Thesis Project Form
Title (tentative): Quantitative 3D Surface Analysis of Thoracic Deformities for Personalized Surgical Planning| Thesis advisor(s): Trò Rossella | E-mail: |
| Address: | Phone: |
Description
Motivation and application domain
Low-cost 3D surface scanning offers a non-invasive approach to assessing thoracic deformities. Building on previous work on Poland syndrome, this thesis will extend surface analysis to cases where neither hemithorax provides a healthy reference, with the aim of supporting preoperative planning and the design, selection, and positioning of patient-specific prostheses.
General objectives and main activities
The thesis will extend an existing 3D mesh-processing pipeline through:
- Development and comparison of reference-surface reconstruction methods for cases where mirroring the contralateral hemithorax is unsuitable, including bilateral deformities.
- Extraction of quantitative descriptors, including estimated volume deficit, maximum and mean depth, affected surface area, asymmetry, and local surface curvature.
- Implementation of interactive mesh colour maps showing depth relative to the reconstructed reference surface and separate maps of local slope to highlight abrupt contour changes.
- Integration of measurements and visualizations into a prototype supporting assessment of the correction required and personalized prosthesis planning.
- Evaluation of measurement repeatability, sensitivity to reference-surface selection, and clinical usefulness with surgeons.
A prototype providing surgeons with quantitative measurements and interactive 3D visualizations to support individualized planning. The reconstructed surface will represent a proposed correction target, whose suitability must be assessed with the surgeon, rather than a uniquely determined healthy anatomy.
- Development and comparison of reference-surface reconstruction methods for cases where mirroring the contralateral hemithorax is unsuitable, including bilateral deformities.
- Extraction of quantitative descriptors, including estimated volume deficit, maximum and mean depth, affected surface area, asymmetry, and local surface curvature.
- Implementation of interactive mesh colour maps showing depth relative to the reconstructed reference surface and separate maps of local slope to highlight abrupt contour changes.
- Integration of measurements and visualizations into a prototype supporting assessment of the correction required and personalized prosthesis planning.
- Evaluation of measurement repeatability, sensitivity to reference-surface selection, and clinical usefulness with surgeons.
A prototype providing surgeons with quantitative measurements and interactive 3D visualizations to support individualized planning. The reconstructed surface will represent a proposed correction target, whose suitability must be assessed with the surgeon, rather than a uniquely determined healthy anatomy.
Training Objectives (technical/analytical tools, experimental methodologies)
Development of skills in 3D mesh processing, geometric modelling, quantitative shape analysis, scientific visualization, and statistical validation using Python and/or 3D Slicer. Particular attention will be given to translating clinical requirements into reproducible measurements and an interpretable interface for surgical planning.
Place(s) where the thesis work will be carried out: DIBRIS; IRCCS Gaslini Children Hospital
Additional information
Pre-requisite abilities/skills: Basic programming skills in Python and/or MATLAB and familiarity with biomedical image processing. Knowledge of 3D geometry, mesh processing, or 3D Slicer is desirable but not required. Interest in developing computational tools for clinical applications and willingness to collaborate with surgeons.
Maximum number of students: 2