Naruhito Hironaga1, Koichi Hagiwara2, Katsuya Ogata2, Mariko Hayamizu2, Tomokazu Urakawa2, Shozo Tobimatsu2. 1. Department of Clinical Neurophysiology, Neurological Institute, Faculty of Medicine, Graduate School of Medical Sciences, Kyushu University, Fukuoka, Japan. Electronic address: hironaga@med.kyushu-u.ac.jp. 2. Department of Clinical Neurophysiology, Neurological Institute, Faculty of Medicine, Graduate School of Medical Sciences, Kyushu University, Fukuoka, Japan.
Abstract
OBJECTIVE: Co-registration between the head shape extracted from anatomical images that are obtained using a 3D digitizer is a non-negligible factor for magnetoencephalographic (MEG) utilization. The study aimed to propose a novel quick system based on a laser scanning technique involving a 3D laser scanner system that allows instant measurement while maintaining high accuracy and reproducibility. METHODS: The measurement duration, accuracy, and reproducibility of the finger representations in response to tactile stimulation between the 3D laser scanner-based method and the conventional magnetic field digitizer-based method were compared in 11 healthy subjects. Day-to-day variance in target registration error (TRE), day-to-day and session-to-session variability in head position indicator error (HRE) and source localization accuracy were evaluated with visualization of the source estimation and analysis of variance (ANOVA). RESULTS: Our proposed one-snapshot approach enabled 3D digitization measurements in <5s, and significantly reduced TRE, while mean HREs were a comparable level. Although there was less dramatic improvement of source localization, we found a significant reduction in session-to-session variability and day-to-day variance using our proposed method. CONCLUSIONS: Our results clearly demonstrated improvements in speed, comfort, accuracy, and reproducibility when using our new MEG co-registration method. SIGNIFICANCE: A systematic improvement in MEG measurement will be beneficial for routine clinical use.
OBJECTIVE: Co-registration between the head shape extracted from anatomical images that are obtained using a 3D digitizer is a non-negligible factor for magnetoencephalographic (MEG) utilization. The study aimed to propose a novel quick system based on a laser scanning technique involving a 3D laser scanner system that allows instant measurement while maintaining high accuracy and reproducibility. METHODS: The measurement duration, accuracy, and reproducibility of the finger representations in response to tactile stimulation between the 3D laser scanner-based method and the conventional magnetic field digitizer-based method were compared in 11 healthy subjects. Day-to-day variance in target registration error (TRE), day-to-day and session-to-session variability in head position indicator error (HRE) and source localization accuracy were evaluated with visualization of the source estimation and analysis of variance (ANOVA). RESULTS: Our proposed one-snapshot approach enabled 3D digitization measurements in <5s, and significantly reduced TRE, while mean HREs were a comparable level. Although there was less dramatic improvement of source localization, we found a significant reduction in session-to-session variability and day-to-day variance using our proposed method. CONCLUSIONS: Our results clearly demonstrated improvements in speed, comfort, accuracy, and reproducibility when using our new MEG co-registration method. SIGNIFICANCE: A systematic improvement in MEG measurement will be beneficial for routine clinical use.
Authors: Sofie S Meyer; James Bonaiuto; Mark Lim; Holly Rossiter; Sheena Waters; David Bradbury; Sven Bestmann; Matthew Brookes; Martina F Callaghan; Nikolaus Weiskopf; Gareth R Barnes Journal: J Neurosci Methods Date: 2016-11-22 Impact factor: 2.390