Mechanics based modeling for image guidance in breast conserving surgery

Abstract

Breast conserving surgery (BCS) is a common procedure for early-stage breast cancer where the tumor is removed with a surrounding margin of healthy tissue. BCS aims to excise breast tumors while preserving cosmesis, but re-excision rates for BCS are high (~17%) despite guideline standardization and new localization technologies. Intraoperative navigation platforms may improve tumor extent and boundary visualization, reducing re-excision rates and improving patient outcomes. This dissertation focuses on improving intraoperative navigation for BCS by developing an image guidance system (IGS) that leverages preoperative imaging and nonrigid registration. More specifically, mechanics based models that compensate for soft tissue deformations in the breast are explored.

The dissertation is structured around three specific aims. The first aim uses supine magnetic resonance (MR) image-to-image registration to inform finite element method (FEM) modeling of surgical deformations. Three FEM models with varying levels of heterogeneity and anisotropy are implemented to evaluate how model complexity affects target accuracy. An intra-subject supine breast MR dataset from healthy volunteers that simulates surgical deformations is used for this evaluation.

In the second aim, breast deformation models are adapted for the intraoperative setting by using regularized Kelvinlet functions for registration. Regularized Kelvinlet functions are analytic solutions that model linear elasticity in an infinite medium. This registration method is compatible with the time constraints and the sparse-data sources available in the surgical environment. Evaluation is performed on the same dataset from the first aim and a breast cancer patient case.

The final aim involves integrating the deformation correction method from the second aim into a BCS-IGS designed for bedside data collection and testing. Breast phantom experiments and a healthy volunteer demonstration are conducted to evaluate system utility. Overall, this dissertation presents a novel image guidance platform to compensate for soft tissue deformations in the breast that may unlock new capabilities in surgical precision for BCS procedures.