P2C-3 An Injury Mimicking Ultrasound Phantom as a Training Tool for Diagnosis of Internal Trauma

Injury mimicking ultrasound phantoms are training devices that can emulate pre- and post-injury conditions at specific regions of human anatomy. As such, they are likely to be useful tools for teaching medical personnel how to recognize trauma conditions from ultrasound images. Due to the increased use of portable ultrasound systems, earlier diagnosis of internal trauma will be feasible at locations such as traffic accidents, earthquakes, battlefields and terrorist attacks. This paper describes a prototype injury mimicking ultrasound phantom for the peritoneal cavity. This phantom is designed by placing agar 'organs' into a half cylinder cavity covered with a distensible latex membrane, that is sufficiently thin to allow ultrasonic wave propagation with little attenuation. Bleeding is emulated by injecting a known fluid volume into the inter-organ space. We have analyzed the distribution of inter-organ fluid volumes at 'pre-injury', 'post-injury' and after a return to 'pre-injury' conditions. In the pre-injury state, the tissue mimicking material represents a 91% volume fraction, with the remaining volume fraction filled with distilled water. We captured a 3D image in the form of a series of equidistant 2D ultrasound images along a scan path at the vertex of the phantom. To reach the post-injury state, we injected fixed fluid volumes up to a total of 600 mL into the phantom. We captured images along the same scan path as with the pre-injury status. To return to the pre-injury state, we extracted the injected fluid volume and captured another set of images along the same scan path. From the 3D image of the phantom, we were able to obtain volume estimations for two regions of inter-organ volume spaces. We found that the volumes for pre-injury and post-injury are significantly different, changing from an average of 5.4 cm³ to 12.2 cm³. Changing the density from 1.0 to 1.08 g/cm³ the injected fluid revealed no shift in fluid accumulation locations. We numerically modeled an ultrasound phantom to simulate the locations where that the injected fluid volume would accumulate for various and organ parameters. In the physical and the numerically modeled phantom, we have found that the fluid tends to distribute uniformly throughout the numerical phantom without a dependence on the density of the injected fluid