Optimizing Ultrasound Analysis for Diverse Patient Anatomy Identification

Central venous catheterization (CVC) poses inherent risks, but these can be significantly reduced by using ultrasound guidance during needle insertion. This study aims to enhance the understanding of patient anatomy as visualized during CVC by employing a semi-automated image analysis method to track the internal jugular vein and carotid artery in recorded ultrasound videos. A total of 50 CVC procedures were conducted at Penn State Health Milton S. Hershey Medical Center, with the ultrasound visualizations recorded for analysis. An algorithm was developed to detect the vessel edges, enabling the calculation of various metrics, including area, position, and eccentricity. The findings revealed typical anatomical variations between the vein and artery, notably with the carotid artery appearing more circular and situated posterior to the vein in most cases. Two instances presented atypical positions of the artery, highlighting the algorithm’s precision in identifying anatomical anomalies. Furthermore, dynamic properties of the vessels were analyzed, showing that the vein compressed to an average of 13.4% of its original size, while the artery expanded by 13.2%. This study offers valuable insights that could enhance the accuracy of training simulations, thereby improving medical education and procedural proficiency. Additionally, the innovative use of automated data analysis techniques in clinical recordings demonstrates the potential for ongoing assessments of patient anatomy, paving the way for future advancements in the field.