553 The role of artificial intelligence in pediatric injuries-a scoping review. (20th November 2022)

• Record Type: Journal Article
• Authors: Naseem, Maleeha; Arshad, Hajra; Ashraf, Sarah
• Is Part Of: Injury prevention. Volume 28(2022)Supplement 2
• Page Start: A85
• Page End: A85
• Publication Date: 2022-11-20
• Subjects: Children's accidents -- Prevention -- Periodicals; Accidents -- Prevention -- Periodicals; 617.1
• DOI: 10.1136/injuryprev-2022-safety2022.252
• Languages: English
• ISSNs: 1353-8047

Abstract

Abstract : Background: In this modern era, clinicians are required to make quick patient diagnosis and therapeutic decisions to gain desirable patient outcomes. AI systems like Artificial neural networks (ANNs), fuzzy expert systems, evolutionary computation and hybrid intelligent systems assists health care workers by productive manipulation of data and knowledge. The role of AI in pediatric injury research cannot be undermined.AI based tools have been utilized in clinical injury prediction as well as prevention in pediatric trauma patients Aims: To explore the utility and application of various AI techniques and tools in pediatric trauma. Methods: The Literature search for this scoping review was performed on PubMed. Studies related to Artificial Intelligence (AI) or its subset (Machine learning, Deep Neural Networks) in pediatric trauma which were in English language were included, whereas studies related to robotics, adult trauma and burn injury were excluded. After removing duplicates, a total of 102 articles were identified. After full text review, 26 articles were included in the review based on our inclusion/exclusion criteria. The literature search and reporting of this Scoping review was conducted according to PRISMA-ScR guidelines. Results: The results were thematically categorized according the category of AI, type of pediatric injuries and modality of injury diagnosis(X rays, CT, Ultrasound or simply numerical data).We identified major AI applications in predictions of 1. Head Injuries 2.Peripheries including fractures and dislocations 3.Outcomes of Injuries and Morbidity/Mortality. The utility of AI tools were mostly in identification and prediction of Head injuries and Head injury outcomes followed by upper extremity injuries including fractures. The major AI techniques were conventional machine learning techniques, Image recognition using CNNs, artificial neural networks, decision support systems and very few reported the utility of NLP(Natural Language processing ). Conclusion: AI has a variety of application in pediatric trauma prediction, mostly in head injury prediction and patient outcomes after head injuries. The scope of pediatric injury research using AI can be expanded to utilizing other domains of AI including speech recognition, NLP and robotics that are still unexplored and underutilized.