With the goal of developing an advanced algorithm to detect pediatric weight errors, we developed a visual annotation tool (VAT) to support the rapid collection of expert-annotated errors. This abstract reports the preliminary evaluation results of VAT in terms of efficiency, accuracy, and usability. Future work involves completing the formal evaluation, use VAT to build a large annotation dataset for algorithm training, and develop a decision support tool to capture weight errors and improve patient safety.

Learning Objective: Evaluate the effectiveness of a visual annotation tool for weight-entry error detection in pediatric weight charts

Authors:

PJ Van Camp (Presenter)
University of Cincinnati

Lei Liu, University of Cincinnati
Cecilia Mahdi, Cincinnati Children’s Hospital Medical Center
Stephen Spooner, University of Cincinnati
Yizhao Ni, University of Cincinnati
Danny Wu (Presenter)
University of Cincinnati

Presentation Materials:

• Abstract/Manuscript - 1