426
IL. Aroyan et al.
simpler tissue models. As we shall see, multiple maxima appear in most of
the models, although the ambiguity decreases significantly when the soft
tissue density and velocity information and the air sacs are included. This
is analogous to the situation in inverse seismologic simulations where ambiguity decreases with the accuracy and completeness of the geologic models
assumed.
The positions of these 'hypocenters' for inverse simulations at the four
frequencies, 25, 50, 75, and 100 kHz, are shown in two PSR projections
(viewed from the right and the front) in Figures 1O.6c and d for the same
direction of return propagation (an inverse (9, <1» = (5°,0°) beam in the
angular coordinate system of Figure 1O.A1a in the Appendix). The maxima
up against the skull is shown when it exists and also the "soft tissue" position (not against the skull) of highest potential energy density for each frequency. The general trends in the positioning of these skull-only focal
maxima are clear. Along an anterior to posterior axis, foci only occur in the
region of the upper narial depression, within a distance of about 3 cm (generally less) from the slope of the frontal bone. Along a transverse axis, they
appear to fall along a line running up the right side of the narial depression, and not on the left. The two lowest foci appear to lie within the vertical range of the right nasal plug, while the main cluster lies roughly 1cm
behind and below the location of the right MLDB.
Consider now the emission patterns that result from placing point sources
at the focal positions of the skull. Figure 10.7 illustrates the far-field emission distributions computed (using the forward simulation and extrapolation techniques discussed in Section 2.1 and Appendix A) for a point source
with a frequency of75 kHz placed at the (9, <1» = (5°,0°) inverse simulation
point of maximum potential energy density shown in Figures 1O.6c and d
(see Appendix A for explanation of the mapping of emission directions).
Peak emission angles (in degrees) and the directivity indexes for these patterns are included in the figure. While a fair amount of energy is projected
to the sides and upward, the skull by itself is capable of forming significant
forward beams solely via reflection off of its upper surface for sources
placed in the vicinity of the right narial depression. Emission simulations
at 25,50, and 100kHz also resulted in forward beams similar to Figure 10.7,
with directivity indexes increasing approximately linearly with the logarithm of frequency.
3.2 Skull and Soft Tissue Simulations
For the skull and soft tissue model, the density and velocity of all points
within the selected forehead region (see Figures 1O.5a and b) were assigned
by the mappings described in Section 2.2. Because these mappings assign
all air CT scan voxels to the background seawater density (1.03 glcc) and
velocity (1,500mls) values, air sac regions were effectively "filled" with seawater in this model.
IL. Aroyan et al.
simpler tissue models. As we shall see, multiple maxima appear in most of
the models, although the ambiguity decreases significantly when the soft
tissue density and velocity information and the air sacs are included. This
is analogous to the situation in inverse seismologic simulations where ambiguity decreases with the accuracy and completeness of the geologic models
assumed.
The positions of these 'hypocenters' for inverse simulations at the four
frequencies, 25, 50, 75, and 100 kHz, are shown in two PSR projections
(viewed from the right and the front) in Figures 1O.6c and d for the same
direction of return propagation (an inverse (9, <1» = (5°,0°) beam in the
angular coordinate system of Figure 1O.A1a in the Appendix). The maxima
up against the skull is shown when it exists and also the "soft tissue" position (not against the skull) of highest potential energy density for each frequency. The general trends in the positioning of these skull-only focal
maxima are clear. Along an anterior to posterior axis, foci only occur in the
region of the upper narial depression, within a distance of about 3 cm (generally less) from the slope of the frontal bone. Along a transverse axis, they
appear to fall along a line running up the right side of the narial depression, and not on the left. The two lowest foci appear to lie within the vertical range of the right nasal plug, while the main cluster lies roughly 1cm
behind and below the location of the right MLDB.
Consider now the emission patterns that result from placing point sources
at the focal positions of the skull. Figure 10.7 illustrates the far-field emission distributions computed (using the forward simulation and extrapolation techniques discussed in Section 2.1 and Appendix A) for a point source
with a frequency of75 kHz placed at the (9, <1» = (5°,0°) inverse simulation
point of maximum potential energy density shown in Figures 1O.6c and d
(see Appendix A for explanation of the mapping of emission directions).
Peak emission angles (in degrees) and the directivity indexes for these patterns are included in the figure. While a fair amount of energy is projected
to the sides and upward, the skull by itself is capable of forming significant
forward beams solely via reflection off of its upper surface for sources
placed in the vicinity of the right narial depression. Emission simulations
at 25,50, and 100kHz also resulted in forward beams similar to Figure 10.7,
with directivity indexes increasing approximately linearly with the logarithm of frequency.
3.2 Skull and Soft Tissue Simulations
For the skull and soft tissue model, the density and velocity of all points
within the selected forehead region (see Figures 1O.5a and b) were assigned
by the mappings described in Section 2.2. Because these mappings assign
all air CT scan voxels to the background seawater density (1.03 glcc) and
velocity (1,500mls) values, air sac regions were effectively "filled" with seawater in this model.