Assessment of an automatic pipeline for probabilistic tractography in drug-resistant epileptic patients

Cattoretti, M. and Berta, L. and Felisi, M. and Sartori, I. (2023) Assessment of an automatic pipeline for probabilistic tractography in drug-resistant epileptic patients. Il nuovo cimento C, 46 (4). pp. 1-4. ISSN 1826-9885

Preview

Text
ncc12689.pdf - Published Version
Download (258kB) | Preview

Official URL: https://www.sif.it/riviste/sif/ncc/econtents/2023/...

Abstract

Diffusion Tensor Imaging is a special technique of Magnetic Resonance Imaging employed for fiber-tracking studies for neurosurgical planning. The aim of this work is to validate the fiber-tracking probabilistic workflow in drug resistant epileptic patients explored by stereoelectroencephalography (SEEG) at ASST GOM Niguarda Hospital. The method is based on anatomical Regions-Of Interest (ROI) definition in a common brain template co-registered on each patient’s image. The pipeline was developed in Python language using the prob trackx algorithm for fiber-tracking available in FSL platform with ROIs as input (http://www.fmrib.ox.ac.uk/fsl/). The accuracy of the method was assessed by automatically comparing reconstructed fibers with pathways previously elaborated during clinical practice, in which ROIs were drawn manually in each patient’s images. In 60% of the cases, automatic fibers were found to be anatomically consistent, if compared to anatomical knowledge, while in 36% of the cases, minimal manual corrections enabled appropriate anatomically consistent fibers to be obtained through a fast process. The proposed fibre tracking methodology suggested the feasibility of the process standardization and the achieved results will be further employed for mutual validation of tractographic process and electrophysiological data, which is the reference method for the clinical validation of fiber-tracking preoperative planning.

Item Type:	Article
Subjects:	500 Scienze naturali e Matematica > 530 Fisica
Depositing User:	Marina Spanti
Date Deposited:	15 Feb 2024 10:41
Last Modified:	15 Feb 2024 10:41
URI:	http://eprints.bice.rm.cnr.it/id/eprint/22701

Actions (login required)

View Item