RGB Model Research Articles

First fluorescence method for monitoring the doping stimulant drug solriamfetol in plasma and urine: A pharmacokinetic study in volunteers’ urine with greenness, whiteness, and blueness assessments

A green, sustainable, simple, and highly sensitive spectrofluorimetric method was developed to determine solriamfetol hydrochloride for the first time in biological fluids and dosage form. To the best of our knowledge, no Fluorimetric methods have been reported for solriamfetol determination in plasma and urine. The method is based on measuring the intrinsic fluorescence of solriamfetol hydrochloride at 522 nm after excitation at 260 nm in ammonium acetate buffer solution at pH 4. All factors influencing the response have been carefully investigated. The method was linear over a concentration range of 10 to 1800 ng/mL (r = 0.9998). The limit of detection (LOD) and limit of quantification (LOQ) values were found to be 3.2 and 9.8 ng/mL, respectively, indicating the excellent sensitivity of the proposed method. The pharmaceutical dosage form, spiked human plasma, and urine samples were successfully analyzed using the described method with average recoveries of 99.79 ± 0.739, 99.27 ± 1.885, and 99.57 ± 1.701, respectively. Plasma samples were first subjected to protein precipitation using 10 % perchloric acid, while urine samples were analyzed directly without any pretreatment. Furthermore, the suggested method was applied effectively on real human volunteers’ urine to measure renal clearance, the cumulative amount of solriamfetol excreted unchanged in the urine, and the percent of the dose recovered. The eco-friendly nature and sustainability of this work were evaluated using trending greenness, whiteness, and blueness assessment metrics. The greenness assessment tools, namely the Analytical GREEnness metric tool (AGREE), Analytical Eco-scale, and complementary green analytical procedure index (complex GAPI) were selected to prove the eco-friendly nature of the method. Additionally, whiteness and blueness assessments were conducted using the Red-Green-Blue model (RGB model), and high Blue Applicability Grade Index (BAGI), respectively. The proposed method is a promising analytical tool for routine analysis in quality control labs, therapeutic drug monitoring, and pharmacokinetics studies.

The determination of bisphenols contained in food-related products with increased accuracy provided by the H-point Standard Addition Method (HPSAM)-based methodology

The basic version of the H-point standard addition method and its variants have been developed with an increased number of measurement conditions (from two to five conditions allowed for differentiation of sensitivity). The methodology adapted to the signal increment and generalized versions was used for the determination of the harmful compounds from the group of bisphenols (bisphenol A, S, and F) using high-performance liquid chromatography with a diode array detector. This approach in theoretical and experimental studies allowed for obtaining high-quality results after eliminating all types of interference effects: multiplicative, additive, and nonlinear. The analysis of food-related samples, such as store receipts and canned food, confirmed the effectiveness of the developed methodology. The analytical results were distinguished by very good accuracy (RE<5 %, RV∼100 %) and precision (CV≤4 %). The approach can be successfully applied to various analytical systems (RGB model) and is characterized by a high greenness value (AGREE approach).

MVP-HOT: A moderate visual prompt for hyperspectral object tracking

The growing attention to hyperspectral object tracking (HOT) can be attributed to the extended spectral information available in hyperspectral images (HSIs), especially in complex scenarios. This potential makes it a promising alternative to traditional RGB-based tracking methods. However, the scarcity of large hyperspectral datasets poses a challenge for training robust hyperspectral trackers using deep learning methods. Prompt learning, a new paradigm emerging in large language models, involves adapting or fine-tuning a pre-trained model for a specific downstream task by providing task-specific inputs. Inspired by the recent success of prompt learning in language and visual tasks, we propose a novel and efficient prompt learning method for HOT tasks, termed Moderate Visual Prompt for HOT (MVP-HOT). Specifically, MVP-HOT freezes the parameters of the pre-trained model and employs HSIs as visual prompts to leverage the knowledge of the underlying RGB model. Additionally, we develop a moderate and effective strategy to incrementally adapt the HSI prompt information. Our proposed method uses only a few (1.7M) learnable parameters and demonstrates its effectiveness through extensive experiments, MVP-HOT can achieve state-of-the-art performance on three hyperspectral datasets.

