Share Email Print

Proceedings Paper

Safe electrode trajectory planning in SEEG via MIP-based vessel segmentation
Author(s): Davide Scorza; Sara Moccia; Giuseppe De Luca; Lisa Plaino; Francesco Cardinale; Leonardo S. Mattos; Luis Kabongo; Elena De Momi
Format Member Price Non-Member Price
PDF $17.00 $21.00

Paper Abstract

Stereo-ElectroEncephaloGraphy (SEEG) is a surgical procedure that allows brain exploration of patients affected by focal epilepsy by placing intra-cerebral multi-lead electrodes. The electrode trajectory planning is challenging and time consuming. Various constraints have to be taken into account simultaneously, such as absence of vessels at the electrode Entry Point (EP), where bleeding is more likely to occur. In this paper, we propose a novel framework to help clinicians in defining a safe trajectory and focus our attention on EP. For each electrode, a Maximum Intensity Projection (MIP) image was obtained from Computer Tomography Angiography (CTA) slices of the brain first centimeter measured along the electrode trajectory. A Gaussian Mixture Model (GMM), modified to include neighborhood prior through Markov Random Fields (GMM-MRF), is used to robustly segment vessels and deal with the noisy nature of MIP images. Results are compared with simple GMM and manual global Thresholding (Th) by computing sensitivity, specificity, accuracy and Dice similarity index against manual segmentation performed under the supervision of an expert surgeon. In this work we present a novel framework which can be easily integrated into manual and automatic planner to help surgeon during the planning phase. GMM-MRF qualitatively showed better performance over GMM in reproducing the connected nature of brain vessels also in presence of noise and image intensity drops typical of MIP images. With respect Th, it is a completely automatic method and it is not influenced by inter-subject variability.

Paper Details

Date Published: 3 March 2017
PDF: 8 pages
Proc. SPIE 10135, Medical Imaging 2017: Image-Guided Procedures, Robotic Interventions, and Modeling, 101352C (3 March 2017); doi: 10.1117/12.2254474
Show Author Affiliations
Davide Scorza, Politecnico de Milano (Italy)
Vicomtech-IK4 (Spain)
Biodonostia Health Research Institute (Spain)
Sara Moccia, Politecnico di Milano (Italy)
Istituto Italiano di Tecnologia (Italy)
Giuseppe De Luca, Politecnico di Milano (Italy)
Lisa Plaino, Politecnico di Milano (Italy)
Francesco Cardinale, Grande Ospedale Metropolitano Niguarda (Italy)
Leonardo S. Mattos, Istituto Italiano di Tecnologia (Italy)
Luis Kabongo, Vicomtech-IK4 (Spain)
Biodonostia Health Research Institute (Spain)
Elena De Momi, Politecnico di Milano (Italy)

Published in SPIE Proceedings Vol. 10135:
Medical Imaging 2017: Image-Guided Procedures, Robotic Interventions, and Modeling
Robert J. Webster III; Baowei Fei, Editor(s)

© SPIE. Terms of Use
Back to Top
Sign in to read the full article
Create a free SPIE account to get access to
premium articles and original research
Forgot your username?