分享:
分享到微信朋友圈
X
综述
磁共振FLAIR序列表现阴性的局灶性皮质发育不良病灶检测的研究进展
俱京涛 陈楠

Cite this article as: Ju JT, Chen N. Research progress of focal cortical dysplasia with FLAIR-negative of magnetic resonance imaging[J]. Chin J Magn Reson Imaging, 2022, 13(7): 164-166, 170.本文引用格式:俱京涛, 陈楠. 磁共振FLAIR序列表现阴性的局灶性皮质发育不良病灶检测的研究进展[J]. 磁共振成像, 2022, 13(7): 164-167, 170. DOI:10.12015/issn.1674-8034.2022.07.033.


[摘要] 局灶性皮质发育不良(focal cortical dysplasia, FCD)是导致药物难治性癫痫的常见原因之一,Ⅰ型在FCD中的占比为38.3%,而Ⅱ型占比为61.7%。手术是治疗FCD的有效方式。术前发现病灶并精准定位是决定手术方式及预后的重要因素。目前对于FCD的诊断主要依赖MRI检查,但是,高达40%的Ⅱ型FCD和85%的Ⅰ型FCD病灶在常规MRI上表现为阴性,给诊断和手术带来极大的难度。随着MRI硬件、软件及后处理技术的发展,极大提高了FCD在常规MRI表现为阴性的检出率(综合诊断增益率为31%),对病灶准确定位、指导手术、降低术后癫痫发作具有重要意义。因此本文就提高常规MRI表现为阴性的FCD检出率的方法进行综述。
[Abstract] Focal cortical dysplasia (FCD) is one of the common causes of drug refractory epilepsy. Type Ⅰ accounts for 38.3% of the FCD lesions, while type Ⅱ accounts for 61.7%. Surgery is an effective way for the treatment of FCD. Preoperative detection and accurate localization of the lesions are important factors affecting the mode of operation and prognosis. At present, the diagnosis of FCD mainly depends on MRI. However, up to 40% of type Ⅱ FCD and 85% of type Ⅰ FCD lesions are negative on conventional MRI, which brings great difficulty to diagnosis and operation. With the development of MRI hardware, software and post-processing technology, the negative detection rate of FCD in conventional MRI is greatly improved (overall diagnostic gain rate 31%). Which is great significance for accurate location of lesions, guiding surgery and reducing postoperative seizures. Therefore, this paper reviews the research progress of improving the detection methods of FCD negative on conventional MRI.
[关键词] 磁共振成像;局灶性皮质发育不良;脑磁图;癫痫;三维容积液体衰减反转恢复序列;双反转恢复序列;液体和白质抑制序列;三维边缘增强梯度回波序列;基于体素的形态学分析;形态学分析程序;基于表面形态学技术;卷积神经网络
[Keywords] magnetic resonance imaging;focal cortical dysplasia;magnetoencephalography;epilepsy;three-dimensional fluid attenuated inversion recover;double inversion recovery sequences;fluid and white matter suppression sequence;fluid and white matter suppression sequence;three-dimensional edge-enhancing gradient echo sequence;voxel-based morphometry;morphometric analysis program;surface-based morphometry;convolutional neural network

俱京涛    陈楠 *  

首都医科大学宣武医院放射与核医学科,北京 100053

陈楠,E-mail:chenzen8057@sina.com

作者利益冲突声明:全体作者均声明无利益冲突。


收稿日期:2021-12-12
接受日期:2022-07-05
中图分类号:R445.2  R742.1 
文献标识码:A
DOI: 10.12015/issn.1674-8034.2022.07.033
本文引用格式:俱京涛, 陈楠. 磁共振FLAIR序列表现阴性的局灶性皮质发育不良病灶检测的研究进展[J]. 磁共振成像, 2022, 13(7): 164-167, 170. DOI:10.12015/issn.1674-8034.2022.07.033.

