D Kozinska1, F Carducci, K Nowinski. 1. Interdisciplinary Center for Mathematical and Computational Modelling, University of Warsaw, ul. Pawinskiego 5a, 02-106, Warsaw, Poland. kozinska@icm.edu.pl
Abstract
OBJECTIVES: We developed a new technique of fully automatic alignment of brain data acquired with scalp sensors (e.g. electroencephalography/evoked potential (EP) electrodes, magnetoencephalography sensors) with a magnetic resonance imaging (MRI) volume of the head. METHODS: The method uses geometrical features (two sets of head points: digitized from the subject and extracted from MRI) to guide the alignment. It combines matching on 3 dimensional (3D) geometrical moments that perform the initial alignment, and 3D distance-based alignment that provides the final tuning. To reduce errors of the initial guessed computation resulting from digitization of the head surface points we introduced weights to compute geometrical moments, and a procedure to remove outliers to eliminate incorrectly digitized points. RESULTS: The method was tested on simulated (Monte Carlo trials) and on real data sets. The simulations demonstrated that for the number of test points within the range of 0.1-1% of the total number of head surface points and for the digitization error in the range of -2-2 mm the average map error was between 0.7 and 2.1 mm. The average distance error was less than 1 mm. Tests on real data gave the average distance error between 2.1 and 2.5 mm. CONCLUSIONS: The developed technique is fast, robust and comfortable for the patient and for medical personnel. It registers scalp sensor positions with MRI head volume with accuracy that is satisfactory for localization of biological processes examined with a commonly used number of scalp sensors (32, 64, or 128).
OBJECTIVES: We developed a new technique of fully automatic alignment of brain data acquired with scalp sensors (e.g. electroencephalography/evoked potential (EP) electrodes, magnetoencephalography sensors) with a magnetic resonance imaging (MRI) volume of the head. METHODS: The method uses geometrical features (two sets of head points: digitized from the subject and extracted from MRI) to guide the alignment. It combines matching on 3 dimensional (3D) geometrical moments that perform the initial alignment, and 3D distance-based alignment that provides the final tuning. To reduce errors of the initial guessed computation resulting from digitization of the head surface points we introduced weights to compute geometrical moments, and a procedure to remove outliers to eliminate incorrectly digitized points. RESULTS: The method was tested on simulated (Monte Carlo trials) and on real data sets. The simulations demonstrated that for the number of test points within the range of 0.1-1% of the total number of head surface points and for the digitization error in the range of -2-2 mm the average map error was between 0.7 and 2.1 mm. The average distance error was less than 1 mm. Tests on real data gave the average distance error between 2.1 and 2.5 mm. CONCLUSIONS: The developed technique is fast, robust and comfortable for the patient and for medical personnel. It registers scalp sensor positions with MRI head volume with accuracy that is satisfactory for localization of biological processes examined with a commonly used number of scalp sensors (32, 64, or 128).
Authors: Ziga Spiclin; Arne Hans; Frank H Duffy; Simon K Warfield; Bostjan Likar; Franjo Pernus Journal: Med Image Comput Comput Assist Interv Date: 2008
Authors: Cindy C Hagan; Will Woods; Sam Johnson; Andrew J Calder; Gary G R Green; Andrew W Young Journal: Proc Natl Acad Sci U S A Date: 2009-11-11 Impact factor: 11.205
Authors: Natalie E Adams; Catarina Teige; Giovanna Mollo; Theodoros Karapanagiotidis; Piers L Cornelissen; Jonathan Smallwood; Roger D Traub; Elizabeth Jefferies; Miles A Whittington Journal: J Neurophysiol Date: 2019-01-30 Impact factor: 2.714