Volume 50 Issue 8
Aug.  2021
Turn off MathJax
Article Contents

Hao Yufan, Feng Zhengyun, Han Chao, Wang Zeyun, Wang Qingfang, Wang Liping, Shi Chenjun, Wu Xu, Peng Yan. Application of high sensitive detection sensor chip in detection of brain glioma disease[J]. Infrared and Laser Engineering, 2021, 50(8): 20210279. doi: 10.3788/IRLA20210279
Citation: Hao Yufan, Feng Zhengyun, Han Chao, Wang Zeyun, Wang Qingfang, Wang Liping, Shi Chenjun, Wu Xu, Peng Yan. Application of high sensitive detection sensor chip in detection of brain glioma disease[J]. Infrared and Laser Engineering, 2021, 50(8): 20210279. doi: 10.3788/IRLA20210279

Application of high sensitive detection sensor chip in detection of brain glioma disease

doi: 10.3788/IRLA20210279
  • Received Date: 2021-04-29
  • Rev Recd Date: 2021-05-06
  • Publish Date: 2021-08-25
  • Terahertz wave can be applied to the rapid qualitative and quantitative identification of substances because of its characteristics of fingerprint identification and non-destructive detection. At present, the detection limit of terahertz technology is in the order of milligram. However, the concentration of the tested substance in the actual biomedical samples is usually in the order of microgram or even below, which limits the detection sensitivity and feasibility. In this study, by taking the glioma biomarkers: inositol (MI) and gamma aminobutyric acid (GABA) as an example, based on the inductive-capacitive (LC) resonance, a metamaterial chip was designed to enhance THz detection sensitivity. Then, the terahertz spectra of the chip covered by MI and GABA at different concentrations was tested to prove that their resonance frequency shifts show different rules with the change of concentration, basing on which the qualitative identification could be achieved. And for known samples of MI and GABA, quantitative analysis could be achieved according to the frequency shifts law. According to calculations based on experimental data, the lower detection limits of proposed chip for these two samples are 3.457 µg and 2.552 µg, respectively, which are three orders of magnitude higher than the detection limit of the conventional tableting method. These results have important reference value for the qualitative and quantitative detection of trace specific substances of diseases in later biomedicine.
  • [1] Li Can, Wang Xin, Zhai Shuang. Detection of diseased brain based on medical image [J]. Journal of Changchun University of Technology, 2020, 41(2): 162-167. (in Chinese) doi:  10.15923/j.cnki.cn22-1382/t.2020.2.10
    [2] Wang Y Y, Wang L P, Li T, et al. Terahertz characteristic absorption spectrum analysis of homocysteine [J]. Acta Optica Sinica, 2019, 39(10): 1030003. (in Chinese) doi:  10.3788/AOS201939.1030003
    [3] Xie Q, Yang H R, Li H G, et al. Explosives identification based on terahertz time domain spectroscopy [J]. Optical Precision Engineering, 2016, 24(10): 2392-2399. (in Chinese) doi:  10.3788/OPE.20162410.2392
    [4] Wang S F, Wang Q C, Peng Y. Mechanism study of terahertz radiation regulation in multi-color laser field [J]. Journal of the Optical Society of America B, 2020, 37(11): 3325-3334. (in Chinese) doi:  10.1364/JOSAB.399626
    [5] Yan Jun, Wang Liping, Li Tian, et al. Quantitative identification of homocysteine in liquid by terahertz technology [J]. Infrared and Laser Engineering, 2019, 48(8): 0819001. (in Chinese) doi:  10.3788/IRLA201948.0819001
    [6] Zhang J F, Yuan X D, Qin S Q. Tunable terahertz and optical metamaterials [J]. Chinese Journal of Optics, 2014, 7(3): 349-364. (in Chinese) doi:  10.3788/CO.20140703.0349
    [7] Zhu Y M, Shi C J, Wu X, et al. Research on algorithm of terahertz spectroscopy in biomedical detection [J]. Acta Optica Sinica, 2020, 41(1): 0130001. (in Chinese)
    [8] Gao Xiang, Liu Xiaoqing, Dai Zijie, et al. Integrated terahertz confocal imaging system based on waveguide structure [J]. Infrared and Laser Engineering, 2019, 48(S2): S219001. (in Chinese) doi:  10.3788/IRLA201948.S219001
    [9] Peng Y, Yuan X R, Zou X, et al. Terahertz identification and quantification of neurotransmitter and neurotrophy mixture [J]. Biomedical Optics Express, 2016, 7(11): 4472-4479. (in Chinese) doi:  10.1364/BOE.7.004472
    [10] Bianch L, Micheli E D, Bricolo A, et al. Extracellular levels of amino acids and choline in human high grade gliomas: An intraoperative microdialysis study [J]. Neurochemical Research, 2004, 29(1): 325-334. doi:  10.1023/B:NERE.0000010462.72557.6d
    [11] Choi I Y, Lee S P, Merkle H, et al. In vivo detection of gray and white matter differences in GABA concentration in the human brain [J]. Neurolmage, 2006, 33(1): 85-93. doi:  10.1016/j.neuroimage.2006.06.016
    [12] Gu H Y, Shi C J, Wu X, et al. Molecular methylation detection based on terahertz metamaterial technology [J]. Analyst, 2020, 145(20): 6705-6712. doi:  10.1039/D0AN01062F
    [13] Wang L P, Wu X, Peng Y, et al. Quantitative analysis of homocysteine in liquid by terahertz spectroscopy [J]. Biomedical Optics Express, 2020, 11(5): 2570-2577. (in Chinese) doi:  10.1364/BOE.391894
    [14] Pan X C, Yao Z H, Xu X L, et al. Fabrication, design and application of THz metamaterials [J]. Chinese Optics, 2013, 6(3): 283-296. (in Chinese)
    [15] Guo L H, Wang X K, Zhang Y. Terahertz digital holographic imaging of biological tissue [J]. Optical Precision Engineering, 2017, 25(3): 611-615. (in Chinese) doi:  10.3788/OPE.20172503.0611
    [16] Luo C W, Zhao Z Y, Song Z Q, et al. Terahertz extraodinary transmission from flexible C-shape split-ring resnoators [J]. Journal of East China Normal University (Natural Science), 2016, 6(1): 0107-06. (in Chinese)
    [17] Singh R, Cao W, Al-Naib I, et al. Ultrasensitive terahertz sensing with high-Q Fano resonances in metasurfaces [J]. Applied Physics Letters, 2014, 105(170): 171101.
    [18] Peng Y, Shi C J, Zhu Y M, et al. Qualitative and quantitative analysis algorithm of terahertz apectroscopy in biomedical detection [J]. Chinese Journal of Lasers, 2019, 46(6): 0614002. (in Chinese) doi:  10.3788/CJL201946.0614002
  • 加载中
通讯作者: 陈斌, bchen63@163.com
  • 1. 

