Department of Biomedical Engineering, Kyung Hee University, Yongin, Gyeonggi 446-701, Korea.
IEEE Trans Med Imaging. 2009 Nov;28(11):1681-7. doi: 10.1109/TMI.2009.2018112. Epub 2009 Apr 14.
We present the first in vivo cross-sectional conductivity image of the human leg with 1.7 mm pixel size using the magnetic resonance electrical impedance tomography (MREIT) technique. After a review of its experimental protocol by an Institutional Review Board (IRB), we performed MREIT imaging experiments of four human subjects using a 3 T MRI scanner. Adopting thin and flexible carbon-hydrogel electrodes with a large surface area and good contact, we could inject as much as 9 mA current in a form of 15 ms pulse into the leg without producing a painful sensation and motion artifact. Sequentially injecting two imaging currents in two different directions, we collected induced magnetic flux density data inside the leg. Scaled conductivity images reconstructed by using the single-step harmonic B(z) algorithm well distinguished different parts of the subcutaneous adipose tissue, muscle, crural fascia, intermuscular septum and bone inside the leg. We could observe spurious noise spikes in the outer layer of the bone primarily due to the MR signal void phenomenon there. Around the fat, the chemical shift of about two pixels occurred obscuring the boundary of the fat region. Future work should include a fat correction method incorporated in the MREIT pulse sequence and improvements in radio-frequency coils and image reconstruction algorithms. Further human imaging experiments are planned and being conducted to produce conductivity images from different parts of the human body.
我们展示了首例人体腿部 1.7 毫米像素大小的磁共振电阻抗断层成像(MREIT)技术的活体横向传导图像。在经过机构审查委员会(IRB)的实验方案审查后,我们使用 3T MRI 扫描仪对四名人类受试者进行了 MREIT 成像实验。采用具有大表面积和良好接触的薄而灵活的碳水凝胶电极,我们可以将高达 9mA 的电流以 15ms 脉冲的形式注入腿部,而不会产生疼痛感和运动伪影。我们连续向两个不同方向注入两种成像电流,以采集腿部内部的感应磁通密度数据。使用单步谐波 B(z)算法重建的传导图像很好地区分了腿部内部的皮下脂肪组织、肌肉、小腿筋膜、肌肉间隔膜和骨骼的不同部位。我们可以观察到骨骼外层的虚假噪声尖峰,主要是由于那里的磁共振信号缺失现象。在脂肪周围,大约有两个像素的化学位移会使脂肪区域的边界变得模糊。未来的工作应包括在 MREIT 脉冲序列中加入脂肪校正方法,以及改进射频线圈和图像重建算法。计划并正在进行进一步的人体成像实验,以从人体不同部位生成传导图像。