[Defense] Statistical Shape Modeling and Post-Operative Shape Estimation in Reconstructive Breast Surgeries
Friday, December 9, 2022
2:00 pm - 3:00 pm
In
Partial
Fulfillment
of
the
Requirements
for
the
Degree
of
Doctor
of
Philosophy
Urmila
Sampathkumar
will
defend
her
proposal
Statistical
Shape
Modeling
and
Post-operative
Shape
Estimation
in
Reconstructive
Breast
Surgeries
Abstract
Simulation and prediction tools for communicating postoperative breast shape changes and for elicitation of patient preferences following reconstructive surgery are limited. We propose a shape simulation method on 3D surface images that allows data-driven deformable, non-rigid localized changes to breast shape. Active Shape Models (ASM) is a statistical approach, that leverages specificity (common features) and variability (features of significant variations in the given dataset) for shape modeling. We previously developed a spherical harmonics-based breast shape model (SPHARM). In this work we utilize the descriptors from SPHARM modeling with ASM to achieve breast shape simulations. The ASM shape parameters are then mapped to semantically and clinically relevant breast features (such as breast height, width, projection, ptosis) through fitting multi variate regression. To obtain a prediction of post-operative breast shape, we utilize data fusion using Kalman Filtering. The ability to preoperatively obtain an estimate of postoperative shape and create desired breast shape simulations will help plastic surgeons in surgical planning and patient expectation management.
Friday,
December
9,
2022
2:00PM
-
3:00PM
CT
Online via
Dr. Fatima Merchant, dissertation advisor
Faculty, students, and the general public are invited.
