Near-infrared Spectroscopy Research Articles

Fishy Forensics: FT-NIR and Machine Learning based authentication of Mediterranean Anchovies (Engraulis encrasicolus)

Seafood authentication and traceability is a challenging issue owing to its complex supply chain and fishing in international waters. NIR spectroscopy has been successfully used to authenticate food of animal and plant origin. In this study, FT-NIR was used to discriminate between Mediterranean anchovies (Engraulis encrasicolus) fished from Adriatic, Balearic, and Tyrrhenian Sea. The spectra were prepared using the standard normal variate (SNV) and the Savitzky-Golay 1st derivative, 2nd order (SG), as well as both together. The model was built, after outlier removal, with linear-support vector machine (L-SVM), polynomial-SVM (P-SVM), k-nearest neighbor (k-NN) and Random Forest (RF). Spectral preprocessing improved model classification accuracy for all algorithms. The data could not be put into groups by linear algorithms like L-SVM and k-NN because the NIR spectra were not linear and had many columns. Non-linear algorithms, P-SVM and RF when coupled with SG+SNV, successfully produced models with maximum robustness. P-SVM and RF models had 100% accuracy in training set and 95.7% and 95.5% accuracy in testing set, respectively and 95.2 and 95.1 accuracy in cross-validation set.

M giants with IGRINS

Context. Neutron-capture elements represent an important nucleosynthetic channel in the study of the Galactic chemical evolution of stellar populations. For stellar populations behind significant extinction, such as those in the Galactic centre and along the Galactic plane, abundance analyses based on near-infrared (NIR) spectra are necessary. Previously, spectral lines from the neutron-capture elements, such as copper (Cu), cerium (Ce), neodymium (Nd), and ytterbium (Yb), have been identified in the H band, while yttrium (Y) lines have been identified in the K band. Aims. Due to the scarcity of spectral lines from neutron-capture elements in the NIR, the addition of useful spectral lines from other neutron-capture elements is highly desirable. The aim of this work is to identify and characterise a spectral line suitable for abundance determination from the most commonly used s-process element, namely barium. Methods. We observed the NIR spectra of 37 M giants in the solar neighbourhood at high spectral resolution and with a high signal-to-noise ratio using the IGRINS spectrometer on the GEMINI South telescope. The full H- and K-bands were recorded simultaneously at R = 45 000. Using a manual spectral synthesis method, we determined the fundamental stellar parameters for these stars and derived the barium abundance from the Ba line (6s5d 3D2 → 6s6p 3P2o) at λair = 23 253.56 Å in the K band. Results. We demonstrate that the Ba line in the K band at 2.33 μm (λ23 253.56) is useful for abundance analyses from the spectra of M giants. The line becomes progressively weaker at higher temperatures and is only useful in M giants and the coolest K giants at supersolar metallicities. Conclusions. We can now add Ba to the trends of the heavy elements Cu, Zn, Y, Ce, Nd, and Yb, which can be retrieved from high-resolution H- and K-band spectra. This opens up the study of nucleosynthetic channels, including the s-process and the r-process, in dust-obscured populations. Thus, these elements can be studied for heavily dust-obscured regions of the Galaxy, such as the Galactic centre.

Optical resonant cavities carving pathways in tunable wavelength sensitive visible-NIR organic photodetectors

Enhancing the photodetection capabilities of organic photodetectors (OPDs) is crucial for advancing applications in medical monitoring, optical communications, image sensing, and robotics, where a strong, focused peak response at a specific designed wavelength is essential for improving sensitivity, wavelength selectivity, and resolution in imaging systems. By controlled integration of ZnO layers within PBDBT:BTP-4F-based OPDs to form a Fabry-Pérot optical cavity, we developed a cost-effective approach to fabricating highly sensitive OPDs by utilizing PBDBT:BTP-4F organic bulk heterojunctions, and extended its detection wavelengths into the near-infrared (NIR) range. Our design integrates a single silver (Ag) layer that significantly enhances peak detection at a wavelength of 830 ​nm, resulting in a remarkably narrow full-width at half maximum (FWHM) wavelength of 30 ​nm and yielding a photoresponse ten times greater than that of non-resonant devices. Furthermore, by varying the thickness of the ZnO layer from 77 ​nm to 620 ​nm, we achieve high spectral tunability, allowing fine adjustments of the resonant peak across a spectrum ranging from ultraviolet (UV) and visible to NIR wavelengths. This sensitive photodetector is also well-suited for applications in photoplethysmography (PPG), effectively detecting pulse signals in the NIR spectrum which has significant potential in medical diagnostics. This work advances the integration of cost-effective, wavelength-selective spectroscopic visible-NIR OPDs, paving the way for the next generation of sensitive photodetectors.

