Wrist trauma is common in children and generally requires radiography for exclusion of fractures, subjecting children to radiation and long wait times in the emergency department. Ultrasound (US) has potential to be a safer, faster diagnostic tool. This study aimed to determine how reliably US could detect distal radius fractures in children, to contrast the accuracy of 2DUS to 3DUS, and to assess the utility of artificial intelligence for image interpretation. 127 children were scanned with 2DUS and 3DUS on the affected wrist. US scans were then read by 7 blinded human readers and an AI model. With radiographs used as the gold standard, expert human readers obtained a mean sensitivity of 0.97 and 0.98 for 2DUS and 3DUS respectively. The AI model sensitivity was 0.91 and 1.00 for 2DUS and 3DUS respectively. Study data suggests that 2DUS is comparable to 3DUS and AI diagnosis is comparable to human experts.

Supracondylar fractures are common injuries in children. Diagnosis typically relies on radiography, which can involve long wait times in the ED, emits ionizing radiation, and can miss non-displaced fractures. Ultrasound (US) has the potential to be a safer, more convenient diagnostic tool, especially with new highly portable handheld 2D point of care US (POCUS). This study aimed to determine the reliability of 2D POCUS for the detection of supracondylar fractures and elbow joint effusions, to contrast the accuracy of 2D POCUS vs. 3DUS vs. radiographs, and to determine whether blinded image interpretation could produce similar results to non-blinded real-time imaging. Fifty-seven children were scanned with 2D POCUS and 3DUS on the affected elbow. US scans were then read by three blinded readers, and the results were compared to gold-standard radiographs. Compared to a gold standard of 30-day radiographic diagnosis, readers of 2D POCUS detected supracondylar fracture and effusion with sensitivities of 0.91 and 0.97, respectively, which were both higher than with 3DUS. Inter-rater reliability of fracture detection was moderate for 2D POCUS (k = 0.40) and 3DUS (k = 0.53). Consensus sensitivities, although high, were lower than reports from some non-blinded studies, indicating that clinical presentation serves as an important factor in detection rates. Our results from consensus US diagnosis support the validity of using 2D POCUS in children for supracondylar fracture and elbow effusion diagnosis.

Wrist trauma is common in children and generally requires radiography for exclusion of fractures, subjecting children to radiation and long wait times in the emergency department. Ultrasound (US) has potential to be a safer, faster diagnostic tool. This study aimed to determine how reliably US could detect distal radius fractures in children, to contrast the accuracy of 2DUS to 3DUS, and to assess the utility of artificial intelligence for image interpretation. 127 children were scanned with 2DUS and 3DUS on the affected wrist. US scans were then read by 7 blinded human readers and an AI model. With radiographs used as the gold standard, expert human readers obtained a mean sensitivity of 0.97 and 0.98 for 2DUS and 3DUS respectively. The AI model sensitivity was 0.91 and 1.00 for 2DUS and 3DUS respectively. Study data suggests that 2DUS is comparable to 3DUS and AI diagnosis is comparable to human experts.

Elbow and wrist fractures are the most common fractures in pediatric populations. Automatic segmentation of musculoskeletal structures in ultrasound (US) can improve diagnostic accuracy and treatment planning. Fractures appear as cortical defects but require expert interpretation. Artificial Intelligence (AI) can provide real-time feedback and highlight key structures, helping lightly trained users perform exams more confidently. However, pixel-wise human expert annotations required for AI model training remain time-consuming and costly. To address this challenge, we propose FlexICL, a novel and flexible in-context learning (ICL) framework designed for segmenting bony regions in US images. We applied it to a label-sparse setting, where only a very small subset of training frames contained human annotations, and the model segmented unseen test frames from different videos. We systematically investigated various image concatenation techniques and training strategies for visual ICL and introduced novel concatenation methods that significantly enhance model performance with limited labeled data. By integrating multiple context focus enhancement strategies, FlexICL achieves robust segmentation performance across four wrist and elbow US datasets while requiring only 5% of the training images. It outperforms state-of-the-art visual ICL models like Painter, MAE-VQGAN, and its backbone ViT-CNN by up to 52% Dice coefficient on 210 US sweeps. These initial results highlight the potential of FlexICL as an efficient and scalable solution for US image segmentation well-suited for medical imaging use cases where labeled data is scarce.