CN108090936A

CN108090936A - The scanning mating plate tomograph imaging method of Qie Lunkefu fluorescence excitations

Info

Publication number: CN108090936A
Application number: CN201711358717.3A
Authority: CN
Inventors: 冯金超; 张娜; 贾克斌; 李哲; 孙中华
Original assignee: Beijing University of Technology
Current assignee: Beijing University of Technology
Priority date: 2017-12-17
Filing date: 2017-12-17
Publication date: 2018-05-29
Anticipated expiration: 2037-12-17
Also published as: CN108090936B

Abstract

The invention discloses the scanning mating plate tomograph imaging methods of Qie Lunkefu fluorescence excitations, belong to field of medical image processing.This method rebuilds the distribution of fluorescence quantum yield using the fluoroscopic image obtained in organism surface highly-sensitive detector.This method includes two processes, first, generating X-ray mating plate scanning organism generation Cerenkov radiation by linear accelerator and being used as internal excitation light source activation fluorogen, i.e. excitation process；Fluorescent yield can be rebuild exactly using external highly-sensitive detector acquisition fluorescence signal, i.e. emission process, this method second is that the fluorescence inspired is penetrated after biological tissue reaches organism surface.

Description

Scanning light slice tomography method based on Cerenkov fluorescence excitation

技术领域technical field

本发明属于医学图像处理领域，涉及一种切伦可夫荧光激发的扫描光片断层成像方法。The invention belongs to the field of medical image processing, and relates to a scanning light slice tomographic imaging method excited by Cherenkov fluorescence.

背景技术Background technique

放射治疗是利用放射线治疗肿瘤的一种方法，在肿瘤治疗中的作用和地位日益突出。中国2015年癌症新发病例约430万，其中91.9万人使用了放射治疗，放射治疗已成为治疗恶性肿瘤的主要手段之一。如果能在放射治疗时在体揭示恶性肿瘤细胞生理学上的活动，将对于恶性肿瘤的治疗有着重要的帮助。Radiation therapy is a method of using radiation to treat tumors, and its role and status in tumor treatment is becoming more and more prominent. There were approximately 4.3 million new cancer cases in China in 2015, of which 919,000 received radiation therapy. Radiation therapy has become one of the main means of treating malignant tumors. If the physiological activities of malignant tumor cells can be revealed in vivo during radiotherapy, it will be of great help to the treatment of malignant tumors.

切伦可夫荧光激发的扫描成像(以下简称CELSI)作为一种新兴的医学成像技术，是一种基于放射产生的切伦可夫光作为激光光源来激发生物体内部的荧光团，并在生物体外通过高灵敏度的探测器接收发射的荧光进行成像的技术。CELSI利用切伦科夫辐射作为内在的激发光源，从而获得了更低的背景噪声和更高的成像灵敏度。此外，CELSI技术利用内在的激发光源激发荧光探针，有效地克服了外部激光光源穿透能力有限的问题，增加了成像深度。另外，光片扫描技术的使用有效地提高了CELSI技术的空间分辨率和深度灵敏度，可以在直径约为1毫米、深度2厘米的病灶进行成像。正是由于CELSI技术具有的这些独特特性，开展该技术研究将促进其在生物医学领域的应用。但是CELSI是一种二维平面成像技术，无法提供荧光探针在生物体内的深度信息，也无法进行定量分析，从而限制了其在生物体中的应用。Cerenkov fluorescence-excited scanning imaging (hereinafter referred to as CELSI), as an emerging medical imaging technology, is based on Cherenkov light generated by radiation as a laser source to excite fluorophores inside organisms, and in biological A technique for imaging in vitro by receiving emitted fluorescence with highly sensitive detectors. CELSI utilizes Cherenkov radiation as the intrinsic excitation light source, resulting in lower background noise and higher imaging sensitivity. In addition, CELSI technology uses the internal excitation light source to excite the fluorescent probe, which effectively overcomes the problem of limited penetration ability of the external laser light source and increases the imaging depth. In addition, the use of light-sheet scanning technology effectively improves the spatial resolution and depth sensitivity of CELSI technology, enabling imaging of lesions with a diameter of about 1 mm and a depth of 2 cm. It is precisely because of these unique characteristics of CELSI technology that research on this technology will promote its application in the field of biomedicine. However, CELSI is a two-dimensional planar imaging technology, which cannot provide depth information of fluorescent probes in vivo, nor can it perform quantitative analysis, thus limiting its application in living organisms.