An evaluation of six techniques for measuring porosity of ribbons produced by roller compaction

Ribbon porosity is a critical parameter to monitor in the roller compaction process. In this study, six techniques for measuring the porosity of solid compacts, i.e., manually by caliper (Caliper), X-ray microtomography (µCT), off-line near-infrared spectroscopy (NIR), laser triangulation (Laser), mercury intrusion porosimetry (MIP), and GeoPyc, were compared using a set of rectangular ribblets of microcrystalline cellulose (MCC). These ribblets, which were compressed at 8–––130 MPa on a compaction simulator, exhibited porosities over the range of 0.09 – 0.52. Subsequently, porosities of MCC ribbons made on a roller compactor at specific roll forces of 1.8 kN/cm and 8.8 kN/cm were measured. The Caliper method is convenient for samples with a simple shape but not suitable for real ribbons. The accuracy of GeoPyc measurement relies on accurate conversion factor (unit in cm3/mm), sample shape and size, and sufficient sample volume percentage in the medium. The µCT data is more accurate at lower porosities (< 0.2), while the MIP data is more accurate at higher porosities (> 0.4). The Laser method has good accuracy and is more reproducible compared to other methods in the ribblets measurement. The NIR method is fast, which makes it suitable for in-line monitoring of changes in ribbon quality, but porosity quantification is sensitive to sample presentation, such as surface curvature and roughness. These insights could assist in the choice of the most appropriate method for monitoring ribbon porosity to guide the development and optimization of a roller compaction process for a given formulation.

Radiative recombination model for BiSeI microcrystals: unveiling deep defects through photoluminescence

Abstract Pnictogen chalcohalides are semiconductors that have emerged as promising materials for energy conversion due to their exceptional optoelectronic properties. Their electronic configuration (ns2), particularly for Bi- and Sb-based compounds, can be a key factor in efficient carrier transport and defect tolerance, similarly, to Pb-perovskites. In the present study, the Bi-containing chalcohalide, bismuth selenoiodide (BiSeI) was synthesized via isothermal heat treatment of binary precursors in evacuated quartz ampoules. The synthesized BiSeI microcrystals exhibited a characteristic needle-like morphology and a near-stoichiometric composition. Both indirect and direct band gap energies of BiSeI were determined by ultraviolet–visible–near-infrared diffuse reflectance spectroscopy, with room temperature values of 1.17 eV and 1.29 eV, respectively. This study presents the first experimental investigation of the photoluminescence properties of BiSeI microcrystals resulting in a recombination model involving multiple defect states. This work provides valuable insights into the defect structure and recombination mechanisms within BiSeI, paving the way for further exploration of its potential in optoelectronic devices.

SN 2021adxl: A luminous nearby interacting supernova in an extremely low-metallicity environment

SN 2021adxl is a slowly evolving, luminous, Type IIn supernova with asymmetric emission line profiles, similar to the well-studied SN 2010jl. We present extensive optical, near-ultraviolet, and near-infrared photometry and spectroscopy covering ∼1.5 years post discovery. SN 2021adxl occurred in an unusual environment, atop a vigorously star-forming region that is offset from its host galaxy core. The appearance of Lyα and O II, as well as the compact core, would classify the host of SN 2021adxl as a “Blueberry” galaxy, analogous to higher redshift, low-metallicity, star-forming dwarf “Green Pea” galaxies. Using several abundance indicators, we find a metallicity of the explosion environment of only ∼0.1 Z⊙, the lowest reported metallicity for a Type IIn SN environment. SN 2021adxl reaches a peak magnitude of Mr ≈ −20.2 mag and since discovery, SN 2021adxl has faded by only ∼4 magnitudes in the r band with a cumulative radiated energy of ∼1.5 × 1050 erg over 18 months. SN 2021adxl shows strong signs of interaction with a complex circumstellar medium, seen by the detection of X-rays, revealed by the detection of coronal emission lines, and through multi-component hydrogen and helium profiles. In order to further understand this interaction, we model the Hα profile using a Monte Carlo electron scattering code. The blueshifted high-velocity component is consistent with emission from a radially thin spherical shell resulting in the broad emission components due to electron scattering. Using the velocity evolution of this emitting shell, we find that the SN ejecta collide with circumstellar material of at least ∼5 M⊙ assuming a steady-state mass-loss rate of ∼4 − 6 × 10−3 M⊙ yr−1 for the first ∼200 days of evolution. SN 2021adxl was last observed to be slowly declining at ∼0.01 mag d−1, and if this trend continues, SN 2021adxl will remain observable after its current solar conjunction. Continuing the observations of SN 2021adxl may reveal signatures of dust formation or an infrared excess, similar to that seen for SN 2010jl.