       局灶性皮质发育不良(focal cortical dysplasia, FCD)是药物难治性癫痫最常见原因之一,约占儿童癫痫的40%~50%[1],其中Ⅰ型在FCD中的占比为38.3%,而Ⅱ型占比为61.7%[2]。手术切除病灶是最有效的治疗方法。目前对于FCD的诊断主要依赖于MRI检查,但大约40%的Ⅱ型FCD和85%的Ⅰ型FCD在常规MRI上没有明确的阳性征象[3],常导致漏诊。对于怀疑FCD但MRI上表现为阴性的患者,临床常用脑磁图(magnetoencephalography, MEG)、脑电图(electroencephalogram, EEG)、颅内脑电图(intracranial electroencepholography, IEEG)进行定位并指导手术切除,但这些方法难以精确定位FCD的边界,可能会导致病灶切除不完全或过度切除,从而导致癫痫复发或脑组织功能损伤[4, 5]。因此,术前准确显示常规MRI上表现为阴性病灶的部位、边界及周围组织关系是手术成功的关键。随着MRI硬件、软件及后处理技术的发展极大提高了MRI阴性FCD的检出率(综合诊断增益率为31%)[6],对病灶准确定位、指导手术、降低术后癫痫发作具有重要的作用,因此本文就提高常规MRI表现为阴性的FCD检出率的方法进行综述。

1 超高场强MRI

       FCD的病理基础是灰质结构异常或白质结构内灰质异位,因此提高灰白质对比度最为关键。超高场强MRI具有亚毫米级分辨率及高灰白质对比度,更易发现常规MRI无法显示的FCD病灶。随着超高场强的7.0 T MRI应用于临床,提高了对FCD的检出率。研究表明,相对于1.5 T和3.0 T MRI,7.0 T MRI对FCD的检出率增加了8%~67%[6]。同时,对7.0 T MRI所获取的3D-T1WI图像进行后处理,其灰白质对比更显著,具有更高的病灶检出敏感性[7]。并且在7.0 T时有助于脑组织功能和分子方面的分析[8],Ⅱb型FCD皮层下“黑线征”解释了病理变化带来的解剖细节改变[9]。众所周知,FCD Ⅱa在MRI检查中很难被诊断,然而最近一项研究[10]表明7.0 T MRI有明显优势,50%的FCD Ⅱa的病灶得到明确诊断。

       虽然高场强的不均匀性会给形态测量带来严重的偏差,如7.0 T MRI下,下颞叶及小脑部位形态学测量的误差大于3.0 T MRI。而利用介电垫可以改善这种磁场不均匀性的影响。同时,双反转恢复(double inversion recovery, DIR)序列和液体和白质抑制(fluid and white matter suppression, FLAWS)序列[11]在7.0 T MRI的应用也可以减轻B1场不均匀导致的信号丢失。因此,很多文献建议应用超高场强MRI提高对FCD病灶的检出。

2 基于脑组织某一特异性信号的序列

       近年来,灰白质特异的序列开始应用于FCD的检测中,对提高FCD的病灶检出率、准确定位及确定周围组织关系起到了重要作用。

2.1 三维容积液体衰减反转恢复序列

       液体衰减反转恢复(fluid attenuated inversion recovery, FLAIR)序列属于反转恢复(inversion recovery, IR)序列,其抑制脑脊液信号的翻转恢复脉冲和产生重T2加权的长TE时间可更敏感地检测蛛网膜下腔和脑实质内的病灶,尤其是临近脑组织—脑脊液交界区的病灶,在FCD检测中具有重要价值[12]。三维容积液体衰减反转恢复(3D-FLAIR)序列可以达到各向同性的薄层扫描,具有高信噪比、高分辨率优势。Saini等[13]报道3D-FLAIR有助于提高FCD病灶的检出,在9例常规MRI表现阴性的患者中,有4例只能通过3D-FALIR检测出来。但也有学者[14]认为,与传统的2D-FLAIR比较,其对FCD病灶的检出率并不显著,但是3D-FLAIR的快速扫描及多方位重建,可以代替常规FLAIR扫描。

