Continuous Inferior Vena Cava Diameter Tracking through an Iterative Kanade-Lucas-Tomasi-Based Algorithm

Area of Science:

• Critical Care Medicine
• Medical Imaging
• Ultrasound Technology

Background:

• Inferior vena cava (IVC) ultrasound is vital for assessing fluid status in critically ill patients.
• Respiratory-induced changes in IVC size reflect volemic status.
• Manual IVC measurement can be time-consuming and subject to inter-observer variability.

Purpose of the Study:

• To develop and validate an automated ultrasound technique for continuous tracking of the IVC diameter.
• To assess the accuracy and precision of this automated method against expert measurements.
• To explore the potential for automated critical care ultrasound applications.

Main Methods:

• Development of an automated algorithm using Kanade-Lucas-Tomasi feature tracking and pyramidal segmentation.
• Continuous tracking of IVC diameter during ultrasound examinations.
• Comparison of automated measurements with those from trained ultrasound physicians in 47 spontaneously breathing patients.

Main Results:

• High agreement between automated and physician measurements was observed (intra-class correlation coefficients: 0.897 for max diameter, 0.967 for min diameter, 0.975 for caval index).
• Over 95% of measurements agreed within 10% of the IVC collapse.
• Observed and reported variability in the cardiac collapsibility index.

Conclusions:

• The developed automated ultrasound technique accurately and precisely tracks IVC diameter.
• This algorithmic approach offers a reliable foundation for future automated critical care ultrasound measures.
• Automated IVC assessment can enhance the evaluation and management of critically ill patients.