Enhancing soil nitrogen measurement via visible-near infrared spectroscopy: Integrating soil particle size distribution with long short-term memory models

Good quality of soil nitrogen data, which is essential for the advancement of both enhanced agricultural management and ecological environment, traditionally depends on labor intensive chemical procedures. Visible near-infrared (Vis-NIR) spectroscopy, acknowledged for its efficiency, environmental compatibility and rapidity, merges as a promising alternative. However, the effectiveness of Vis-NIR measurement models are significantly compromised by soil particle size distribution (PSD), presenting a substantial challenge in improving the measurement accuracy and reliability. Here an innovative deep learning methodology that integrates PSD with Vis-NIR spectroscopy was proposed for the measurement of nitrogen content in soil samples. By leveraging the LUCAS dataset, different strategies for integrating PSD with Vis-NIR spectral data in deep learning models were explored, revealing that our proposed InSGraL framework, which incorporated mixed features of PSD and spectra as LSTM inputs achieves superior performance. Compared to models utilizing solely Vis-NIR data, InSGraL exhibits a 39.47 % reduction in RMSE and a 42.55 % decrease in MAE, and demonstrates robust performance across various land cover types, achieving an R2 of 0.94 on grassland samples. Moreover, Shapley Additive exPlanations (SHAP) analysis revealed that incorporating PSD modifies the spectral input importance distribution, effectively mitigating spectral interference from particle size while highlighting critical wavelengths previously obscured. This study provides an innovative modeling strategy to mitigate the influence of PSD by integrating it within deep learning framework using Vis-NIR, contributing a deeper understanding of the relationship between PSD and Vis-NIR spectra for the measurement of nitrogen content and offering an effective means to attain soil nitrogen data.

Digital model of biochemical reactions in lactic acid bacterial fermentation of simple glucose and biowaste substrates

As concerns about the environmental impacts of biowaste disposal increase, lactic acid bacterial fermentation is becoming increasingly popular. Current academic research is aimed at the process optimization by developing digital bioreactors. The primary focus is to develop a digital model mimicking the biochemical reactions. In the light of this, this paper intended to build a digital model of biochemical reactions during the fermentation process of both glucose and biowaste substrates, including white pasta and organic municipal waste. For this purpose, near-infrared (NIR) and mid-infrared (MIR) spectroscopy techniques were used to collect spectral information during the fermentation process. Next, the samples were analyzed by High Pressure Liquid Chromatography (HPLC) to measure their glucose, fructose, arabinose, xylose, disaccharide, lactic acid, and acetic acid contents. The results showed that learning algorithms trained on MIR spectra accurately estimated the biochemical reactions for both glucose and biowaste substrates. For the glucose substrate, the results showed R-squared of 0.97 and RMSE of 4.69 g/L for glucose, and R-squared of 0.98 and RMSE of 2.74 g/L for lactic acid. In the case of biowaste substrate, estimations included glucose (R-squared = 0.97, RMSE = 4.69 g/L), fructose (R-squared = 0.88, RMSE = 1.47 g/L), arabinose (R-squared = 0.98, RMSE = 0.55 g/L), xylose (R-squared = 0.93, RMSE = 1.11 g/L), disaccharide (R-squared = 0.90, RMSE = 0.55 g/L), total sugar (R-squared = 0.98, RMSE = 3.79 g/L), lactic acid (R-squared = 0.98, RMSE = 2.74 g/L), and acetic acid (R-squared = 0.97, RMSE = 0.36 g/L). Regarding NIR spectral data, the predictive models were accurate when the substrate was glucose, however, they failed to accurately estimate the chemical reactions in the case of biowaste substrate. The findings of this study can be used to fulfill the requirements for a continuous fermentation process with the objective of maximizing lactic acid production.