2.2 DIR

       DIR同时抑制白质和脑脊液信号,突出灰质像素特征,与常规T2、FLAIR序列相比,明显提高组织间的对比度,而清楚显示FCD病灶轮廓,具有较好的敏感性和特异性。研究发现,DIR对常规MRI表现阴性的FCD和微小病变检出率显著提高,其诊断敏感度提高约45%[15]。同时,3D-DIR的敏感度高于3D-FLAIR[16, 17],最近,Beheshti等[18]利用3D-DIR图像和机器学习相结合检测阴性FCD,能达到与氟代脱氧葡萄糖正电子发射型计算机断层显像(18F-fluorodeoxyglucose-positron emission tomography, 18F-FDG-PET)类似的阳性率,这一研究可能在一定程度上解决了FCD对PET的依赖性,从经济方面考虑,这更容易让患者接受。但DIR成像依赖于反转时间(inversion time, TI),灰白质狭窄的TI值限制了其在婴儿或髓鞘发育不完全的大脑中的应用。

2.3 FLAWS序列

       FLAWS序列由Tanner等[19]首次提出,其一次扫描中在两个不同的TI获得两组三维高空间分辨率图像:抑制白质信号图像(fluid and white matter suppression 1, FLAWS1)以及抑制脑脊液信号的图像(fluid and white matter suppression 2, FLAWS2)。然后计算出一组合成的最小体素的灰质图像,同时抑制白质和脑脊液信号,突出了灰质可视化特性,其效果类似DIR,但白质抑制程度更显著。

       因其高三维空间分辨率,对清晰显示皮层下细微结构变化很有意义。对常规MRI表现阴性的FCD的可视化比传统3D-FLAIR更为显著,其敏感度和特异度分别为71.9%和71.1%,且对“Transmantle 征”的显示明显高于其他序列[20]。因其较好的分割性能,基于FLAWS的形态学分析(morphometric analysis program, MAP)更提高了细微FCD的检出率[21],且能减弱高场强B1场不均匀性的影响。在7 T MRI下与DIR相比,FLAWS获得了更均匀的白质信号抑制、更好的灰质可视化、更低的比吸收率(specific absorption ratio, SAR)值[11]。然而FALWS中脑组织对比噪声比(contrast-to-noise ratio, CNR)较低,扫描时间较长。

2.4 三维边缘增强梯度回波序列

       三维边缘增强梯度回波(3D edge-enhancing gradient echo, 3D-EDGE)序列是对反转恢复梯度回波序列进行了调整,优化了TI,使灰质和白质信号强度相当,但极性相反,以达到灰白质交界区信号相互抵消,呈现低信号带,这类似于水-脂界面的勾边效应。

       2020年Middlebrooks等[22]首次将3D-EDGE序列用于FCD检查,该序列依赖于灰白质边界体素中的信号,包含神经元和白质混合物的发育不良体素,其焦点集中在交界区。与磁化准备的快速梯度扫描(magnetization prepared rapid acquisition gradient echo with two inversion times, MP2RAGE)和FLAIR相比,脑灰白质连接处的对比度增加了近6倍,更好地勾勒出了致痫灶的全部范围,使手术精准切除致痫灶成为可能。该作者另一篇文献[23]中肯定了3D-EDGE在深部脑刺激(deep brain stimulation, DBS)中准确定位灰质核团的优势,脑内灰质核团在此序列上边界显示清晰。该序列是一项全新的技术,因其样本量较小,检测效果的确定性仍需研究。

2.5 其他序列

       扩散张量成像(diffusion tensor imaging, DTI)在FCD中的应用是利用水分子扩散反映神经轴突的数量。以此来区分FCD中灰质和白质,这比T1和T2加权更可靠。该序列可以避免常规MRI检测的大脑皮层的“假增厚”或由于异常物质沉积导致信号的相互抵消而出现的假阴性。研究表明,基于体素的DTI和高斯机器学习过程(gaussian processes for machine learning, GPML)的多模态MRI数据比传统的支持向量机(support vector machine, SVM)具有更高的敏感度(76%)[24]。利用球面平均技术(the spherical mean technique, SMT)和神经突定向扩散和密度成像(neurite orientation dispersion and density imaging, NODDI)的测量,有助于表征FCD亚型[25]。与常规检查比较DTI可探及耐药性FCD病灶周围更广泛白质异常[26],这一发现有可能会改善FCD术前评估的标准。目前,DTI对于检测FCD具体特异性未见报道。

