دانلود رایگان ترجمه مقاله نقشه برداری قبل از جراحی مغز در زمان فلج مغزی – اسپرینگر ۲۰۱۵

نقشه برداری قبل از جراحی مغز در زمان فلج مغزی با استفاده همزمان از EEG و MRI کاربردی با بزرگ نمایی بالا: امکان سنجی و نخستین نتایج

Presurgical brain mapping in epilepsy using simultaneous EEG and functional MRI at ultra‑high field: feasibility and first results

Abstract

Objectives The aim of this study was to demonstrate that eloquent cortex and epileptic-related hemodynamic changes can be safely and reliably detected using simultaneous electroencephalography (EEG)–functional magnetic resonance imaging (fMRI) recordings at ultra-high field (UHF) for clinical evaluation of patients with epilepsy. Materials and methods Simultaneous EEG–fMRI was acquired at 7 T using an optimized setup in nine patients with lesional epilepsy. According to the localization of the lesion, mapping of eloquent cortex (language and motor) was also performed in two patients. Results Despite strong artifacts, efficient correction of intra-MRI EEG could be achieved with optimized artifact removal algorithms, allowing robust identification of interictal epileptiform discharges. Noise-sensitive topographyrelated analyses and electrical source localization were also performed successfully. Localization of epilepsyrelated hemodynamic changes compatible with the lesion were detected in three patients and concordant with findings obtained at 3 T. Local loss of signal in specific regions, essentially due to B1 inhomogeneities were found to depend on the geometric arrangement of EEG leads over the cap. Conclusion These results demonstrate that presurgical mapping of epileptic networks and eloquent cortex is both safe and feasible at UHF, with the benefits of greater spatial resolution and higher blood-oxygenation-level-dependent sensitivity compared with the more traditional field strength of 3 T.

۱ Introduction

Functional magnetic resonance imaging (fMRI) is a noninvasive technique capable of detecting hemodynamic changes related to functional brain activity. The most widely used acquisition methods rely on the blood-oxygenation-level-dependent (BOLD) effect, which arises due to a local modification of the magnetic susceptibility caused by the paramagnetic properties of deoxyhemoglobin [1]. The ability to acquire fMRI at ultra-high field (UHF) offers the opportunity to greatly enhance BOLD contrast sensitivity and to subsequently improve the spatial resolution or decrease the number of events required to observe a significant effect [2, 3]. Furthermore, the intravascular signal contribution from draining veins decreases with magnetic field strength [4], allowing a more accurate localization. These benefits open the possibility to better characterize epileptic networks using simultaneous electroencephalography (EEG) and to enhance our understanding of negative BOLD responses [5]. EEG–fMRI at UHF would permit more precise identification of epileptogenic areas and functional vital cortex during presurgical evaluations. Better EEG–fMRI sensitivity may be beneficial to patients with few interictal epileptiform discharges (IED) or those with mitigated results at 3 T. However, the acquisition of EEG during fMRI, especially at UHF, suffers from various artifacts that compromise data quality. First, gradient artifacts completely obscure the EEG during fMRI acquisition. The rapid switching of magnetic field gradients produces electromagnetic induction into the loops formed by the wires of the EEG, the amplitude of which depends on gradient slew rates [6]. If EEG acquisition is synchronized with the MR clock, gradient artifacts are strictly periodic and reproducible and can easily be removed using moving average artifact template subtraction [7, 8]. Second, any motion or vibration of electrodes and leads in the static magnetic field produce induced electromotive forces in EEG that are proportional to the field strength [9]. In particular, pulse artifacts (PA), which are due to a combination of nodding head motion following each heartbeat, scalp expansion inducing motion of the electrodes near superficial arteries, and Hall effect produced by blood flow [10, 11], strongly contaminate the EEG. Contrary to gradient artifacts, PA are nonstationary and highly variable between successive heartbeats. The use of averaged artifact subtraction (AAS) is not sufficient to accurately remove PA, especially at higher fields [8, 12, 13], where both amplitude and variability greatly increase with magnetic field strength [14]. A second step using independent component analysis (ICA) [15] or optimal basis set (OBS) [16] is generally used to remove residuals. At UHF, vibrations induced by the helium pump [17] or by the ventilation system [18] could also seriously affect EEG quality, but they can be limited by using an optimized setup with ultrashort bundled wires [19]. Spontaneous head motion may also affect EEG recordings at higher field and compromise subsequent gradient and PA correction using moving average template subtraction. The fMRI realignment parameters can be used to improve gradient artifact removal [20], and piezoelectric sensors or additional loops on the EEG caps can be used to filter out motion artifacts [9, 21, 22]. Furthermore, the presence of EEG electrodes and conductive wires can alter the homogeneity of the static magnetic field (B0) and disrupt the radiofrequency (RF) field (B1), leading to local signal dropout and distortion in the vicinity of EEG electrodes and to a global decrease of signal-to-noise ratio (SNR) [23]. These artifacts mainly depend on field strength and can significantly impact image quality at UHF. However, given that physiological noise is also reduced, temporal SNR of functional images is less affected, even in the presence of a dense-array EEG cap [24]. In addition to substantial artifacts, simultaneous recordings of EEG and fMRI at UHF raise important concerns regarding patient safety. First, the presence of EEG materials may alter the transmit B1 field distribution, resulting in unpredicted local specific absorption rate (SAR) modification [25]. Second, the radiofrequency pulse wavelength decreases with static magnetic field intensity, increasing the risk of resonant antenna effects in the wires, especially at 7 T [26, 27]. In this report, preliminary results obtained in nine patients with epilepsy using simultaneous EEG–fMRI at 7 T are presented. Our aim was to demonstrate that using an optimized setup and appropriate artifacts removal algorithms, eloquent cortex and epileptic-related hemodynamic changes can be safely and reliably detected at UHF for clinical presurgical evaluation of epileptic patients. To our knowledge, this is the first report of simultaneous EEG– fMRI acquisition at UHF in patients with epilepsy