Near-infrared Spectral Homogeneity of the Didymos System Before and After the DART Impact* * Based on observations made with the Italian Telescopio Nazionale Galileo (TNG) operated on the island of La Palma by the Fundación Galileo Galilei of the Istituto Nazionale di Astrofisica (INAF) at the Spanish Observatorio del Roque de los Muchachos of the Instituto de Astrofisica de Canarias

We spectroscopically characterized the Didymos system, target of the Double Asteroid Redirection Test (DART)/Light Italian Cubesat for Imaging of Asteroids (LICIACube) space mission, close in time to the DART impact event, during six nights between 2022 August and November at Telescopio Nazionale Galileo. Here, we show that near-infrared (NIR) spectra (0.75–2.25 μm) look mostly similar within the same night and between different nights. They are in good agreement with the only spectrum previously available in the literature, observed almost 20 years before those reported in this paper. During one of the observing nights we also obtain spectroscopy information on the ejecta tail induced by the DART impact. The spectrum of the ejecta tail is also very similar to Didymos/Dimorphos itself. All of these aspects seem to suggest that the Didymos system in the NIR looks mostly homogeneous, with very subtle spectral variations.

Growth, first-principles calculations and optical properties of a novel near-infrared Ho: NLGW laser crystal

A rigid structure was constructed using an isotopic doping strategy, and a novel near-infrared 1 at% Ho3+: NaLa0.93Gd0.06(WO4)2 (Ho: NLGW) laser crystal with a size of Φ 18 mm × 30 mm was successfully grown by the Czochralski method. X-ray rocking curves and Rietveld refinement indicate that the grown crystal possesses high crystalline quality and crystallizes in the I41/a space group with lattice parameters: a = b = 5.3518 Å, c = 11.6438 Å and V = 333.5051 Å3. The Ho: NLGW crystal exhibits a low coefficient of thermal expansion (8.472 × 10−6 K−1), and the c-axis thermal conductivity increases from 1.235 W/(m × K) to 1.404 W/(m × K). The thermal properties of the Ho: NLGW crystal surpass those of other Ho3+-doped tungstate series and mainstream near-infrared laser crystals, suggesting that it is easier to obtain excellent laser gain. For revealing the excellent thermal properties, the elastic moduli of the NLW and NLGW crystals have been calculated by first principles. With the introduction of Gd3+, the structural stiffness of the NLW as a whole has been increased to optimize the mechanical properties in each axial direction and to obtain the reason for the larger Debye temperature. Finally, the transmission, absorption and near-infrared emission spectra of the Ho: NLGW crystal were investigated. The crystal exhibits a high transmittance of 88.5 % and an emission cross-section of 0.84 × 10−20 cm2 calculated according to the J-O theory. Experimental results indicate that the grown Ho: NLGW crystal can achieve large laser output more easily and are expected to be an excellent medium for 2 μm solid-state lasers.

Enhancing forensic blood detection using hyperspectral imaging and advanced preprocessing techniques

Bloodstains are pivotal in forensic investigations as they can provide crucial DNA information about individuals involved in a crime. Traditional methods for bloodstain detection, including chemical tests and forensic lights, have limitations such as non-specificity to human blood and susceptibility to false positives. In forensic blood detection, molecular spectroscopy is crucial for identifying the unique spectral fingerprints of blood, which arise from its molecular composition. Hyperspectral imaging (HSI) leverages this principle by capturing a wide spectrum of light for each pixel in an image, allowing for the detailed analysis of various substances. HSI has emerged as a promising alternative, offering non-contact, rapid, and cost-effective detection of bloodstains by analyzing the visible and near-infrared electromagnetic spectrum. This study explores the application of HSI for blood detection, addressing challenges such as spectral mixing, time-related changes in bloodstain spectra, and data complexity. The study introduces a novel framework to optimize HSI data, enhancing the accuracy and efficiency of bloodstain classification, called the Fast Extraction (FE) framework. It includes two stages. The main method in the first one is the Enhancing Transformation Reduction (ETR) method to reduce the dimension and complexity of the HSI. The second stage contains a compatible classification model to enhance feature extraction and classification. Our approach is validated using the HyperBlood datasets and many evaluation methods, demonstrating superior performance compared to state-of-the-art deep learning models. It provides high accuracy (97%–100 %) for all HSIs, overcoming various difficulties and blood collection times.

