Gliomas, the most prevalent primary brain tumors, exhibit complex genetic and epigenetic variations, including ATRX mutations. Existing ATRX status diagnostics, like immunohistochemistry, DNA, and RNA sequencing, face limitations. Terahertz spectroscopy, known for its interaction with biological materials, holds potential for ATRX diagnosis due to its non-invasive genetic and structural insights. This study proposes an innovative methodology integrating deep learning and terahertz spectroscopy for ATRX assessment. The approach begins by transforming one-dimensional terahertz data into two-dimensional images, enhancing data richness. A Deep Convolutional Generative Adversarial Network (DCGAN) augments the image dataset, addressing data scarcity. DCGAN generates realistic images by training a generator and discriminator in tandem. Subsequently, a Residual Network (ResNet) extracts features from augmented images, tackling the vanishing gradient issue. The ResNet model captures crucial complexities essential for accurate ATRX prediction. Extracted features feed into a classifier for final prediction. The study encompasses 22 patients with 440 terahertz spectral data. Dataset contained 220 ATRX-positive and 220 ATRX-negative spectral data. Employing terahertz data and deep learning, the model achieved up to 90.64% accuracy in diagnosing ATRX status. This research introduces a novel approach integrating terahertz spectroscopy and deep learning for enhanced precision in glioma ATRX diagnosis. The method's potential impact extends to personalized treatment and improved prognosis. Moreover, it underscores the broader utility of terahertz spectroscopy and deep learning in advancing genetic alteration diagnostics in diverse cancers.
Extracting spectral parameters of materials is an essential application of terahertz spectroscopy technology, based on the characteristics of coherent detection. However, the commonly used algorithm for extracting spectral parameters requires a parallel and smooth surface as a prerequisite, and the surface roughness will affect the extraction result. Nevertheless, the effect degree and the mechanism are not precise before. Firstly, in this research, optical parameters of samples with different roughness extracted by the algorithm are displayed. Therefore, the effect degree of different roughness on the optical extraction algorithm is clarified. After that, the mechanism of the influence is analyzed through the method of microelement modeling. As a result, it shows that when the sample surface is slightly rough (roughness<60μm), it will not significantly impact the extraction results. The shape of the optical curve will not be significantly distorted. As the roughness increases, the change of the statistical distribution of the Fresnel coefficient and the phase change are the reason for the attenuation of the terahertz wave amplitude and the decrease of the extraction accuracy.
Terahertz time-domain spectroscopy has been widely applied in performing dielectric analysis for various materials. However, air voids trapped in samples will significantly affect the characterization. In this study, the refractive index of mixture samples consisting of iron trioxide and polytetrafluoroethylene in five different mass ratios were measured with terahertz time-domain spectroscopy. In order to extract the intrinsic refractive index of iron trioxide, the effective medium model of CRI was then applied to remove refractive index of PTFE in the sample. The extracted refractive indices were presented as a variable parameter decrease with the increase of iron trioxide content, which also corresponds to the increase of air voids in the tablets. The correlation between trapped air voids and analyte composition roots from the coarse rust particles used in pellets compression. This study shows that the influence of air porosity should be considered when terahertz waves are utilized to characterize dielectric property of coarse material.
Traditional terahertz time-domain spectroscopy (THZ-TDS) can provide a broadband spectral response of the measured object and has been widely used. Notwithstanding, for applications that require real-time quantitative detection such as security inspection, more attention should be paid to the terahertz response around the absorption characteristics of the sample. Terahertz frequency domain spectroscopy (THz-FDS) using frequency modulation continuous wave can better meet the real-time requirements of security applications. In this paper, an analytical method is established to achieve accurate prediction of molar concentration for organics. The absorption coefficient spectra of samples with different molar concentrations are obtained by using traditional THz-TDS in a wide frequency range, and then the THz-FDS based on photomixing technology is applied to locking a narrow band range near the absorption peak for rapid quantitative analysis. The scheme was verified by taking α-lactose monohydrate as an example, and the results showed that the mean square error of concentration prediction was only 0.025 under the interference of water vapor environment. It may shed light on terahertz rapid quantitative detection of organic compounds in realistic security scene.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.