发明内容Contents of the invention

为克服上述背景技术描述的问题，本发明首次提出了切伦科夫激发的荧光扫描断层成像方法。In order to overcome the problems described in the background technology above, the present invention proposes a Cerenkov excited fluorescence scanning tomography method for the first time.

本发明采用的技术方案为切伦可夫荧光激发的扫描光片断层成像方法，该方法利用在生物体表高灵敏度探测器获取的荧光图像来重建荧光量子产额的分布。该方法包括两个过程，一是通过直线加速器产生X射线光片扫描生物体产生切伦科夫辐射并作为内部激发光源激发荧光团，即激发过程；二是激发出的荧光穿透生物组织到达生物体表后利用外在的高灵敏度探测器获取荧光信号，即发射过程。The technical scheme adopted in the present invention is a scanning light slice tomography method excited by Cherenkov fluorescence, which utilizes fluorescence images acquired by high-sensitivity detectors on the surface of living organisms to reconstruct the distribution of fluorescence quantum yields. The method includes two processes, one is to scan the living body through the X-ray light sheet generated by the linear accelerator to generate Cerenkov radiation and use it as an internal excitation light source to excite the fluorophore, that is, the excitation process; the other is the excited fluorescence to penetrate the biological tissue to reach After the surface of the organism, an external high-sensitivity detector is used to obtain the fluorescent signal, that is, the emission process.

为准确地描述激发和发射过程，本方法采用耦合的扩散近似方程，扩散近似方程形式为：In order to accurately describe the excitation and emission process, this method uses a coupled diffusion approximation equation, and the form of the diffusion approximation equation is:

其中，下标x和m分别代表激发和发射两个波段，▽为梯度算子，Φ_x(r)是r处激发波段的切伦科夫光强，Φ_m(r)是发射波段r处的荧光光强，μ_ai(r)(i＝x,m)代表光学吸收系数，D_i(r)(i＝x,m)代表光学扩散系数，μ_af(r)是荧光团对激发光的吸收系数，η是荧光团量子效率，ημ_af(r)是需要重建的荧光量子产额。对于公式(1)中的光源项q(r)，它是一个内在的激发光源而不是外在的，是通过放射治疗中直线加速器产生的切伦科夫辐射。考虑到大于6兆电子伏的X射线在生物组织中衰减小，且光束厚度小于5毫米，因此将切伦科夫辐射简化为一个均匀分布的光片光源。Among them, the subscripts x and m represent the excitation and emission bands respectively, ▽ is the gradient operator, Φ _x (r) is the Cerenkov light intensity at the excitation band at r, and Φ _m (r) is the emission band at r , μ _ai (r)(i=x,m) represents the optical absorption coefficient, D _i (r)(i=x,m) represents the optical diffusion coefficient, μ _af (r) is the fluorophore’s response to the excitation light , η is the fluorophore quantum efficiency, and ημ _af (r) is the fluorescence quantum yield that needs to be reconstructed. For the light source term q(r) in Equation (1), it is an intrinsic excitation light source rather than an extrinsic one, which is Cherenkov radiation generated by a linear accelerator in radiotherapy. Considering that X-rays greater than 6 MeV have little attenuation in biological tissues, and the beam thickness is less than 5 mm, the Cerenkov radiation is simplified as a uniformly distributed light sheet source.

