03.08.2015
Ultrasound is used in medicine for several therapeutic purposes; besides other applications, ultrasound can be used to selectively and locally heat tissue since absorbed ultrasound energy is converted into heat. During high-intensity therapeutic ultrasound (HITU) applications, for example, tumors are selectively heated and thus destroyed by overheating. The two most common methods for determining this increase in temperature are measurements with magnetic resonance imaging (MRI) and measurements using thermocouples. MRI thermometry is, on the one hand, not invasive and is therefore used in some applications for in vivo monitoring of the temperature distribution during treatment, but on the other hand, it is very costly and time-consuming. Additionally, it provides a comparatively poor spatial and temporal resolution and is usually not traceable. Measurements using thermocouples, however, are typically not performed in vivo, but in tissue-mimicking materials (so-called phantoms) whose measured heating can be used for calculating the heating in the body (under the same sonication conditions) or for verifying temperature prediction algorithms. Measurements with thermocouples are, on the contrary, easy to perform and traceable, but the determination of two-dimensional temperature distributions is very time-consuming, since measurements are possible only point-wise and before each point-wise measurement time allotment for complete cooling has to be given. Furthermore, the metal thermocouples are not MRI compatible, so that a direct comparison of the two methods is not possible.
Therefore, a setup was recently developed and accomplished at PTB for fast and easy measurements of two-dimensional temperature distributions in tissue phantoms. The basis of this approach is thermotropic liquid crystal foils ("TLC foils") that are commonly used, for example, to visualize the temperature in aquariums or wine coolers. These foils are reversibly thermochromic - that means that they change their color when heated and take on their original color when cooled again. With the developed setup it is possible, firstly, to independently and traceably validate MRI temperature measurements; secondly, to perform quick and uncomplicated constancy tests of therapeutic ultrasound devices ("Does the device still induce the same temperature distribution?"); and thirdly, to experimentally verify calculation algorithms that predict temperature distributions with little effort.
Figure 1: Left: Visualization of the temperature increase induced by the sound field of a high-intensity therapeutic ultrasound transducer (bottom of the picture) as color change of a TLC foil inside a transparent tissue-mimicking phantom. Right: Schematic setup for quantitative recording of the color changes; relevant abbreviations are explained in the text.
For the realization of the setup, it was necessary to especially develop transparent tissue-mimicking materials (TMM in Figure 1) that allow the observation of the color change of the TLC foils with a camera (C). Furthermore, a reproducible illumination of the foils and effective protection of the camera against the sound field and the magnetic field of the MRI had to be implemented. The former was realized through the use of an LED and diffuse lighting via a light guide (LG) and a dispersing prism (P). The protection against the sound field has been ensured by the use of highly absorbent oil (OIL) and an indirect observation of the TLC foils via a totally reflecting surface (G). The entire setup was designed and accomplished MRI compatible by the exclusive use of appropriate materials - only the camera has to be kept away from the magnetic field of the MRI scanner by an adequate installation (see Figure 2).
Figure 2: The developed setup (black box in the middle of the figure, the camera is situated at the right end of the gray tube) in operation in an MRI scanner at “Fraunhofer-Institut für Bildgestützte Medizin” in Bremen.
Furthermore, for quantitative temperature measurements with the developed setup a software for automatic recording of the color information with the camera and subsequent conversion of the recorded pictures into two-dimensional temperature maps was realized and a traceable calibration of the assignment of "color" to "temperature" with an adjustable electric heating foil and calibrated thermocouples was implemented. Corresponding studies have shown here that among the different possibilities of color analysis the processing in the HSV color space (hue, saturation, value) provides a higher accuracy and repeatability than in the RGB color space (red, green, blue).
Contact persons:
Julian Haller, FB 1.6, AG 1.62, E-Mail: 
julian.haller @ptb.de
Volker Wilkens, FB 1.6, AG 1.62, E-Mail: volker.wilkens @ptb.de