    沈阳化工大学材料科学与工程学院 沈阳 110142

  1. 本站搜索
  2. 百度学术搜索
  3. 万方数据库搜索
  4. CNKI搜索

Figures(5)

Article Metrics

Article views(353) PDF downloads(32) Cited by()

Related
Proportional views

Application of high sensitive detection sensor chip in detection of brain glioma disease

doi: 10.3788/IRLA20210279
  • College of Optical-Electrical and Computer Engineering, University of Shanghai for Science and Technology, Shanghai 200093, China

Abstract: Terahertz wave can be applied to the rapid qualitative and quantitative identification of substances because of its characteristics of fingerprint identification and non-destructive detection. At present, the detection limit of terahertz technology is in the order of milligram. However, the concentration of the tested substance in the actual biomedical samples is usually in the order of microgram or even below, which limits the detection sensitivity and feasibility. In this study, by taking the glioma biomarkers: inositol (MI) and gamma aminobutyric acid (GABA) as an example, based on the inductive-capacitive (LC) resonance, a metamaterial chip was designed to enhance THz detection sensitivity. Then, the terahertz spectra of the chip covered by MI and GABA at different concentrations was tested to prove that their resonance frequency shifts show different rules with the change of concentration, basing on which the qualitative identification could be achieved. And for known samples of MI and GABA, quantitative analysis could be achieved according to the frequency shifts law. According to calculations based on experimental data, the lower detection limits of proposed chip for these two samples are 3.457 µg and 2.552 µg, respectively, which are three orders of magnitude higher than the detection limit of the conventional tableting method. These results have important reference value for the qualitative and quantitative detection of trace specific substances of diseases in later biomedicine.

