Biomedical Engineering Reference
In-Depth Information
Table 26.1
Condensation for the parameter setup of constitutive models for parenchyma and cap-
sule as used for the FE-analysis of the aspiration test. The tissues were modeled fully incom-
pressible. RP: reduced polynomial, QLV: quasi-linear viscoelasticity in terms of Prony-series, RB:
Rubin-Bodner. Note that the capsule model is expressed in terms of membrane properties
RP
QLV
RB
C
10
(kPa)
C
20
(kPa)
g
1
(-)
τ
1
(s)
g
2
(-)
τ
2
(s)
m
μ
0
(N/mm)
q
(-)
Parenchyma
0
.
7
4
.
5
0
.
49
1
.
9
0
.
19
6
.
5
Capsule
4
.
81
0
.
052
26.3.3 Open Surgery Measurements
Aspiration measurements were performed on the fully perfused organ after open-
ing of the abdomen and mobilization of the liver for resection. Typically the whole
measurement procedure did not exceed 5 min. The measuring sites were defined on
the part of the organ to be resected, which then was available for biopsy acquisition.
First, the aspiration measurement was realized 3-times on normal tissue for refer-
ence. Subsequently, the measurement was repeated 3-times on the lesion. The tissue
was not preconditioned. The aspiration cycle was set for a step decrease to a target
suction pressure of 200 mbar with a hold time of 9 s. One measuring cycle took
about 30 s. The measurement procedure was established based on initial trials, as
follows: the probe head is kept perpendicular and just in contact with the organ sur-
face, keeping the downforce as small as possible, while the surgeon visually tracks
the motion of the organ due to the artificial respiration of the patient; once a stable
situation has been accomplished, the operator initiates the measurement; the sur-
geon tracks the motion of the liver keeping the probe in a constant relative position
to avoid changes in the contact conditions and dependence of the measurement on
the far-field boundary conditions; at the end of each measuring cycle, the surgeon
carefully retracts the probe and visually checks the measurement site.
The repeatability of the measurements was assessed based on the stiffness pa-
rameter (Sect.
26.2.1.2
). The coefficient of variation was calculated as the standard
deviation of normalized with respect to the measuring site mean value. The coeffi-
cient of variation for the present data resulted in 3
.
9 % for the reference sites, 4
.
8%
for the lesion sites, and 4
.
2 % overall. Considering previous data (Mazza et al.,
2007
;
Nava et al.,
2008
), with typical values of the coefficient of variation in the range of
20 %, a fivefold increase in repeatability was achieved. Most probably these im-
provements are due to the minimization and better control of the contact force, and
are due to the fact that always the same surgeon performed the measurements.
Figure
26.8
(a) reports all measured stiffness values for normal tissue and lesion
with the corresponding connective tissue content as measured by histo-pathological
analysis (Hollenstein,
2011
). Although a weak separation between lesion (higher
values) and normal tissue can be observed, no significant correlation could be found
between stiffness value and connective tissue content. More interesting findings can
be worked out when considering the relative difference between the 'reference' site
