Create

Design Project Gallery

Project Search
Filter projects by keyword, program, course, or submission year.

Search Fields

Clear all filters

Synthetic-to-Real Learning for 3D Shape Prediction of FBG-Sensorized Stylets

Project Description:

This project develops a hybrid framework for real-time prediction of flexible needle shape during minimally invasive procedures. It leverages Fiber Bragg Grating (FBG) sensors to measure curvature and combines a physics-based Lie-group model with machine learning to reconstruct and predict 3D trajectories. The goal is to accurately estimate future needle position (e.g., next 30 mm) using both synthetic and real data, reducing placement errors and improving guidance without relying on continuous imaging.

Project Photo:

Diagram showing a multicore fiber embedded in a medical needle, with multiple sensing nodes and cross-sectional fiber layout used to measure strain and estimate curvature for 3D shape reconstruction.

FBG-based shape-sensing stylet used for real-time curvature measurement and 3D needle shape prediction.

Project Poster

Open full size poster in new tab (PDF)

Student Team Members

Sichen Deng

Course Faculty

Russell Taylor

Project Mentors, Sponsors, and Partners

I. IULIAN IORDACHITA
Yinsong Ma
Jacynthe Francoeur