Funding outcome

Point-of-care ultrasound (POCUS) offers a noninvasive diagnostic tool for various physicians. Physicians can use POCUS to diagnose and treat musculoskeletal (MSK) injuries in children, specifically wrist and elbow injuries, with higher accuracy and speed. These injuries are common in children, and waiting for X-rays can unnecessarily increase wait times and radiation exposure. Physicians can use POCUS to diagnose MSK injuries accurately and efficiently, potentially reducing the need for radiographs like X-rays or CT scans. However, despite its advantages, using POCUS in clinical practice has been limited by clinicians not having the training to use POCUS nor being able to interpret ultrasound images. This limitation increases diagnostic variability and reduces POCUS's effectiveness as a diagnostic tool. This project aims to overcome these challenges by evaluating the effectiveness of artificial intelligence AI-aided interpretation of POCUS scans in pediatric patients with wrist and elbow injuries. I will conduct ultrasound scanning on pediatric patients at the Stollery, emergency department sites, and health clinics. AI algorithms will analyze the scans to interpret ultrasounds and identify diagnostic features of MSK injuries. AI-aided readings will be compared to clinician diagnoses to validate the accuracy of the AI tool. Clinicians will be presented with their initial diagnoses and AI-driven interpretations to evaluate how AI influences clinical decision-making. This project will determine whether AI-aided POCUS is a valid and feasible diagnostic tool and assess its impact on clinical workflows. The expected outcome is to demonstrate AI-aided POCUS as an accurate and accessible diagnostic modality. This project will leverage AI to enhance POCUS utility. This study aims to use a more efficient modality for diagnosing pediatric patients with wrist and elbow injuries in practice by addressing challenges in ultrasound interpretation and validating AI-aided POCUS.