  • 脑是高度发达的器官,是支配和调节人的一切生理活动的中枢;当脑的结构和功能产生异常时,会引起一系列异常的神经活动,同时也可能导致机体其它器官功能紊乱,进而对人体健康产生危害。其中,脑胶质瘤是最常见的原发性颅脑恶性肿瘤,其具有发病率高、复发率高、死亡率高的特点。常见的脑胶质瘤疾病检测技术主要以伊红染色法等为主的病理学染色技术以及计算机断层扫描(CT)、正电子发射断层扫描(PET)、磁共振成(MRI)为主的脑影像技术[1]。虽然它们各有优点,但在应用于脑胶质瘤疾病研究时,依旧存在一定固有的局限性,例如:组织病理学染色检测周期长;CT和PET检查需要标记放射性元素;MRI的成像分辨率低和敏感性限等。这些局限性将在一定程度上阻碍生物医学准确检测疾病发生后的病理变化以及全掌握脑疾病的发病机制。因此,脑胶质瘤的深入研究迫切地需要新检测技术的突破和发展。

    太赫兹(THz)波是波长范围在0.03~3.00 mm区间的电磁波。它的频率处于微波和红外光波之间,是电子学到光子学的过渡区域[2]。与其他波段相比较,太赫兹波用于生物医学检测时具有如下优点:(1)其能量为毫电子伏特,不会因为电离从而破坏被检测物质本身的属性[2-5];(2)大量生物分子的转动和振动频率在太赫兹频段范围内[4-6],因此可用作物质指纹普识别;(3)与其他化学检测方法(如试剂盒法、质谱法等)相比,太赫兹光谱法可以实现快速无损的样品检测[5-8]

    但是常规的太赫兹光谱检测通常在测量之前将待测物压制为片剂,随着样品浓度的降低,物质的太赫兹特征吸收峰将迅速减小,检测下限在毫克量级。而实际生物医学样本中待测物的浓度通常在微克量级甚至以下,现有方法限制了其检测灵敏度和可行性。以脑组织为例,人脑组织细胞中含有数千种物质,但当疾病发生时,仅几种物质的浓度发生变化,其中仅两三种物质标志脑胶质瘤发生和病变程度,且含量均在5%~10%内[9],如:肌醇(MI)、γ-氨基丁酸(GABA)等。其中MI作为真核生物中许多二级信使的结构基础起着重要的作用。在100 g脑组织中含有约174.3 µmol,且在脑胶质瘤中的含量增加约100%[10]。GABA是大脑中各种抑制神经元和间神经元的主要递质。在正常脑组织的灰质中,GABA浓度在(1.30±0.36) µmol/g范围内,白质中GABA浓度在(0.16±0.16) µmol/g范围内[11],且在脑胶质瘤中含量升高了约9倍[12]。因此,一些研究人员提议将太赫兹光谱法和超材料生物传感器相结合以提高检测限,其优点是:(1)样品消耗低,灵敏度高;(2)无标记检测;(3)快速响应和测量步骤简单[13-15]

    通过常规的压片法太赫兹检测,发现对于微克量级别的样本,太赫兹与样本的共振响应非常微弱,导致光谱特征峰无法识别。因此,文中基于电容电感效应设计了一款太赫兹超材料芯片,以提升太赫兹检测的灵敏度和检测下限。文中以脑胶质瘤标志物MI和GABA的检测为例,将涂有样品的超材料芯片放入太赫兹时域系统中检测,分析了不同浓度的MI和GABA的透射光谱、芯片谐振峰频移量及频移规律,从而实现对MI和GABA的定性识别与定量分析。

  • 纯品肌醇(MI)和γ-氨基丁酸(GABA)(粉末,纯度>99%)均从Sigma Aldrich购置,并按制造商要求存储。对于纯样品的特征峰测试,将纯样品与聚乙烯粉末混合。MI和GABA与聚乙烯粉末的质量混合比 (mg∶mg)分别为32.14∶100和30∶100。用3 t (1 t=1000 kg)的力将混合粉末压制成13 mm的片剂。采样过程中的质量损失均控制在1%以内。

    在超材料实验中,将纯品以不同质量加入至1 mL水(色谱纯)中混合均匀,得到MI的浓度分别为0.28、0.55、0.69、1.10、1.39、2.21、2.78、4.42、5.55、8.83和11.10 µmol/mL,GABA的浓度分别为0.67、2.67、5.33、10.67和21.34 µmol/mL。然后用电动移液枪每次取50 µL滴在超材料芯片表面,彻底干燥进行测试,其中通过固定移液枪的高度来确保每次样品的覆盖面积一致(约为直径5 mm的圆形范围内),使得样本厚度与样本量呈线性关系,因此在光斑下的等效样品量与实际样品量之间为线性关系。

  • 笔者所在团队采用的实验装置为经典的太赫兹时域光谱(THz-TDS)系统,系统动态范围为~60 dB,其有效光谱范围为0.5~3.0 THz,以1024个单点累积数和1.9 GHz的分辨率进行测量。每个样本制备3个样品,重复测量4次得到误差条,所有测量都在低于3%的环境湿度下进行。

