Category Ranking
98%
Total Visits
921
Avg Visit Duration
2 minutes
Citations
20
Article Abstract
Objectives: Point-of-care ultrasonography (POCUS) for skin and soft tissue infections (SSTIs) has been integrated into routine clinical care in pediatric emergency medicine (PEM). Despite its widespread utilization, empirical data on skill development are required to inform standards of care. We sought to evaluate the accuracy of POCUS for the detection of SSTIs, and to estimate a learning curve as providers gained experience.
Methods: A database was created at a single urban pediatric emergency department (ED) for all POCUS studies performed for cellulitis and abscess among children from August 2009 to January 2020. Providers who completed at least 10 total studies were included. The accuracy of each study was asynchronously rated by 6 POCUS experts. We report each provider's learning curve. Within the provider, the studies were ordered temporally. Mixed-effect logistic regression was used to model study accuracy with study accuracy as the dependent variable, within-provider examination temporal order as a fixed effect, and random effects for provider-level intercept and longitudinal slope.
Results: A total of 3408 ultrasounds by 87 providers were included. The median number of ultrasounds per provider was 32 (range: 10 to 184). Accuracy increased significantly with each additional scan [odds ratio: 1.02 (1.01, 1.03)]. Significant variability was found across provider-level intercepts (P < 0.001) but not across slopes (P = 0.215). To reach mean accuracy levels of 90% and 95%, averages of 29 and 80 scans were required, respectively.
Conclusions: POCUS accuracy for detecting pediatric SSTIs improved with experience, with a mean of 29 scans required to reach 90% accuracy.
Download full-text PDF |
Source |
|---|---|
| http://dx.doi.org/10.1097/PEC.0000000000003467 | DOI Listing |
Publication Analysis
Top Keywords
Similar Publications
Two step approach for detecting and segmenting the second mesiobuccal canal of maxillary first molars on cone beam computed tomography (CBCT) images via artificial intelligence.
BMC Oral Health
September 2025
Oral and Maxillofacial Radiology Department, Cairo university, Cairo, Egypt.
Aim: The purpose of this study was to assess the accuracy of a customized deep learning model based on CNN and U-Net for detecting and segmenting the second mesiobuccal canal (MB2) of maxillary first molar teeth on cone beam computed tomography (CBCT) scans.
Methodology: CBCT scans of 37 patients were imported into 3D slicer software to crop and segment the canals of the mesiobuccal (MB) root of the maxillary first molar. The annotated data were divided into two groups: 80% for training and validation and 20% for testing.
Systematic review and meta-analysis of the learning curve and mentoring in robotic pancreatectomy: transitioning from novice to expert robotic surgeon.
J Robot Surg
September 2025
Department of General Surgery, Giglio Hospital Foundation, Cefalu', Italy.
The adoption of robotic pancreatectomy has grown significantly in recent years, driven by its potential advantages in precision, minimally invasive access, and improved patient recovery. However, mastering these complex procedures requires overcoming a substantial learning curve, and the role of structured mentoring in facilitating this transition remains underexplored. This systematic review and meta-analysis aimed to comprehensively evaluate the number of cases required to achieve surgical proficiency, assess the impact of mentoring on skill acquisition, and analyze how outcomes evolve throughout the learning process.
View Article and Find Full Text PDFExplainable deep learning framework for brain tumor detection: Integrating LIME, Grad-CAM, and SHAP for enhanced accuracy.
Med Eng Phys
October 2025
Biomedical Device Technology, Istanbul Aydın University, Istanbul, 34093, Istanbul, Turkey. Electronic address:
Deep learning approaches have improved disease diagnosis efficiency. However, AI-based decision systems lack sufficient transparency and interpretability. This study aims to enhance the explainability and training performance of deep learning models using explainable artificial intelligence (XAI) techniques for brain tumor detection.
View Article and Find Full Text PDFTwo-Step Semi-Automated Classification of Choroidal Metastases on MRI: Orbit Localization via Bounding Boxes Followed by Binary Classification via Evolutionary Strategies.
AJNR Am J Neuroradiol
September 2025
From the Department of Radiology, Memorial Sloan Kettering Cancer Center, New York, New York, United States of America (J.S.S., B.M., S.H., A.H., J.S.), and Department of Aerospace Engineering, Indian Institute of Technology Madras, Chennai, Tamil Nadu, India (H.S.).
Background And Purpose: The choroid of the eye is a rare site for metastatic tumor spread, and as small lesions on the periphery of brain MRI studies, these choroidal metastases are often missed. To improve their detection, we aimed to use artificial intelligence to distinguish between brain MRI scans containing normal orbits and choroidal metastases.
Materials And Methods: We present a novel hierarchical deep learning framework for sequential cropping and classification on brain MRI images to detect choroidal metastases.
Predicting left-without-being-seen in an emergency department as a dynamic risk.
Am J Emerg Med
September 2025
University of Toronto, Rotman School of Management, Canada.
Study Objective: Accurately predicting which Emergency Department (ED) patients are at high risk of leaving without being seen (LWBS) could enable targeted interventions aimed at reducing LWBS rates. Machine Learning (ML) models that dynamically update these risk predictions as patients experience more time waiting were developed and validated, in order to improve the prediction accuracy and correctly identify more patients who LWBS.
Methods: The study was deemed quality improvement by the institutional review board, and collected all patient visits to the ED of a large academic medical campus over 24 months.