采用罗宾型边界条件：Robin-type boundary conditions are used:

式(2)中，下标x和m分别代表激发和发射两个波段，是边界的单位法向量，A是取决于边界光学反射系数偏差的特定常量。In formula (2), the subscripts x and m represent two bands of excitation and emission, respectively, is the boundary The unit normal vector of , A is a specific constant that depends on the deviation of the boundary optical reflectance.

切伦科夫激发的荧光扫描断层成像方法就是利用发射过程中采集的荧光图像来重建荧光量子产额，这是典型的病态逆问题。为重建荧光量子产额，基于正则化理论，第k次迭代的荧光产额变化量通过式(3)中的迭代方程更新得到：The Cherenkov-induced fluorescence scanning tomography method uses the fluorescence images collected during the emission process to reconstruct the fluorescence quantum yield, which is a typical ill-conditioned inverse problem. In order to reconstruct the fluorescence quantum yield, based on the regularization theory, the fluorescence yield variation of the kth iteration Through the update of the iterative equation in formula (3):

其中，k表示迭代次数，λ是正则化参数，Φ^meas是测量的荧光图像，是计算的发散过程荧光图像，即通过求解方程(1)得到；T表示矩阵转置，J_k是雅克比矩阵，其形式为：where k represents the number of iterations, λ is the regularization parameter, Φ ^meas is the measured fluorescence image, is the calculated fluorescence image of the divergence process, which is obtained by solving equation (1); T represents the matrix transpose, J _k is the Jacobian matrix, and its form is:

其中，表示数学中的偏导算子，I_i(i＝1,...,NM)表示测量的荧光图像Φ^meas中第i个测量值，雅克比矩阵J的大小为(M×N)×NN。M表示光片扫描的数量，N表示边界测量点的数目，NN表示将生物体有限元离散后节点的数目。雅克比矩阵J_k是通过有限元方法对公式(1)进行离散后得到的。in, Represents the partial derivative operator in mathematics, I _i (i=1,...,NM) represents the i-th measurement value in the measured fluorescence image Φ ^meas , and the size of the Jacobian matrix J is (M×N)×NN . M represents the number of light sheet scans, N represents the number of boundary measurement points, and NN represents the number of nodes after the finite element of the organism is discretized. The Jacobian matrix J _k is obtained by discretizing the formula (1) through the finite element method.

如果当前迭代次数k大于最大迭代次数k_max或者时，式(2)停止迭代，否则继续迭代直到满足停止条件。If the current iteration count k is greater than the maximum iteration count k _max or When , formula (2) stops iterating, otherwise it continues until the stopping condition is satisfied.

附图说明Description of drawings

图1为仿体形状，图中黑色的圆点表示探测器位置；Figure 1 is the shape of the phantom, and the black dot in the figure indicates the position of the detector;

图2为扫描方式示意图,箭头表示光片的扫描角度,直线表示光片位置。Figure 2 is a schematic diagram of the scanning method, the arrow indicates the scanning angle of the light sheet, and the straight line indicates the position of the light sheet.

图3为荧光团分布示意图。图中黑色区域表示荧光团的位置。Figure 3 is a schematic diagram of the distribution of fluorophores. The black area in the figure indicates the position of the fluorophore.

图4为重建的荧光产额分布。Figure 4 shows the reconstructed fluorescence yield distribution.

具体实施方式Detailed ways

下面根据具体实施示例与附图对本发明进行说明。The present invention will be described below based on specific implementation examples and accompanying drawings.