  • 物质的太赫兹特征吸收峰来源于太赫兹波与物质分子的振动/转动之间的共振吸收[15]。笔者所在团队将准备好的MI和GABA压片放入太赫兹时域光谱(THz-TDS)系统进行对纯品测试,得到MI和GABA的实际光谱。MI的THz谱如图1(a)所示,在MI有效质量为32.14 mg时,纯品样本对应的特征峰分别在1.47 THz和2.34 THz,而在MI有效质量为2.23 mg时,因特征峰幅度过小而无法有效识别。GABA的THz谱则如图1(b)所示,在GABA有效质量为30 mg时,纯品样本对应的特征峰分别在1.07 THz、1.49 THz和1.99 THz处,其在1.47 THz和2.34 THz处无特征吸收,且在GABA有效质量为1.65 mg时,同样无法识别其特征峰。由此知,传统压片法直接进行MI和GABA物质检测的下限均在毫克量级,无法满足实际样本中低浓度(微克、纳克级别)检测的应用需求。

    Figure 1.  (a) THz characteristic spectrum of MI; (b) THz characteristic spectrum of GABA. The error bars are marked on the curves in blue

  • 该节中,为实现上述脑胶质瘤中特异性物质MI和GABA的高灵敏、低浓度检测,设计了一款超材料传感芯片,下面进行详细阐述。

  • 基于电容电感效应,使用COMSOL软件先设计了一款特征透射型谐振环。为获得较强的透射响应,提高灵敏度,衬底的吸收应越低越好[15]。因此笔者课题组选用熔石英材料,衬底厚度为t=500 µm。通过太赫兹光谱系统测试了它的介电常数。在2.0~2.6 THz范围内,其平均相对介电常数为4+0.05i。超材料谐振周期为p=35 µm,衬底表面用50 nm厚的金(Au)属薄膜(灰色阴影部分)在衬底上形成周期性金属开口谐振环SRR,谐振环的几何参数如图2(a)所示,环开口的大小g=3 µm,内圆直径D2=9.8 µm,外圆直径D1=16 µm,线宽d=3.1 µm。

    Figure 2.  (a) Structure parameters of THz resonant ring chip; (b) Structure multiple layout drawings; (c) Theoretical map data; (d) The normalized electric field distribution calculating by COMSOL

    由于在COMSOL仿真时,纳米级别厚度的金层相对微米级别大小的结构会使网格剖分极为精密,导致非常庞大的计算量。因此,在熔石英衬底的上表面添加一个完美电导体代替实际的金层,减少计算量。其中完美电导体是一层电阻可以忽略不计的介质,由于金在太赫兹波段是一种良好的导电材料,在仿真时可以近似为完美电导体。根据该结构计算得到的太赫兹特征吸收峰位于2.41 THz处,如图2(c)所示。其对应的归一化电场分布如图2(d)所示,可知电场模限制在了圆环状的裂隙处,这是因为当电场偏振方向即垂直于y轴时,即电场偏振方向为x轴方向时,开口环结构能够在金属表面引起感应环流,从而引起载流子分布不均,这种共振被称为LC共振。在介电特性方面,由于LC共振的存在,材料在对应的频率存在介电损耗,最终使其透射频谱出现单峰特性[16]。在样本的覆盖下,超材料芯片的等效电容电感发生变化,导致LC共振频率发生变化。且不同样本的介电常数不同,因此最终超材料芯片的等效电容电感的变化会随物质的种类发生不同变化趋势,导致出现不同的LC共振规律。基于该原理,笔者课题组将该超材料芯片用于生物传感测试。

  • 根据理论设计的参数,芯片加工得到超材料芯片在200倍显微镜下的部分结构如图3(a)所示。该芯片放置于太赫兹时域光谱系统中测得的太赫兹光谱如图3(b) 所示,可以看到结构的谐振峰中心频率处于2.38 THz处,对于实验和仿真的光谱,其特征峰的位置(频率)和强度(透过率)皆存在不同。对于频率上的不同,这是由于在仿真时,考虑到金的电导率良好,为了提升计算效率,笔者所在团队使用一层完美电导体来代替实际的金层,因而最终超结构表面的等效电容电感存在差异,LC振荡的频率也就存在偏移。而透过率的差异则是由于两点:(1)仿真时的完美电导体是一个不计厚度的平面,因此计算时不会考虑金层所产生的介电损耗,而实际结构的金层具有一定厚度,因此会对太赫兹波产生吸收;(2)由于加工技术的限制,存在公差,无法完美实现仿真时的设计。

