of Radiology.
Marty Stempniak has covered healthcare since 2012, with his byline appearing in the American Hospital Association's member magazine, Modern Healthcare and McKnight's. Prior to that, he wrote about village government and local business for his hometown newspaper in Oak Park, Illinois. He won a Peter Lisagor and Gold EXCEL awards in 2017 for his coverage of the opioid epidemic.