S. Raith 1, M.Eder 1, F. v Waldenfels 1, J. Jalali 2, A.Volf 1, L. Kovacs 1

(1)

element simulation

S. Raith1_{, M.Eder}1_{, F. v Waldenfels}1_{, J. Jalali}2_{, A.Volf}1_{, L. Kovacs}1

1_{Research Group CAPS (Computer Aided Plastic Surgery) - Klinik und Poliklinik für Plastische} Chirurgie und Handchirurgie, Klinikum rechts der Isar, Technische Universität München, Ismaninger

Str. 22, D-81675 Munich, Germany

2_{Institute of Medical Engineering at the Technische Universität München (IMETUM), Boltzmannstr.} 11, D-85748 Garching, Germany

1 10/17/2012

(2)

9. Weimarer Optimierungs- und Stochastiktage

Clinic for Plastic Surgery and Handsurgery

Klinikum rechts der Isar

Technische Universität München

medical staff:

engineering staff:

Who is research group CAPS?

Computer Aided Plastic Surgery

10/17/2012 2

Klinikum rechts der Isar _{IMETUM Garching}

PD Dr. Laszlo Kovacs

Head of the group Dr. M. Eder

Dr. F.v.Waldenfels _{A. Volf} _{J. Jalali} _{S. Raith}

(3)

•

Breast cancer is very common: 140.337 surgical interventions

in Germany in 2010

[source: S. B. Deutschland, „Statistisches Bundesamt,“ [Online]. Available:

https://www.destatis.de/DE/ZahlenFakten/GesellschaftStaat/Gesundheit/Krankenhaeuser/Tabellen/DiagnosenWeiblich.html. [4 9 2012].]

•

Additionally there are reductions, asymmetry corrections,

reconstructions and augmentations

•

There is a need for better planning in breast surgery to

overcome todays ad-hoc heuristic approaches

–

Software tools from engineering science might be used for planning



Simulations of the mechanical behavior of the breast’s soft tissues

is evident

Current Situation and Motivation

(4)

 Krouskop, T., Wheeler, T., Kallel, K., Garra, B., Hall, T.,

1998. Elastic moduli of breast and prostate tissues under compression. Ultrason Imaging. 20, 260-274.

 Wellman, P., Howe, R., Dalton, E., Kern, K., 1999. Breast tissue stiffness in compression is correlated to histological diagnosis. Harvard Biorobotics Laboratory; Availabe at http://biorobotics.harvard.edu/pubs/1999/mechprops.pdf Accessed March 01, 2012

 Azar, F., Metaxas, D., Schnall, M., 2001. A deformable finite element model of the breast for predicting mechanical deformations under external perturbations. Acad Radiol. 8, 965-975.

 Samani, A., Bishop, J., Yaffe, M.J., Pelwes, D., 2001. Biomechanical 3D finite element modeling of the Human Breast using MRI data. IEEE Trans Med Imaging. 20, 271-279.

 Samani, A., Plewes, D., 2004. A method to measure the hyperelastic parameters of ex vivo breast tissue samples. Phys Med Biol. 49, 4395-4405.

 Tanner, C., Schnabel, J.A., Hill, D.L., Hawkes, D.J., Leach, M.O., Hose, D.R., 2006. Factors influencing the accuracy of biomechanical breast models. Med Phys. 33, 1758-1769.

 Samani, A., Zubovits, J., Plewes, D., 2007. Elastic moduli of normal and pathological human breast tissues: an inversion-technique-based investigation of 169 samples. Phys. Med Biol. 52, 1565-1576.

 Rajagopal, V., Lee, A., Chung, J., Warren, R., Highnam R.,

2008. Creating individual-specific biomechanical models of the breast for medical image analysis. Acad Radiol. 15, 1425–1436.

 Del Palomar, A.P., Calvo, B., Herrero, J., López, J., Doblaré, M., 2008. A finite element model to accurately predict real deformations of the breast. Med Eng Phys. 30, 1089-1097.

 Lapuebla-Ferri, A., Del Palomar, A.P., Herrero, J., Jiménez-Mocholí, A.J., 2011. A patient-specific FE-based methodology to simulate prosthesis insertion during an augmentation mammoplasty. Med Eng Phys. 33, 1094-1102.