       利用磁共振波谱(magnetic resonance spectroscopy, MRS)、磁化转移成像(magnetization transfer imaging, MTI)、磁化转移比(magnetization transfer ratio, MTR)以及动脉自旋标记(arterial spin labeling, ASL)同样能提高FCD的检出率,对检测FCD有较好的敏感性,但是由于其特异性较差,可作为其他检查手段的补充。

3 MEG

       MEG的优势在于无创且其信号不受颅骨、皮肤及其他组织的影响。其高时空分辨率定位更为准确,可以探测到脑电图难以定位的脑深部的磁场变化,如脑沟底部病变、岛叶癫痫,与EEG相比诊断增益率约25%[27]。MRI在髓鞘发育不全大脑中检测FCD很困难,MEG在幼儿耐药性癫痫检测中具有明显的优势[28]。有学者[29]在对1000例大样本癫痫患者回顾性研究中证明MEG在癫痫患者的术前评估、颅内电极放置准确性及预后都有积极的效果。Aydin等[30]将EEG、MEG、MRI联合应用,利用EMEG(EEG和MEG联合)定位、MRI区域放大扫描(层厚0.5 mm)等对微小病变的精确定位很有帮助,有可能减少颅内植入性电极的使用,从而减少对脑组织的额外损伤。由于MEG价格高昂,限制了其普遍推广。

4 图像后处理技术

4.1 基于体素的形态学分析

       基于体素的形态学分析(voxel-based morphometry, VBM)在MR图像上通过相应的标准对脑组织进行分割,以Z值反映组间(连接图、厚度图、扩展图)体素的差异,从而量化脑组织形态学异常。该技术对于视觉上细微的FCD病变具有较高的准确性。Wang等[31]对150例病例回顾性研究发现VBM的阳性率为43%、敏感度为90%、特异度为67%,且MAP+患者预后较好。Chen等[32]研究发现定量体积的VBM阳性率为87.5%、敏感度和特异度分别为93.9%和79.6%,并指出致痫灶同样存在于皮层萎缩区。VBM与PET-MRI联合应用可以明显提高扣带回微小病变的检出率[33]。Martin等[34]对144名患者进行4种VBM,在不同的检验水准下,除了基于归一化的T2-FLAIR的形态学测量(normalized fluid-attenuated inversion recovery-based VBM, nFSI-VBM)外,基于T1灰质体积的形态学测量(gray matter volume-based VBM, GMV-VBM)、基于T1灰质浓度的形态学测量(gray matter concentration-based VBM, GMC-VBM)、基于T1连接图的形态学测量(junction map-based VBM, JM-VBM)均具有较高的假阳性率。因此VBM还需匹配EEG和PET,共同提高检测的特异性。VBM对图像的伪影比较敏感,会导致假阳性或假阴性率的增加。

4.2 形态学分析程序

       形态学分析程序(morphometric analysis program, MAP)是基于SPM和MATLAB软件,以高分辨率MRI图像为基础,通过配准、分割、计算机模型分别生成MAP “扩展”图和“连接”图,反映灰白质交界区信号差异。MAP检测阴性FCD Ⅱ型病变敏感度与PET、MEG和单光子发射CT相当,明显高于传统视觉分析[35]。利用成人数据库对幼儿癫痫诊断,对难以测化的阴性扣带回致痫灶及对微小病灶有较高的敏感性[36]。最近一项研究利用人工神经网络(artificial neural network, ANN)分类器扩展MAP18程序,实现了FCD检测的完全自动化,敏感度为84.0%,特异度为84.3%,而且不受扫描位置、机型及扫描参数的限制[37]。但MAP不能清晰显示病灶的边界。

4.3 基于表面形态学技术

       基于表面形态学技术(surface-based morphometry, SBM)通过构建MRI图像脑表面各种参数(包括皮层厚度、曲率、脑沟深度以及折叠指数等)并利用Freesurfer软件计算构建灰质内、外表面的特征来表征FCD。由于参数较多,SBM是检测阴性FCD较准确且理想的方法。而将表面形态的互补指标(称为“doughnut”方法)融入该技术,其诊断敏感度由59%提升至73%[38]。配备了机器学习的自动分析的SBM在耐药性FCD Ⅱ型病变的检测中敏感度(73.7%)和特异度(90.0%)也较高[39]。但是SBM计算烦琐,工作量大。Ganji等[40]在机器学习的自动分析SBM中加入了ANN使FCD Ⅱ型病变的检测的敏感度、特异度及准确度分别提高到96.7%、100%和98.6%,即使在MRI阴性的患者中准确度也能达到91.3%。

