Research participants:
C.A.T. van den Berg,
A.L.H.M.W. van Lier,
A.A.C. de Leeuw,
B.W. Raaymakers and
J.J.W. Lagendijk When heat is deposited in human tissue, there are several mechanisms which transport heat as is demonstrated in Figure 1. First of all, thermal conductivity transports heat in the direction of the temperature gradient. A second mechanism capable of transporting much more energy is convective heat transport. When a large vessel with cold blood enters a piece of heated tissue, it starts exchanging heat with the tissue environment. As the vessel branches become smaller, this heat exchange will become more significant up to a point where the blood is in thermal equilibrium with the tissue.
Figure 1. When heat is deposited in human tissue, the heat will be distributed by thermal conductance and convective heat transport due to blood vessels.
In our group we have develop a unique thermal simulation package called DIVA capable of including the effect of discrete vessels. In figure 2 the effect of including the whole vesseltree is illustrated. The full vessel tree results in a very inhomogeneous temperature distribution with "cold tracks" around the larger vessel which are not yet in thermal equilibrium. As we strip the smaller vessels from the vessel tree, the temperature distribution becomes more homogeneous.
Figure 2: The inclusion of the full vasculature results in heterogeneous temperature distributions. Including only the larger vessels, results in more homogeneous temperature distributions.
An example of the possibilities of DIVA is the simulation of the induced temperature rise due to a mobile phone. See Figure 3. For this purpose we built first a dielectric model of human head based upon segmentation of MR images of a patient. FDTD simulations were applied to calculated the RF energy deposition distribution in the human brain due to a mobile phone. To translate this into a temperature distribution, we constructed a detailed thermal model. MR angiography was applied to make a 3D vessel model of the arterial and venous vasculature of a human head. This demonstrate that mobile phone use leads to a negligible temperature rise of 0.11 degrees Celsius for an antenna with an average emitted power of 0.25 W
Figure 3: Thermal model of the human head using a discrete vessel model consisting of arteries (red) en veins (blue). b. The RF energy deposition of mobile phone held near the ear portraited on the outer skin . c. The resulting temperature rise in three-orthogonal planes.
Another example of our thermal modelling research is the simulation of a regional hyperthermia treatment of the prostate. In regional hyperthermia radiowaves are focussed on a tumour to heat the tumour tissue. The induced temperature is dependent on various processes. However, a dominant factor is the local blood flow and the temperature of the incoming blood. If the blood temperature is lower than the surrounding tissue, the blood cools the tissue. If on the other hand, the blood is preheated at some upstreams location, it will heat the tissue. To investigate this phenomenon for the prostate we used DIVA together with various imaging modalities to acquire information about the vasculature and local tissue perfusion.
With imaging techniques like Dynamic Contrast Enhanced MRI (DCE-MRI) and DCE multi slice CT, it is possible to obtain 3D blood flow maps of the prostate. See Figure 4. A detailed model of the vasculature of the pelvis can be made with multi-slice CT angiography. See figure 5. Much care in this project was given to the thermal impact of small vessels (0.1-1 mm diameter) in the prostate. The prostate vasculature was reconstructed using cryo-microtome slicing from post-mortem prostates.
Figure 4: Two examples of measured perfusion maps obtained by Dynamic Contrast Enhanced MultiSlice CT imaging for two different patients. Regions of elevated perfusion were found which corresponded with the tumour location found by ultrasound and palpation.
Figure 5: Angiogram of pelvic vessels obtained by Multi Slice CT angiography. Depicted are the Iliac arterial (red) and venous (blue) branches. The Inferior Mesentric Artery is shown in yellow. The blood supply of the prostate was found to vary among patients. Both the Iliac arteries as well as the Inferior Mesentric artery were found to be involved in the prostatic blood supply.
Combining all the data from the various modalities a detailed thermal model of the prostate could be made. This model allowed the investigation of pre-heating of blood in the vessels supplying the prostate, the effect of the systemic body temperature on the tumour temperature and the influence of the venous return temperature from the legs on the tumour temperature. See Figure 6 and 7.
Figure 6: Result of temperature calculations at 5x5x5 mm.
a:) Tomogram
b:) Homogeneous perfusion
c:) Perfusion map
d:) Perfusion map + pelvic vasculature
Note the "cold spot" in tumour marked by the arrow in c and d.
Figure 7: Pre-heating depicted in a false colour scale over the prostate vasculature. As can be seen, the thermal equilibration of the blood in the vessels with the tissue takes place for an important part in the small vessels of the prostate capsule.