首先，建立一个长、宽分别为10cm和6cm的矩形仿体，如图1所示。在仿真实验中，假定仿体是均匀的。激发波段的μ_ax为0.009mm-1,D_x为0.25mm-1,发射波段的μ_am为0.006mm-1,D_m为0.26mm-1。表光学扩散系数试验中共设置了67个荧光检测器，将这些检测器等距离的放置在仿体的上边界，如图1所示。在试验中，从仿体自上到下每间隔0.5cm就采用光片扫描一次，共扫描21次，扫描方式示意图如图2所示。在本实验中，将直径为10mm，荧光产额浓度为0.008mm-1的圆形荧光团放置在中心为(0,0)mm位置处，如图3所示。为了重建荧光产额，该仿体被离散化为2747个点和5280个三角形。最大迭代次数k_max设为10，停止阈值ε设为0.01。通过本方法计算，可以得出如图4所示的荧光产额重建结果。实验结果表明，本方法可以准确地对荧光产额进行重建。First, create a rectangular phantom whose length and width are 10cm and 6cm respectively, as shown in Figure 1. In the simulation experiments, it is assumed that the phantom is uniform. The μ _ax in the excitation band is 0.009mm-1, the D _x is 0.25mm-1, the μ _am in the emission band is 0.006mm-1, and the D _m is 0.26mm-1. A total of 67 fluorescence detectors were set up in the optical diffusion coefficient test, and these detectors were placed equidistantly on the upper boundary of the phantom, as shown in Figure 1. In the experiment, from the top to the bottom of the phantom, the light sheet was used to scan once at an interval of 0.5 cm, for a total of 21 scans. The schematic diagram of the scanning method is shown in Figure 2. In this experiment, a circular fluorophore with a diameter of 10mm and a fluorescence yield concentration of 0.008mm-1 was placed at the center of (0,0)mm, as shown in Figure 3. To reconstruct the fluorescence yield, the phantom was discretized into 2747 points and 5280 triangles. The maximum number of iterations k _max is set to 10, and the stopping threshold ε is set to 0.01. Through calculation by this method, the fluorescence yield reconstruction results shown in Figure 4 can be obtained. Experimental results show that this method can accurately reconstruct the fluorescence yield.

Claims

1. The tomography method of the scanning light sheet excited by Cherenkov fluorescence is characterized in that: the method utilizes a fluorescence image acquired by a high-sensitivity detector on the biological surface to reconstruct the distribution of fluorescence quantum yield; the method comprises two processes, namely, an X-ray film generated by a linear accelerator is used for scanning an organism to generate Cerenkov radiation and is used as an internal excitation light source to excite a fluorophore, namely, an excitation process; secondly, after the excited fluorescence penetrates through biological tissues and reaches the body surface of a living body, an external high-sensitivity detector is used for acquiring a fluorescence signal, namely an emission process;

to accurately describe the excitation and emission processes, the method employs a coupled diffusion approximation equation of the form:

wherein, subscripts x and m represent excitation and emission bands respectively,. v is gradient operator,. phi_x(r) Cerenkov intensity at the excitation band at r, phi_m(r) is the intensity of the fluorescence light at the emission band r, μ_ai(r) (i ═ x, m) represents an optical absorption coefficient, D_i(r) (i ═ x, m) represents an optical diffusion coefficient, μ_af(r) is the absorption coefficient of the fluorophore for the excitation light, η is the quantum efficiency of the fluorophore, η μ_af(r) is the fluorescence quantum yield to be reconstructed; for the light source term q (r) in equation (1), which is an intrinsic excitation light source rather than extrinsic, it is cerenkov radiation generated by a linear accelerator in radiotherapy; considering that the attenuation of X-rays larger than 6 MeV in biological tissues is small, and the thickness of a light beam is less than 5mm, the Cerenkov radiation is simplified into a uniformly distributed light sheet light source;

the robin type boundary conditions are adopted:

<mrow> <mtable> <mtr> <mtd> <mrow> <msub> <mi>&Phi;</mi> <mrow> <mi>x</mi> <mo>,</mo> <mi>m</mi> </mrow> </msub> <mrow> <mo>(</mo> <mi>r</mi> <mo>)</mo> </mrow> <mo>+</mo> <mn>2</mn> <msub> <mi>AD</mi> <mrow> <mi>x</mi> <mo>,</mo> <mi>m</mi> </mrow> </msub> <mrow> <mo>(</mo> <mi>r</mi> <mo>)</mo> </mrow> <mo>&lsqb;</mo> <mover> <mi>v</mi> <mo>&RightArrow;</mo> </mover> <mrow> <mo>(</mo> <mi>r</mi> <mo>)</mo> </mrow> <mo>&CenterDot;</mo> <msub> <mi>&Phi;</mi> <mrow> <mi>x</mi> <mo>,</mo> <mi>m</mi> </mrow> </msub> <mrow> <mo>(</mo> <mi>r</mi> <mo>)</mo> </mrow> <mo>&rsqb;</mo> <mo>=</mo> <mn>0</mn> </mrow> </mtd> <mtd> <mrow> <mi>r</mi> <mo>&Element;</mo> <mo>&part;</mo> <mi>&Omega;</mi> </mrow> </mtd> </mtr> </mtable> <mo>-</mo> <mo>-</mo> <mo>-</mo> <mrow> <mo>(</mo> <mn>2</mn> <mo>)</mo> </mrow> </mrow>

in formula (2), the subscripts x and m represent excitation and emission bands, respectively,is a boundaryA is a specific constant depending on the deviation of the boundary optical reflection coefficient;

the fluorescence scanning tomography method excited by Cerenkov is to reconstruct fluorescence quantum yield by using a fluorescence image acquired in the emission process, which is a typical ill-conditioned inverse problem; for reconstructing fluorescence quantum yield, based on regularization theory, the amount of fluorescence yield change of the kth iterationUpdated by the iterative equation in equation (3) to obtain:

<mrow> <msubsup> <mi>&delta;&eta;&mu;</mi> <mrow> <mi>a</mi> <mi>f</mi> </mrow> <mi>k</mi> </msubsup> <mo>=</mo> <msup> <mrow> <mo>(</mo> <msubsup> <mi>J</mi> <mi>k</mi> <mi>T</mi> </msubsup> <msub> <mi>J</mi> <mi>k</mi> </msub> <mo>+</mo> <mi>&lambda;</mi> <mi>I</mi> <mo>)</mo> </mrow> <mrow> <mo>-</mo> <mn>1</mn> </mrow> </msup> <msubsup> <mi>J</mi> <mi>k</mi> <mi>T</mi> </msubsup> <mrow> <mo>(</mo> <msup> <mi>&Phi;</mi> <mrow> <mi>m</mi> <mi>e</mi> <mi>a</mi> <mi>s</mi> </mrow> </msup> <mo>-</mo> <msubsup> <mi>&Phi;</mi> <mi>k</mi> <mrow> <mi>c</mi> <mi>a</mi> <mi>l</mi> <mi>c</mi> </mrow> </msubsup> <mo>)</mo> </mrow> <mo>-</mo> <mo>-</mo> <mo>-</mo> <mrow> <mo>(</mo> <mn>3</mn> <mo>)</mo> </mrow> </mrow>

where k denotes the number of iterations, λ is a regularization parameter, Φ^measIs a measured image of the fluorescence light,is a calculated fluorescence image of the divergent process, i.e. obtained by solving equation (1); t denotes the matrix transposition, J_kIs a Jacobian matrix of the form:

wherein,in the mathematics of representationPartial derivative operator of, I_i(i 1.., NM) represents a measured fluorescence image Φ^measFor the ith measurement value, the size of the Jacobian matrix J is (M multiplied by N) multiplied by NN; m represents the number of optical sheet scans, N represents the number of boundary measurement points, and NN represents the number of nodes after a finite element of an organism is dispersed; jacobi matrix J_kIs obtained by dispersing the formula (1) by a finite element method.

2. The cherenkov fluorescence-excited scanning light sheet tomography method of claim 1, wherein: if the current iteration number k is larger than the maximum iteration number k_maxOrAnd (3) stopping iteration by the formula (2), otherwise, continuing iteration until a stopping condition is met.