4.4 卷积神经网络

       卷积神经网络(convolutional neural network, CNN)将大脑分割成多个模块,通过机器学习建立病变脑组织与正常脑组织特征模型,然后在多个模块中通过一一比对输出符合病变特征的脑组织区域,该技术具有较高的检测准确性及分割精度,其对FCD检出的敏感度无论是基于T1图像的FCD补丁技术(90%)[41]还是基于FLAIR图像的激活最大化和卷积定位技术(83%)[42]均较高,与VBM、SBM进行比较,其对FCD检出的特异度和精确度分别为85%和88%,而且避免了手动特征提取带来的偏差。Thomas等[43]通过加入多分辨率的注意门控模块,基于混合跳过连接所提出的模块的稳健性解决了线性基线架构存在的缺陷,而且利用较少的参数就可以得到与上述同等的效果,提高了临床的实用性。

5 小结

       综上所述,随着MRI场强的不断提升、新序列及各种后处理技术的发展,对常规MRI表现阴性的FCD的检出率有了显著提高,对疾病的早期诊断、准确定位、手术指导及预后判断具有重要意义。然而各种新序列及新技术诊断效能还需要大样本的研究及长时间随访进行确认。寻求一种合适的组合扫描方案和后处理技术以提高FCD的诊断敏感性及特异性,将是未来着重研究的方向。

