Yet another innovative use for Microsoft’s Kinect sensor: researchers at the Hospital of the University of Pennsylvania are using Kinect to estimate whole-body volume in order to provide more accurate CT dose estimations. Several methods exist that estimate whole-body effective radiation dose, however many of those do not incorporate patient size, which can result in inaccurate estimations in small patients such as children and in large adults. Also, current methods usually do not incorporate the size of the region of anatomy being imaged.
The researchers are using the Kinect’s camera and infrared depth sensors in combination with software developed by Radimetrics to estimate patient volume. The patient has to stand 2.5 meters in front of the camera for several seconds, during which a depth map is registered and a skeletonization of the person is computed. The captured data is used to calculate the volume in real-time. The system was tested in seven volunteers, in which measurements were demonstrated to be repeatable and provided equivalent information to patient weight in terms of consistency.
The system is not quite ready for prime-time yet, as amongst other things it was found that volume estimations varied with patient positioning, which requires further study. To improve the system, the researchers suggest acquiring perpendicular depth maps by using two Kinect cameras positioned at a 90-degree angle to one another, or by recording depth maps with the patient in consecutive frontal and lateral positions with respect to a single camera. The research was presented recently at SIIM 2012 in Orlando.
More info at Aunt Minnie…