Fluorescence-based determination of tobramycin using a portable smartphone-based device

The study presents the development of a robust portable smartphone-based device for determining tobramycin in drugs and urine by a fluorescence-based method. As existing analytical methods for the determination of tobramycin often require complex equipment and time-consuming procedures, the proposed approach offers simplicity, portability and real-time monitoring capabilities. The spectrofluorimetric method is based on the reaction of tobramycin and fluorescamine in an alkaline medium. The measurement of analytical signals was enabled with a specially designed device integrating a 3D printed adapter, LEDs, a macro lens, an Arduino-based microcontroller, and asmartphone. The traditional spectrofluorimetry was used as a reference method. Preliminary and optimisation studies were conducted to select the most suitable experimental conditions (e.g. reagent concentration and reaction time) as well as instrumental parameters (e.g. selection of the appropriate RGB model component, ISO, and white balance). The performance of the method was validated, and then verified against conventional spectrofluorimetry, showing comparable accuracy and precision. The method implemented in a smartphone-based device showed a linear range of 0.10–1.00 mgL-1 with a detection limit of 0.03 mgL-1. Synthetic samples were tested to verify analytical usability, followed by analysis of pharmaceutical and urine samples. The method showed good precision (CV<3.4 %) and recovery values (92.3–114.1 %), consistent with spectrofluorimetric results (CV<5.9 %, recovery 92.3–119.8 %). Comparison with other methods highlighted the simplicity and efficiency of the developed approach. The research reveals that it is competitive with existing well-established methods. The proposed device offers reliable results with low reagent and sample consumption, making itsuitable for on-the-go analysis in non-laboratory conditions.

FCS-Net: Feather condition scoring of broilers based on dense feature fusion of RGB and thermal infrared images

Assessing the feather condition of broilers is crucial for monitoring the animal welfare status and detecting the occurrence of feather pecking activities. Currently, the feather condition of individual broilers is manually scored by trained experts. To provide a more objective and efficient tool for feather condition scoring, a novel deep learning-based model, named Feather Condition Scoring Network (FCS-Net), was proposed based on RGB and thermal infrared images. The FCS-Net model combined the ResNet18 architecture with the proposed Dense Feature Fusion (DFF) module, which can effectively learn the feature mapping relationship between RGB and thermal infrared images. Before inputting the images into the network, an image registration process was conducted to align the RGB and thermal infrared images. The results showed that the FCS-Net model had a good performance for feather condition scoring, with the Accuracy of 97.02%, the Precision of 96.99%, the Recall of 97.04%, the F1 of 97.01%, and the Inference speed of 15.34 fps. Compared to the ResNet18_RGB model, which only utilise RGB images, the FCS-Net model showed notable improvements in Accuracy by 4.02%, Precision by 3.90%, Recall by 4.08%, and F1 by 4.01%. Moreover, it was observed that the FCS-Net model focused more on the back region of the broilers through heatmap visualization. Furthermore, the algorithm was compared with six typical image recognition algorithms including VGG16, ResNet18, SE-ResNet18, DenseNet121, Mobilenet_V2, and Shufflenet_V2_x1_0, as well as the state-of-the-art (SOTA) feather condition assessment methods. The results showed that the FCS-Net model achieved better performance than the six algorithms and the SOTA feather condition assessment methods. This study provided a valuable reference for automated monitoring of feather condition scoring of broilers in smart farming.

Spectrofluorimetric and smartphone-based detection methods for determination of gentamicin

This paper presents the development of spectrofluorimetric and smartphone-based detection methods for gentamicin determination using fluorescamine as a reagent. The research included selecting excitation (415 nm) and emission (489 nm) wavelengths, reaction time, and conditions like reagent concentration, and pH of the reaction medium. Moreover, the optimal operating parameters of the smartphone camera, like ISO, white balance, camera shutter, and RGB model channel were selected. Analytical parameters of the developed spectrofluorimetric and smartphone-based methods were estimated including the linear range: 0.04–15.00 mg dm−3 and 0.18–1.20 mg dm−3, respectively, limits of detection and quantification: 0.01 and 0.04 mg dm−3, and 0.06 and 0.18 mg dm−3, respectively, and precision (CV, n = 6): 5.2% and 2.8%, respectively. The proposed approaches were successfully applied to determine gentamicin in pharmaceutical samples. The obtained results were consistent with values declared by manufacturers and satisfactory recovery values, 93.2–113.6% were obtained for both spectrofluorimetric and smartphone-based methods. The developed fluorimetric method with smartphone-based detection provides a low limit of detection specific to spectrofluorimetric methods whereas the measurement system is a simple, easily accessible, compact, and low-cost device. Hence, it can become a competitive alternative to other gentamicin determination methods.Graphical abstract

Optimization and Evaluation of Potentiometric Sensing Membranes for Vildagliptin Determination: A Sustainable Approach Incorporating Whiteness and Greenness Assessment