[1]
Xue H, Cai LX, Dong S, et al. Clinical characteristics and post-surgical outcomes of focal cortical dysplasia subtypes[J/OL]. J Clin Neurosci, 2016 [2022-1-26]. https://www.jocn-journal.com/article/S0967-5868(15)00360-4/fulltext. DOI: 10.1016/j.jocn.2015.04.022.
[2]
Yao K, Duan ZJ, Zhou J, et al. Surgical histopathologic findings of 232 Chinese children cases with drug-resistant seizures[J/OL]. Brain Behav, 2020 [2022-1-26]. https://onlinelibrary.wiley.com/doi/10.1002/brb3.1565. DOI: 10.1002/brb3.1565.
[3]
Blumcke I, Spreafico R, Haaker G, et al. Histopathological Findings in Brain Tissue Obtained during Epilepsy Surgery[J]. N Engl J Med, 2017, 377(17): 1648-1656. DOI: 10.1056/NEJMoa1703784.
[4]
Veersema TJ, Swampillai B, Ferrier CH, et al. Long-term seizure outcome after epilepsy surgery in patients with mild malformation of cortical development and focal cortical dysplasia[J]. Epilepsia Open, 2019, 4(1): 170-175. DOI: 10.1002/epi4.12289.
[5]
Wagstyl K, Whitake K, Raznahan A, et al. Atlas of lesion locations and postsurgical seizure freedom in focal cortical dysplasia: A MELD study[J]. Epilepsia, 2022, 63(1): 61-74. DOI: 10.1111/epi.17130.
[6]
Van Lanen RHGJ, Colon AJ, Wiggins CJ, et al. Ultra-high field magnetic resonance imaging in human epilepsy: A systematic review[J/OL]. Neuroimage Clin, 2021 [2022-1-26]. https://www.sciencedirect.com/science/article/pii/S2213158221000462. DOI: 10.1016/j.nicl.2021.102602.
[7]
Chen C, Xie JJ, Ding F, et al. 7T MRI with post-processing for the presurgical evaluation of pharmacoresistant focal epilepsy[J/OL]. Ther Adv Neurol Disord, 2021 [2022-1-26]. https://journals.sagepub.com/doi/full/10.1177/17562864211021181. DOI: 10.1177/17562864211021181.
[8]
Opheim G, van der Kolk A, Markenroth Bloch K, et al. 7T Epilepsy Task Force Consensus Recommendations on the Use of 7T MRI in Clinical Practice[J]. Neurology, 2021, 96(7): 327-341. DOI: 10.1212/WNL.0000000000011413.
[9]
Bartolini E, Cosottini M, Costagli M, et al. Ultra-High-Field Targeted Imaging of Focal Cortical Dysplasia: The Intracortical Black Line Sign in Type IIb[J]. AJNR Am J Neuroradiol, 2019, 40(12): 2137-2142. DOI: 10.3174/ajnr.A6298.
[10]
Liu T, Liang H, Cui J, et al. Clinical Application of 7T Magnetic Resonance Imaging in Patients with Focal Cortical Dysplasia IIa and Epilepsy[J]. Stereotact Funct Neurosurg, 2021, 99(3): 212-220. DOI: 10.1159/000511152.
[11]
Urushibata Y, Kuribayashi H, Fujimoto K, et al. Advantages of fluid and white matter suppression (FLAWS) with MP2RAGE compared with double inversion recovery turbo spin echo (DIR-TSE) at 7T[J/OL]. Eur J Radiol, 2019 [2022-1-26]. https://www.ejradiology.com/article/S0720-048X(19)30153-6/fulltext. DOI: 10.1016/j.ejrad.2019.04.019.
[12]
Kokkinos V, Kallifatidis A, Kapsalaki EZ, et al. Thin isotropic FLAIR MR images at 1.5T increase the yield of focal cortical dysplasia transmantle sign detection in frontal lobe epilepsy[J/OL]. Epilepsy Res, 2017 [2022-1-26]. https://www.sciencedirect.com/science/article/abs/pii/S0920121116303710?via%3Dihub. DOI: 10.1016/j.eplepsyres.2017.02.018.
[13]
Saini J, Singh A, Kesavadas C, et al. Role of three-dimensional fluid-attenuated inversion recovery (3D FLAIR) and proton density magnetic resonance imaging for the detection and evaluation of lesion extent of focal cortical dysplasia in patients with refractory epilepsy[J]. Acta Radiol, 2010, 51(2): 218-225. DOI: 10.3109/02841850903433805.
[14]
Tschampa HJ, Urbach H, Malter M, et al. Magnetic resonance imaging of focal cortical dysplasia: Comparison of 3D and 2D fluid attenuated inversion recovery sequences at 3T[J/OL]. Epilepsy Res, 2015 [2022-1-26]. https://www.sciencedirect.com/science/article/abs/pii/S0920121115300243?via%3Dihub. DOI: 10.1016/j.eplepsyres.2015.07.004.
[15]
