基于磁场强度双变分的磁共振电特性成像方法研究

The magnetic resonance electrical properties tomography(MREPT)combines the advantages of high contrast for electrical impedance tomography(EIT)and high spatial resolution and imaging quality for magnetic resonance imaging(MRI),which requires no electrodes and also without external energy injected imaging body. On the biological tissue, MREPT is a real non-invasive and non-contact measurement technology. At present，there are mainly two kinds of imaging algorithms for MREPT. One is iterative algorithms which have time-consuming iterative proves and poor astringency. Another kind is non iterative algorithm that base on the RF field imaging technology and assume the longitudinal magnetic field strength as zero. But this algorithm has large reconstruction error of electric parameters. In view of this, this project proposes a non-iterative magnetic resonance electrical property tomography based on dual variational equation of RF field magnetic field strength. The new method of using high strength and ultra high field MRI system, break out the non reasonable assumption that the longitudinal magnetic field strength is zero, and accurate reconstruct the biological tissue conductivity, dielectric constant and RF electromagnetic field. At the same time, according to the RF electromagnetic field, the SAR is calculated accurately. Through simulations and experiments configured with different RF coil mode and with biological tissues of different electromagnetic parameters added in, pattern of the influence of the B1 field uniformity on the distribution of electric parameters, contribution of the radial magnetic intensity to the reconstruction of the electrical parameters and pattern of the distribution of the value of SAR are investigated. The robustness of the reconstruction algorithm to noise is studied with different noise levels, giving the technique a further progress towards real application. New imaging method with high resolution and high contrast -- which provides a powerful analysis tool and platform for early diagnosis of cancer -- is derived with independent intellectual property rights.

磁共振电特性成像结合了电阻抗成像高对比度以及磁共振成像高空间分辨率和高成像质量的优势，无需电极也无需外加能量注入成像体，是一种真正的非侵入性和无损伤测量方法。目前磁共振电特性成像算法主要有两种，一种是非线性迭代类算法，迭代过程耗时，而且收敛性有待研究，另一种则是基于射频场成像技术，假定纵向磁场强度为零的求解方法，重建的电特性误差大。针对此本项目提出一种基于射频场磁场强度双变分的磁共振电特性成像方法，新方法利用高场强和超高场强MRI系统，突破目前射频场纵向磁场强度为零的不合理假设，准确重建生物组织的电导率、介电常数和SAR。基于不同射频线圈激励考察射频场均匀性、纵向磁场强度对电特性参数分布和SAR值的影响规律，研究不同噪声情况下，重建算法的抗噪性，使该技术进一步向应用推进，形成具有自主知识产权的新型高分辨率、高对比度、高精度的磁共振电特性成像方法，为癌症的早期诊断提供一个有力的分析工具和平台

磁共振电特性成像结合了电阻抗成像高对比度以及磁共振成像高空间分辨率和高成像 质量的优势，无需电极也无需外加能量注入成像体，是一种真正的非侵入性和无损伤测量 方法。目前磁共振电特性成像算法主要有两种，一种是非线性迭代类算法，迭代过程耗时 ，而且收敛性有待研究，另一种则是基于射频场成像技术，假定纵向磁场强度为零的求解 方法，重建的电特性误差大。针对此本项目提出一种基于射频场磁场强度双变分的磁共振 电特性成像方法，新方法利用高场强和超高场强MRI系统，突破目前射频场纵向磁场强度 为零的不合理假设，准确重建生物组织的电导率、介电常数和SAR。基于不同射频线圈激 励考察射频场均匀性、纵向磁场强度对电特性参数分布和SAR值的影响规律，研究不同噪 声情况下，重建算法的抗噪性，使该技术进一步向应用推进，形成具有自主知识产权的新 型高分辨率、高对比度、高精度的磁共振电特性成像方法，为癌症的早期诊断提供一个有 力的分析工具和平台。