    Figure 3.  Physical test of THz resonant ring chip. (a) Partial structural photos of metamaterial chip under microscope (200×); (b) THz spectrum measured by the chip. The error bar is marked on the curve in grey

  • 2.3.1 超材料传感芯片对MI样品的响应

    图4(a)所示,在超材料芯片试验中,随着芯片表面MI样品的增加,同时出现了峰频移和峰幅度变化两个现象:(1)峰幅度下降:随着样品浓度增加,其对太赫兹波产生更多吸收,导致光谱透过率降低。(2)峰向低频移动:发现超材料芯片的谐振峰随着样品浓度的变化而移动,因为超材料对其表面覆盖样品的相对介电常数非常敏感。随着相对介电常数的变化,谐振峰的位置也会随之改变[17-18]。同时笔者所在团队还从图4(a)中明显观察到MI浓度增加时,芯片的谐振峰都向低频移动(红移),提取不同浓度的频移量来分析其与浓度的函数关系,对数据进行拟合,见图4(b),通过函数拟合得到线性方程,f(n)=a+b·n,函数中f(n)为频移量(THz),n为样品的浓度(µmol/mL),函数系数为:a=0.00225,b=0.01235,决定系数(R2)=0.9805的拟合优度。

    Figure 4.  Testing results of MI. (a) Terahertz spectra of MI covered metamaterial chips with different concentrations; (b) The change of formant frequency shift of chip under different concentration of MI. The error bars are marked on the curves/points

    2.3.2 超材料传感芯片对不同样品的响应

    另外,用同一芯片测试了脑胶质瘤另一标志物γ-氨基丁酸(GABA),如图5(a)所示,随着芯片表面GABA样品的增加,同样出现了峰频移和峰幅度变化两个现象。提取不同浓度的频移量来分析其与浓度的函数关系,对数据进行拟合,如图5(b)所示,通过函数合得到非线性方程f(n)=a·exp(b·n) + c·exp(d·n),函数中f(n)为频移量(THz),n为样品的浓度(µmol/mL),函数系数为:a=0.07539,b=0.03814,c=−0.07471,d=−0.2549,可以获得决定系数(R2)= 0.9999的拟合优度。

    Figure 5.  Testing results of GABA. (a) Terahertz spectra of metamaterial chips covered with different concentrations of GABA; (b) The change of formant frequency shift of chip under different concentrations of GABA.The error bars are marked on the curves/points

  • 图4图5的结果可知,所设计的超材料芯片对于MI和GABA具有不同的响应函数,虽然由于实验中制备样本和系统抖动都会产生误差,导致函数拟合优度有差异,但是,依旧可以通过对未知样本的浓度累积测试,从而通过识别其频移曲线的参数来定性识别样本的种类。因为MI和GABA的介电常数不同,所以对于覆盖了MI和GABA的超材料芯片,其等效电容电感不同,导致LC振荡频率不同,因此具有不同的频移规律。此外,通过对实验中最低浓度的测试样品的数据计算可得,所设计的芯片对MI和GABA实际检测下限分别达3.457 µg和2.552 µg,优于传统的压片检测三个数量级。

  • 文中基于电容电感效应设计了一款超材料传感芯片并将其应用于脑胶质瘤中特异性物质MI和GABA的检测。根据芯片测试结果可知:(1)相比于传统压片法,采用超材料传感芯片检测MI和GABA能够达到微克量级,依据样品的最低浓度进行计算,可得实际检测下限分别达3.457 µg和2.552 µg,优于传统的压片检测三个数量级,明显提高了检测灵敏度;(2)芯片对于MI和GABA具有超材料特征峰变化规律,因而能够基于这种变化规律实现对于MI和GABA的定性和定量识别。这也表明,所设计的超材料传感芯片不仅能用于脑胶质瘤疾病检测,还可以应用于其他疾病检测,因为不同疾病标志物具有不同的介电常数,使得超材料传感芯片对其具有不同的响应。此外,通过内标法还可以实现通过超材料传感芯片对混合物的定性和定量分析[14]。这些结果证明了将超材料传感芯片应用于脑胶质瘤诊断的可行性,但是对于真实的组织样本,其组分复杂,目前的芯片灵敏度仍然欠缺。因此,需要同时从芯片频移响应和系统的分辨率增强的方向上进行改进。

Reference (18)

Catalog

    /

    DownLoad:  Full-Size Img  PowerPoint
    Return
    Return