Rugg-Gunn FJ, Boulby PA, Symms MR, et al. Imaging the neocortex in epilepsy with double inversion recovery imaging[J]. Neuroimage, 2006, 31(1): 39-50. DOI: 10.1016/j.neuroimage.2005.11.034.
[16]
Sone D, Sato N, Kimura Y, et al. Quantitative analysis of double inversion recovery and FLAIR signals in temporal lobe epilepsy[J/OL]. Epilepsy Res, 2021 [2022-1-26]. https://www.sciencedirect.com/science/article/abs/pii/S092012112030591X. DOI: 10.1016/j.eplepsyres.2020.106540.
[17]
Park CC, Thongkham DW, Sadigh G, et al. Detection of Cortical and Deep Gray Matter Lesions in Multiple Sclerosis Using DIR and FLAIR at 3T[J]. J Neuroimaging, 2021, 31(2): 408-414. DOI: 10.1111/jon.12822.
[18]
Beheshti I, Sone D, Maikusa N, et al. Accurate lateralization and classification of MRI-negative 18F-FDG-PET-positive temporal lobe epilepsy using double inversion recovery and machine-learning[J/OL]. Comput Biol Med, 2021 [2022-1-26]. https://www.sciencedirect.com/science/article/abs/pii/S0010482521005990. DOI: 10.1016/j.compbiomed.2021.104805.
[19]
Tanner M, Gambarota G, Kober T, et al. Fluid and white matter suppression with the MP2RAGE sequence[J]. J Magn Reson Imaging, 2012, 35(5): 1063-1070. DOI: 10.1002/JMRI.23532.
[20]
Chen X, Qian T, Kober T, et al. Gray-matter-specific MR imaging improves the detection of epileptogenic zones in focal cortical dysplasia: A new sequence called fluid and white matter suppression (FLAWS)[J/OL]. Neuroimage Clin, 2018 [2022-1-26]. https://www.sciencedirect.com/science/article/pii/S2213158218302511?via%3Dihub. DOI: 10.1016/j.nicl.2018.08.010.
[21]
Sun K, Yu T, Yang D, et al. Fluid and White Matter Suppression Imaging and Voxel-Based Morphometric Analysis in Conventional Magnetic Resonance Imaging-Negative Epilepsy[J/OL]. Front Neurol, 2021 [2022-1-26]. https://www.frontiersin.org/articles/10.3389/fneur.2021.651592/full. DOI: 10.3389/fneur.2021.651592.
[22]
Middlebrooks EH, Lin C, Westerhold E, et al. Improved detection of focal cortical dysplasia using a novel 3D imaging sequence: Edge-Enhancing Gradient Echo (3D-EDGE) MRI[J/OL]. Neuroimage Clin, 2020 [2022-1-26]. https://www.sciencedirect.com/science/article/pii/S2213158220302862. DOI: 10.1016/j.nicl.2020.102449.
[23]
Middlebrooks EH, Okromelidze L, Lin C, et al. Edge-enhancing gradient echo with multi-image co-registration and averaging (EDGE-MICRA) for targeting thalamic centromedian and parafascicular nuclei[J]. Neuroradiol J, 2021, 34(6): 667-675. DOI: 10.1177/19714009211021781.
[24]
Wang Y, Zhou Y, Wang H, et al. Voxel-based automated detection of focal cortical dysplasia lesions using diffusion tensor imaging and T2-weighted MRI data[J/OL]. Epilepsy Behav, 2018 [2022-1-26]. https://www.epilepsybehavior.com/article/S1525-5050(18)30084-2/fulltext. DOI: 10.1016/j.yebeh.2018.04.005.
[25]
Lorio S, Adler S, Gunny R, et al. MRI profiling of focal cortical dysplasia using multi-compartment diffusion models[J]. Epilepsia, 2020, 61(3): 433-444. DOI: 10.1111/epi.16451.
[26]
Jin B, Lv Z, Chen W, et al. Perilesional white matter integrity in drug-resistant epilepsy related to focal cortical dysplasia[J/OL]. Seizure, 2021 [2022-1-26]. https://www.seizure-journal.com/article/S1059-1311(21)00258-2/fulltext. DOI: 10.1016/j.seizure.2021.07.027.
[27]
Duez L, Tankisi H, Hansen PO, et al. Electromagnetic source imaging in presurgical workup of patients with epilepsy: A prospective study[J/OL]. Neurology, 2019 [2022-1-26]. https://n.neurology.org/content/92/6/e576. DOI: 10.1212/WNL.0000000000006877.
[28]
Otsubo H, Ogawa H, Pang E, et al. A review of magnetoencephalography use in pediatric epilepsy: an update on best practice[J]. Expert Rev Neurother, 2021, 21(11): 1225-1240. DOI: 10.1080/14737175.2021.1910024.
[29]
