Machine Learning Algorithm Detection of Confluent B-Lines.

OBJECTIVE: B-lines are a ring-down artifact of lung ultrasound that arise with increased alveolar water in conditions such as pulmonary edema and infectious pneumonitis. Confluent B-line presence may signify a different level of pathology compared with single B-lines. Existing algorithms aimed at B-line counting do not distinguish between single and confluent B-lines. The objective of this study was to test a machine learning algorithm for confluent B-line identification. METHODS: This study used a subset of 416 clips from 157 subjects, previously acquired in a prospective study enrolling adults with shortness of breath at two academic medical centers, using a hand-held tablet and a 14-zone protocol. After exclusions, random sampling generated a total of 416 clips (146 curvilinear, 150 sector and 120 linear) for review. A group of five experts in point-of-care ultrasound blindly evaluated the clips for presence/absence of confluent B-lines. Ground truth was defined as majority agreement among the experts and used for comparison with the algorithm. RESULTS: Confluent B-lines were present in 206 of 416 clips (49.5%). Sensitivity and specificity of confluent B-line detection by algorithm compared with expert determination were 83% (95% confidence interval [CI]: 0.77-0.88) and 92% (95% CI: 0.88-0.96). Sensitivity and specificity did not statistically differ between transducers. Agreement between algorithm and expert for confluent B-lines measured by unweighted κ was 0.75 (95% CI: 0.69-0.81) for the overall set. CONCLUSION: The confluent B-line detection algorithm had high sensitivity and specificity for detection of confluent B-lines in lung ultrasound point-of-care clips, compared with expert determination.

A Quest for Better Strategy in Pediatric Intraosseous Placement Using Radiographic Measurements in Patients Younger Than 2 Years.

ABSTRACT: Intraosseous (IO) needles are used to obtain vascular access in pediatric patients during emergent situations. Recent literature has raised concern about high rates of IO malposition in younger children. Despite the widespread use of IO access in the pediatric population, there is scarce evidence regarding the ideal needle length or optimal access site. This study uses a radiographic approach to determine the appropriate IO needle length and access site to minimize the risk of malposition in children younger than 2 years. Radiographs of the lower extremities were obtained from the electronic database from a single tertiary care center. Using lateral views, anteroposterior measurements were obtained at 2 axial planes, located 1 cm superior to distal femur physis and 1 cm inferior to distal tibia physis. Based on the measurements, we calculated the probable needle tip positions if the needle was placed to the hub at the skin level using the EZ-IO (Teleflex Ltd, Wayne, PA) preset needle sizes. For subjects younger than 6 months, the 25-mm needle minimized malposition in the femur site with a 45.7% appropriate position rate, and the 15-mm needle minimized malposition in the tibia site with a 57.1% appropriate position rate. For the older age groups, we did not find a standard needle that would consistently minimize malposition in the femur site. For the tibia site, the 25-mm needle minimized malposition risk, with appropriate position rates of 81.0%, 87.5%, and 91.1% in the 6- to 12-month, 13- to 18-month, and 19- to 24-month groups, respectively.