基于太赫兹介电特性的胶质瘤细胞浓度依赖性与病理分级鉴别研究

Study on Concentration Dependence and Pathological Grading Identification of Glioma Cells Based on Terahertz Dielectric Properties

摘要: 目的：研究不同浓度人脑胶质瘤细胞HS683在太赫兹频段(110~1000 GHz)的介电特性，分析细胞浓度对介电常数实部(ε′)、损耗因子(εʺ)的影响，并通过临床石蜡包埋样本验证介电常数在胶质瘤病理分级鉴别中的应用价值，为胶质瘤的诊断、分级及治疗提供生物电磁学基础数据和临床技术支撑。方法：采用矢量网络分析仪测量并计算1.62 × 10⁶、3.25 × 10⁶、6.5 × 10⁶和1.3 × 10⁷ cells/mL四个浓度HS683细胞悬液的介电常数；利用Pearson相关分析研究细胞浓度与介电参数的相关性；增设临床验证实验，收集到5例经病理确诊的人脑胶质瘤石蜡包埋样本，在最优检测频段检测其介电常数实部，对高、低级别胶质瘤样本的介电参数差异进行分析。结果：随着频率增加，所有浓度的HS683细胞介电参数都有所降低，呈现出色散的典型特征。中高频段(501~700 GHz)为细胞浓度检测的最优区间，该频段细胞浓度与介电常数实部相关性最强。临床石蜡样本检测结果显示，高级别与低级别胶质瘤石蜡样本切片的介电常数实部存在显著统计学差异，以介电常数实部为指标可以有效地区分高级别与低级别胶质瘤。结论：HS683细胞的太赫兹介电特性具有明显的浓度依赖性和频率依赖性；介电常数实部可作为胶质瘤病理分级鉴别的潜在特征参数，太赫兹介电特性检测技术在胶质瘤的精准诊断、病理分级及临床治疗中具有重要的生物物理基础价值和临床转化潜力。

Abstract: Objective: To investigate the dielectric properties of human glioma cell line HS683 at different concentrations in the terahertz band (110~1000 GHz), analyze the effects of cell concentration on the real part of permittivity (ε′) and loss factor (εʺ), and verify the application value of permittivity in the pathological grading identification of glioma using clinical paraffin-embedded samples, so as to provide basic bioelectromagnetic data and clinical technical support for the diagnosis, grading and treatment of glioma. Methods The dielectric constants of HS683 cell suspensions at four concentrations (1.62 × 10⁶, 3.25 × 10⁶, 6.5 × 10⁶ and 1.3 × 10⁷ cells/mL) were measured and calculated using a vector network analyzer. Pearson correlation analysis was performed to evaluate the correlation between cell concentration and dielectric parameters. For clinical verification, five pathologically confirmed human glioma paraffin-embedded samples were collected, and their real part of permittivity was detected in the optimal frequency band to analyze the differences in dielectric parameters between high-grade and low-grade glioma samples. Results: The dielectric parameters of HS683 cells at all concentrations decreased with increasing frequency, showing typical dispersion characteristics. The middle-high frequency band (501~700 GHz) was the optimal range for cell concentration detection, in which the correlation between cell concentration and the real part of permittivity was the strongest. The detection results of clinical paraffin samples showed that there were significant statistical differences in the real part of permittivity between high-grade and low-grade glioma paraffin sections, and the real part of permittivity could effectively distinguish high-grade from low-grade glioma. Conclusion: The terahertz dielectric properties of HS683 cells exhibit obvious concentration dependence and frequency dependence. The real part of permittivity can be used as a potential characteristic parameter for the pathological grading identification of glioma. Terahertz dielectric property detection technology has important biophysical basic value and clinical translation potential in the accurate diagnosis, pathological grading and clinical treatment of glioma.