Publications dealing with mechanical behaviour of

female breast soft tissue

(5)

Publication

Krouskop et al. (1998)

Wellman et al. (1999)

Azar et al. (2001)

Samani et al. (2001)

Tanner et al. (2006)

Del Palomar et al. (2008)

Rajagopal et al. (2008)

Lapuebla-F. et al. (2011)

Theoretical constitutive models

utilized by the different authors

10/17/2012 5

Constitutive Model

Piecewise linear elastic

Exponential Hyperelastic

Polynomial Elastic

Mooney-Rivlin

Linear Elastic

Various Material Models

Neo-Hookean

Different testing methods such as ex vivo material tests (uni- or biaxial tension, shear)

or numerical simulations(FEA)

(6)

9. Weimarer Optimierungs- und Stochastiktage 6 10/17/2012

(7)

•

Taking advantage of modern imaging technologies

–

MRI imaging

•

inner anatomy of the test persons’ chests

•

possible in prone or supine position

–

3-D surface scanning

•

possible in standing position

Method for in vivo calculation of the breast’s

material parameters

(8)

•

Taking advantage of modern imaging technologies

Method for in vivo calculation of the breast’s

material parameters

10/17/2012 8

3-D surface scan Standing position

(9)

10/17/2012 9

Principle workflow of the

material model validation

Machbarkeitsstudie

D

(10)

9. Weimarer Optimierungs- und Stochastiktage 10

3-D Laser Scan im Brustbereich

10/17/2012

scanning device:

(11)

stitching and merging of the 3 separate surface scan data

Utilized software tool:

3-D laser scans of the breast region

(12)

(13)

(14)

10/17/2012 14

•

3 different compartments are defined:

•

thorax wall, pectoral muscle and soft tissue (fat

and glandular)

Segmentation of the relevant anatomical

compartments

(15)

Finite element mesh for simulation

(16)

Finite element mesh for simulation

10/17/2012 16

System boundaries (green)

-> fixed boundary condition

Thorax treated as rigid

-> fixed boundary condition

Muscle also considered to be rigid -> fixed boundary condition

(17)

(18)

10/17/2012 18

•

Apparent deformation due to gravity loading is non negligible

•

-> for physically meaningful calculations a stress free starting configuration

has to be found

(19)

Inverse algorithm for the determination of the

unloaded breast shape

deformed configuration

segmented from MRI data

First estimation of the

unloaded configuration

new forward calculation

of the deformed shape

g

19

Comparison

(20)

better approximation of the undeformed configuration

reaction forces may be determined form

the comparison to the undeformed configuration

comparison via

prescribed displacements

!F

!D

Prob 04

g

forward calculation

10/17/2012

(21)

Iterativer Prozess

21 10/17/2012

(22)

10/17/2012 22

variation of material parameters according to Tanner et al. and further

Neo Hookean (Young’s modulus varying, PR=0.49)

Simulation results with different sets of

material properties

* Tanner et al., Factors influencing the accuracy of biomechanical breast models, Am. Assoc. Phys. Med. 33 (2006) 1758-1769

(23)

10/17/2012 23

Registration of surfaces in one common

coordinate system

(24)

•

Coordinate systems of simulation an scans have to be

registered for the comparison.

•

Calculation of 3D deviation (so called Hausdorff distance)

and summed over the surface as one scalar value

3-D surface comparison

(25)

Comparison of different simulation

results with

material parameters too stiff

matching material parameters

(26)

Principle workflow of the

parameter identification loop

10/17/2012 26

D

Optimization loop

Preprocessing

MR Imaging 3-D surface scan FEA Mesh merged surface

D

Material parameters 3D compare scalar value of accordance FEA result

(27)

10/17/2012 27

Results:

(28)

10/17/2012 28

Patient individual optimal material parameters:

Young‘s modulus: 0.000827 MPa

(Factor 1.06 to Tanner et al.)

Poisson‘s ratio: 0.5 (perfectly incompressible materail)

Results:

(29)

10/17/2012 29

Results:

clearly defined optimal set of parameters

Material parameters: Poisson‘s ratio always at 0.5 Young‘s Modulus as

a factor to the

parameters of Tanner at al. Output: average deviation in mm

(30)

10/17/2012 30

•

In vivo approximation of material properties is feasible

–

Based on a combination of:

•

3-D surface scans of standing position

•

MRI in prone lying position

–

non-invasive approach for material parameter determination

–

Patient individual determination of material parameters is possible

•

Acquired material parameters may be used for surgical

planning and operation simulations

Conclusion

(31)

•

application to training set of test persons to get better

statistical evidence (18 are yet available)

•

using more complex material models

–

Mooney-Rivlin, Ogden etc…

–

with more parameters than E and nu (or G and K, respectively)

•

material is yet considered to be homogeneous and isotropic

–

glandular tissue is not considered

Outlook

31 10/17/2012

(32)

10/17/2012 32

Determined material parameters may be used in clinical

surgery planning e.g. breast reconstruction

(33)

10/17/2012 33

Thank you for your attention

www.caps.me.tum.de

Contact: Stefan Raith

stefan.raith@caps.me.tum.de