Selecting ns2 electron (Sb3+) and d3 electron (Cr3+) co-doped in lead-free halide double perovskite to achieve blue and near-infrared luminescence

The double-perovskite Cs2NaInCl6 has emerged as a highly favored candidate due to its remarkable stability and eco-friendly properties. However, it is noteworthy that this material does not demonstrate the desired optical performance in the crucial visible and near-infrared (NIR) spectrum, which may limit its potential applications in certain domains. We develop such characteristics in Cs2NaInCl6 by co-doping Sb3+ (s-electron) and Cr3+ (d-electron) ions. Doping with Sb3+ allows for 1S0→3P1 transition absorption, which then emits blue photoluminescence of self-trapped exciton (STE). By co-doping Sb3+ and Cr3+ ions, it causes the transfer of excitation energy from STE to Cr3+, yielding strong NIR emission at 950 nm. Temperature-dependent Photoluminescence (PL), PL excitation, PL decay and Tanabe-Sugano diagram provide insight into photophysical processes. Furthermore, a novel dual-modal optical information encryption based on the integration of blue and NIR luminescence printing patterns using Cs2NaInCl6: Sb3+ and Cs2NaInCl6: Sb3+/Cr3+, respectively, is successfully demonstrated.

Evaluation of foveal avascular zone changes in patients with type 1 diabetes using optical coherence topography angiography

Aim To evaluate vascular changes in the foveal avascular zone in pediatric patients with type 1 diabetes by using optical coherence tomography angiography (OCTA). Patients and methods This cross-sectional study was performed in Benha University Hospital at Department of Ophthalmology. The patients were obtained from the Pediatric Endocrinology Clinic in Benha University Hospital. The study was conducted on 48 patients (96 eyes) divided into two groups: group A, 24 diabetic children with type 1 diabetes, group B, 24 healthy children of the same age and sex who met the inclusion criteria. The patients were collected from April 2022 to April 2023. The patients were photographed using the Optovue AngioVue OCTA System. Results Our results show that according to the foveal avascular zone it was slightly wider in type 1 diabetic cases with no clinically evident diabetes retinopathy (DR) than control. There were insignificant differences between the two groups as regards foveal density, but as regards foveal thickness, it was significantly lower in diabetic cases than in the control group. Conclusion We discovered that OCTA could consistently identify microvascular changes in the Deep and Superficial Capillary Plexus prior to the clinical indications’ onset of DR. In order to assess the OCTA potential for the screening of DR in children with T1DM, it is necessary to conduct numerous prospective studies.

Accidental Corneal Intrastromal Intraocular Lens Implantation During Cataract Surgery: A Case Report

Purpose To report an unusual but important complication of routine cataract surgery, accidental intrastromal intraocular lens (IOL) implantation. Methods During cataract surgery in a 66-year-old man, the IOL was delivered into the intrastromal space of the cornea accidentally. The lens was removed immediately and carefully by enlarging the corneal tunnel. Results Anterior segment optical coherence topography (AS-OCT) was used for postoperative follow-up of the patient, which showed a viscoelastic substance in the stroma of the cornea that was washed out. Ther patient's final visual acuity was 20/20. Conclusions A uniplanar corneal incision with wound-assisted IOL insertion can lead to intrastromal IOL injection. If a cataract surgeon encounters such a complication, AS-OCT could be a useful imaging technique for follow-up. [ Journal of Refractive Surgery Case Reports . 2024;4(4):e36–e39.]

Hemodynamics and vascular oxygenation measured at the forehead during changes in respiration: A SPA-fNIRS Study

Cerebral blood flow is influenced by respiration, primarily through changes in the CO2 concentration of arterial blood. The objective of this study was to investigate the effect of changes in arterial CO2 concentration induced by respiratory changes on oxygenation and hemodynamics in the cerebral and extracerebral tissue layers of the forehead. We used systemic physiology augmented functional near-infrared spectroscopy (SPA-fNIRS) to assess changes in forehead hemodynamics and oxygenation as well as systemic physiology in 20 healthy subjects. All participants performed two respiratory tasks (breath-holding and hyperventilation). In our SPA-fNIRS study we found that changes in breathing affected hemodynamics and oxygenation in both the extracerebral and cerebral tissue layers of the forehead in characteristic ways, depending on the two respiratory tasks. The results show that extracerebrovascular reactivity (ECVR) exists in parallel with the well-known cerebrovascular reactivity (CVR): CVR and ECVR must be considered when performing fNIRS neuroimaging studies involving changes in respiration.