Rampp S, Stefan H, Wu X, et al. Magnetoencephalography for epileptic focus localization in a series of 1000 cases[J]. Brain, 2019, 142(10): 3059-3071. DOI: 10.1093/brain/awz231.
[30]
Aydin Ü, Rampp S, Wollbrink A, et al. Zoomed MRI Guided by Combined EEG/MEG Source Analysis: A Multimodal Approach for Optimizing Presurgical Epilepsy Work-up and its Application in a Multi-focal Epilepsy Patient Case Study[J]. Brain Topogr, 2017, 30(4): 417-433. DOI: 10.1007/s10548-017-0568-9.
[31]
Wang ZI, Jones SE, Jaisani Z, et al. Voxel-based morphometric magnetic resonance imaging (MRI) postprocessing in MRI-negative epilepsies[J]. Ann Neurol, 2015, 77(6): 1060-1075. DOI: 10.1002/ana.24407.
[32]
Chen X, Qian T, Maréchal B, et al. Quantitative volume-based morphometry in focal cortical dysplasia: A pilot study for lesion localization at the individual level[J/OL]. Eur J Radiol, 2018 [2022-1-26]. https://www.ejradiology.com/article/S0720-048X(18)30224-9/fulltext. DOI: 10.1016/j.ejrad.2018.06.019.
[33]
Sun K, Ren Z, Yang D, et al. Voxel-based morphometric MRI post-processing and PET/MRI co-registration reveal subtle abnormalities in cingulate epilepsy[J/OL]. Epilepsy Res, 2021 [2022-1-26]. https://www.sciencedirect.com/science/article/pii/S0920121121000218?via%3Dihub.
[34]
Martin P, Winston GP, Bartlett P, et al. Voxel-based magnetic resonance image postprocessing in epilepsy[J]. Epilepsia, 2017, 58(9): 1653-1664. DOI: 10.1111/epi.13851.
[35]
Wagner J, Weber B, Urbach H, et al. Morphometric MRI analysis improves detection of focal cortical dysplasia type II[J]. Brain, 2011, 134(Pt 10): 2844-2854. DOI: 10.1093/brain/awr204.
[36]
Stecher X, Schonstedt V, Manterola C, et al. Morphometric analysis program: Detection of epileptic foci in young children using an adult normative database: Initial experience[J]. Epilepsia Open, 2021, 6(1): 235-238. DOI: 10.1002/epi4.12456.
[37]
David B, Kröll-Seger J, Schuch F, et al. External validation of automated focal cortical dysplasia detection using morphometric analysis[J]. Epilepsia, 2021, 62(4): 1005-1021. DOI: 10.1111/epi.16853.
[38]
Adler S, Wagstyl K, Gunny R, et al. Novel surface features for automated detection of focal cortical dysplasias in paediatric epilepsy[J/OL]. Neuroimage Clin, 2016 [2022-1-26]. https://www.sciencedirect.com/science/article/pii/S2213158216302674?via%3Dihub. DOI: 10.1016/j.nicl.2016.12.030.
[39]
Jin B, Krishnan B, Adler S, et al. Automated detection of focal cortical dysplasia type Ⅱ with surface-based magnetic resonance imaging postprocessing and machine learning[J]. Epilepsia, 2018, 59(5): 982-992. DOI: 10.1111/epi.14064.
[40]
Ganji Z, Hakak MA, Zamanpour SA, et al. Automatic Detection of Focal Cortical Dysplasia Type Ⅱ in MRI: Is the Application of Surface-Based Morphometry and Machine Learning Promising?[J/OL]. Front Hum Neurosci, 2021 [2022-1-26]. https://www.frontiersin.org/articles/10.3389/fnhum.2021.608285/full. DOI: 10.3389/fnhum.2021.608285.
[41]
Wang H, Ahmed SN, Mandal M. Automated detection of focal cortical dysplasia using a deep convolutional neural network[J/OL]. Comput Med Imaging Graph, 2020 [2022-1-26]. https://www.sciencedirect.com/science/article/abs/pii/S0895611118304610. DOI: 10.1016/j.compmedimag.2019.101662.
[42]
Feng C, Zhao H, Li Y, et al. Automatic localization and segmentation of focal cortical dysplasia in FLAIR-negative patients using a convolutional neural network[J]. J Appl Clin Med Phys, 2020, 21(9): 215-226. DOI: 10.1002/acm2.12985.
[43]
Thomas E, Pawan SJ, Kumar S, et al. Multi-Res-Attention UNet: A CNN Model for the Segmentation of Focal Cortical Dysplasia Lesions from Magnetic Resonance Images[J]. IEEE J Biomed Health Inform, 2021, 25(5): 1724-1734. DOI: 10.1109/JBHI.2020.3024188.

上一篇 基于磁共振成像的人工智能技术在垂体腺瘤中的应用研究现状
下一篇 磁共振T2定量成像技术在肿瘤评价中的应用进展
  
诚聘英才 | 广告合作 | 免责声明 | 版权声明
联系电话:010-67113815
京ICP备19028836号-2