Using the Probability Density Function-Based Channel-Combination Bloch-Siegert Method Realizes Permittivity Imaging at 3T

• Published in: Bioengineering (Basel) Vol. 11; no. 7; p. 699
• Main Authors: Wang, Jiajia, Gao, Yunyu, Xin, Sherman Xuegang
• Format: Journal Article
• Language: English
• Published: Switzerland MDPI AG 10-07-2024
• Subjects: Accuracy; Algorithms; B1 mapping technology; Bloch–Siegert method; Electrical properties; electrical properties tomography; Error analysis; Gadolinium; Helmholtz equations; Image reconstruction; Leg; Magnetic properties; Magnetic resonance imaging; Maximum likelihood method; Measurement techniques; Muscles; Noise measurement; Permissible error; Permittivity; Physiology; Probability density functions; relative permittivity; Signal to noise ratio; Spatial distribution; Tomography; Bloch–Siegert method; B1 mapping technology; relative permittivity; electrical properties tomography
• ISSN: 2306-5354; 2306-5354
• DOI: 10.3390/bioengineering11070699

Magnetic resonance electrical properties tomography (MR EPT) can retrieve permittivity from the B1+ magnitude. However, the accuracy of the permittivity measurement using MR EPT is still not ideal due to the low signal-to-noise ratio (SNR) of B1+ magnitude. In this study, the probability density function (PDF)-based channel-combination Bloch-Siegert (BSS) method was firstly introduced to MR EPT for improving the accuracy of the permittivity measurement. MRI experiments were performed using a 3T scanner with an eight-channel receiver coil. The homogeneous water phantom was scanned for assessing the spatial distribution of B1+ magnitude obtained from the PDF-based channel-combination BSS method. Gadolinium (Gd) phantom and rats were scanned for assessing the feasibility of the PDF-based channel-combination BSS method in MR EPT. The Helmholtz-based EPT reconstruction algorithm was selected. For quantitative comparison, the permittivity measured by the open-ended coaxial probe method was considered as the ground-truth value. The accuracy of the permittivity measurement was estimated by the relative error between the reconstructed value and the ground-truth value. The reconstructed relative permittivity of Gd phantom was 52.413, while that of rat leg muscle was 54.053. The ground-truth values of relative permittivity of Gd phantom and rat leg muscle were 78.86 and 49.04, respectively. The relative error of average permittivity was 33.53% for Gd and 10.22% for rat leg muscle. The results indicated the high accuracy of the permittivity measurement using the PDF-based channel-combination BSS method in MR EPT. This improvement may promote the clinical application of MR EPT technology, such as in the early diagnosis of cancers.