MENU

Click here to

×

Are you sure ?

Yes, do it No, cancel

Implementation of a Patient Motion Monitoring System Using Kinect V2 Sensor

D Lee1*, M Chetvertkov1, S Oh1, J Sohn1,2, (1) Allegheny Health Network, Pittsburgh, PA, (2) Drexel University College of Medicine, Philadelphia, PA

Presentations

(Sunday, 7/12/2020)   [Eastern Time (GMT-4)]

Room: AAPM ePoster Library

Purpose: Patient motion often compromises dosimetric accuracy. Different motion monitoring systems are used in clinics to account for that, but they are mostly expensive and most monitoring systems mounted provide less (or zero) mobility to share with other imaging and treatment systems. In this study, we proposed a motion monitoring technique with Kinect V2 sensor to (1) improve installation flexibility and (2) maintain monitoring system consistency in radiotherapy procedures.
Methods: A Kinect sensor was utilized to continuously obtain a marker (40mm × 50mm) position located at 600mm ~ 700mm distance. The angle of marker on a hinge frame is adjusted for directly facing the Kinect sensor and the marker is placed at where to monitor motion. Motion monitoring is comprised of three steps: (1) Determine a reference marker position and its minimum and maximum ranges by analyzing marker positions acquired prior to real-time monitoring, (2) monitor a target motion in real-time with/without the intervention of imaging and treatment using given warning messages (i.e., ±3mm threshold of ranges) and (3) write a log of measured and analyzed data for further uses. A patient motion monitoring system using Kinect sensor was implemented on C# and .Net framework platform and evaluated whilst monitoring simulated marker motions. Root mean square (RMS) of centroid marker motions was measured in the presence of simulated 10mm, 20mm, 30mm, and 40mm displacements.
Results: Centroid marker motions in RMS were measured at 10.4mm, 19.7mm, 30.6mm and 39.2mm across the four displacements and their differences between simulated and measured centroid motions were 0.4mm, -0.3mm, 0.6mm and 0.8mm.
Conclusions: This study demonstrated sub-millimeter accuracy in real-time motion monitoring and this will be potential for motion monitoring of cancer patients during medical imaging and radiotherapy. In addition, visual guidance with motion traces is highly considered as a function of improving dosimetry accuracy.

Download ePoster [PDF]

Keywords

Not Applicable / None Entered.

Taxonomy

Not Applicable / None Entered.

Contact Email