Non-destructive Measurement of Potassium Content in Low-Potassium Radish and Turnip Using Visible Near-Infrared Spectroscopy

Non-destructive Measurement of Potassium Content in Low-Potassium Radish and Turnip Using Visible Near-Infrared Spectroscopy

Comparing different motion correction approaches for resting-state functional connectivity analysis with functional near-infrared spectroscopy data.

Motion artifacts are a notorious challenge in the functional near-infrared spectroscopy (fNIRS) field. However, little is known about how to deal with them in resting-state data. We assessed the impact of motion artifact correction approaches on assessing functional connectivity, using semi-simulated datasets with different percentages and types of motion artifact contamination. Thirty-five healthy adults underwent a 15-min resting-state acquisition. Semi-simulated datasets were generated by adding spike-like and/or baseline-shift motion artifacts to the real dataset. Fifteen pipelines, employing various correction approaches, were applied to each dataset, and the group correlation matrix was computed. Three metrics were used to test the performance of each approach. When motion artifact contamination was low, various correction approaches were effective. However, with increased contamination, only a few pipelines were reliable. For datasets mostly free of baseline-shift artifacts, discarding contaminated frames after pre-processing was optimal. Conversely, when both spike and baseline-shift artifacts were present, discarding contaminated frames before pre-processing yielded the best results. This study emphasizes the need for customized motion correction approaches as the effectiveness varies with the specific type and amount of motion artifacts present.

Atypical prefrontal neural activity during an emotional interference control task in adolescents with autism spectrum disorder: A functional near-infrared spectroscopy study

Atypical prefrontal neural activity during an emotional interference control task in adolescents with autism spectrum disorder: A functional near-infrared spectroscopy study

Characteristic Changes of Prefrontal and Motor Areas in Patients with Type 2 Diabetes and Major Depressive Disorder During a Motor Task of Tai Chi Chuan: A Functional Near-Infrared Spectroscopy Study.

This cross-sectional study aims to identify the characteristic changes of prefrontal and motor areas during a tai chi chuan task in patients with Type 2 diabetes mellitus (T2DM) and major depressive disorder (MDD) using wearable functional near-infrared spectroscopy (fNIRS). Three parallel groups (T2DM with DD group, T2DM group, and healthy group) were recruited from December 10, 2022, to May 31, 2023. Participants in three groups conducted a motor task of tai chi chuan designed by Eprime 3.0, and fNIRS was used to monitor the brain activation, functional connectivity (FC), and lateralization of prefrontal and motor areas. Correlation analyses were performed to examine the relationship between depressive symptoms and the function of prefrontal and motor areas. Ninety elder adults (aged ≥ 60), including 30 patients with T2DM and MDD, 30 patients with T2DM, and 30 healthy subjects, were enrolled. In contrast with the patients with T2DM and healthy subjects, the patients with T2DM and MDD had decreased activation and abnormal lateralization in prefrontal and motor areas and decreased FC among supplementary motor area, motor area, and dorsolateral prefrontal cortex (DLPFC). Furthermore, the oxyhemoglobin (HbO2) concentration value of DLPFC in patients with T2DM and MDD was negatively associated with scores of Hamilton Depression Scale-24 (HAMD-24). Patients with T2DM and MDD had characteristic functional changes in prefrontal and motor areas. DLPFC may be a potential target of diagnosis and intervention for patients with T2DM and MDD.

CT-Net: an interpretable CNN-Transformer fusion network for fNIRS classification.

Functional near-infrared spectroscopy (fNIRS), an optical neuroimaging technique, has been widely used in the field of brain activity recognition and brain-computer interface. Existing works have proposed deep learning-based algorithms for the fNIRS classification problem. In this paper, a novel approach based on convolutional neural network and Transformer, named CT-Net, is established to guide the deep modeling for the classification of mental arithmetic (MA) tasks. We explore the effect of data representations, and design a temporal-level combination of two raw chromophore signals to improve the data utilization and enrich the feature learning of the model. We evaluate our model on two open-access datasets and achieve the classification accuracy of 98.05% and 77.61%, respectively. Moreover, we explain our model by the gradient-weighted class activation mapping, which presents a high consistent between the contributing value of features learned by the model and the mapping of brain activity in the MA task. The results suggest the feasibility and interpretability of CT-Net for decoding MA tasks.