# FEM simulation of breast deformation with semi-fluid representation

## TL;DR

This paper introduces a new simulation method for breast deformation in cancer surgery, using a semi-fluid model to improve accuracy.

## Contribution

A novel finite element method is proposed to simulate semi-fluid breast deformation, incorporating tissue mobility and anchoring structures.

## Key findings

- The tumor center registration error was 11.87 ± 4.05 mm with the proposed method.
- Semi-fluid representation reduced the tumor’s Hausdorff distance from 12.89 ± 6.24 mm to 11.50 ± 4.38 mm.
- The method improves breast posture conversion accuracy by considering fluidic tissue behavior.

## Abstract

In image-guided surgery for breast cancer, the representation of the breast deformation between planning and surgery plays a key role. The breast deforms significantly and behaves as a fluid with some constraints. Concretely, the deep fat layer in the breast deforms fluidly due to its incomplete fixation to the chest wall, while the anchoring structures by fascia avoid excessive deformation. In this study, we propose a method to simulate the semi-fluid deformation of the breast, considering the fluidic properties of the adipose tissue under the constraints of the anchoring structures.

The proposed method prioritizes anatomical features of the breast, enhancing tissue mobility near the chest wall and modeling the anchoring structure of the fascia along the inframammary fold. To simulate semi-fluid deformation, constraint force from anchoring structure is applied to prone-positioned breast model, using a finite element method.

The results of the evaluation indicate a tumor center registration error of 11.87 ± 4.05 mm. Additionally, we verified how semi-fluid representation affects the registration error. The tumor’s Hausdorff distance decreased from 12.89 ± 6.24 mm to 11.50 ± 4.38 mm with considering semi-fluidity.

The results showed that the use of semi-fluid representation tends to reduce registration errors. Therefore, it was suggested that the proposed method could improve the accuracy of breast posture conversion.