In the current environmental context, incorporating green analytical chemistry principles into analytical chemistry analysis shows great promise. In drug analysis, the most important objective is to develop techniques that are cost-efficient, eco-friendly, and of high quality. Our study highlights the novelty of utilizing greenness and whiteness approaches in developing a solid-state electrochemical sensor for Vildagliptin (VILD) detection. The sensor employs potentiometric transduction and underwent a dual-phase adjustment process. Initially, various ionophores were evaluated to enhance sensor specificity, followed by the incorporation of a multi-walled carbon nanotube layer as a mediator for ion-to-electron conversion. The selected ionophore was then employed to complete the assessment of VILD. The linearity range obtained was 1.00 × 10−5 M – 1.00 × 10−2 M with a limit of detection of 7.94 × 10−6 M. This sensor demonstrated effectiveness in the selective determination of VILD in bulk powdered material and pharmaceutical formulations. Environmental impact evaluation was performed using the Green Analytical Procedure Index (GAPI) and the Analytical Greenness Calculator (AGREE), comparing it to a previous method. Additionally, The RGB model assessed both method’s whiteness. Our investigation reveals the proposed method as an excellent eco-friendly approach compared to the reported method.

MatUp: Repurposing Image Upsamplers for SVBRDFs

AbstractWe propose MatUp, an upsampling filter for material super‐resolution. Our method takes as input a low‐resolution SVBRDF and upscales its maps so that their rendering under various lighting conditions fits upsampled renderings inferred in the radiance domain with pre‐trained RGB upsamplers. We formulate our local filter as a compact Multilayer Perceptron (MLP), which acts on a small window of the input SVBRDF and is optimized using a data‐fitting loss defined over upsampled radiance at various locations. This optimization is entirely performed at the scale of a single, independent material. Doing so, MatUp leverages the reconstruction capabilities acquired over large collections of natural images by pre‐trained RGB models and provides regularization over self‐similar structures. In particular, our light‐weight neural filter avoids retraining complex architectures from scratch or accessing any large collection of low/high resolution material pairs – which do not actually exist at the scale RGB upsamplers are trained with. As a result, MatUp provides fine and coherent details in the upscaled material maps, as shown in the extensive evaluation we provide.

Utilizing two different sustainable and green spectrophotometric approaches using derivative ratio spectra for the determination of a ternary severely overlapped mixture: application to veterinary formulation

Mathematical manipulation technique has proven to be a very powerful tool for efficient processing and handling of highly overlapped spectra. This work introduced two green and sustainable approaches for the successful recovery of the ternary mixture, Tylosin tartarate (TYL), Sulfadimdine (SLD), and Trimethoprim (TRI). The approaches were constructed using three different methods, derivative ratio spectrum zero-crossing method (DRSZ), double divisor ratio spectra derivative method (DDRD), and factorized derivative ratio method coupled with spectrum subtraction (FDRM-SS). The two approaches succeeded in recovering the three drugs (Linearity range achieved were 0.5–5 µg/mL for TYL, 0.3–1.3 µg/mL for SLD, and 0.3–5 µg/mL for TRI), giving convenient standard deviations and satisfactory recovery percentages. The recommended methods have been verified in accordance with (ICH) guidelines. When the results were statistically compared to the official methods, no significant difference was found. Both AGREE—Analytical GREEnness Metric Approach and Software, and White Analytical Chemistry (WAC) RGB model gave scores of 0.93 and 97.2%, respectively, which proved that the approaches were eco-friendly and abiding by the sustainability principles.

Glaucoma Detection through a Novel Hyperspectral Imaging Band Selection and Vision Transformer Integration.

Conventional diagnostic methods for glaucoma primarily rely on non-dynamic fundus images and often analyze features such as the optic cup-to-disc ratio and abnormalities in specific retinal locations like the macula and fovea. However, hyperspectral imaging techniques focus on detecting alterations in oxygen saturation within retinal vessels, offering a potentially more comprehensive approach to diagnosis. This study explores the diagnostic potential of hyperspectral imaging for glaucoma by introducing a novel hyperspectral imaging conversion technique. Digital fundus images are transformed into hyperspectral representations, allowing for a detailed analysis of spectral variations. Spectral regions exhibiting differences are identified through spectral analysis, and images are reconstructed from these specific regions. The Vision Transformer (ViT) algorithm is then employed for classification and comparison across selected spectral bands. Fundus images are used to identify differences in lesions, utilizing a dataset of 1291 images. This study evaluates the classification performance of models using various spectral bands, revealing that the 610-780 nm band outperforms others with an accuracy, precision, recall, F1-score, and AUC-ROC all approximately at 0.9007, indicating its superior effectiveness for the task. The RGB model also shows strong performance, while other bands exhibit lower recall and overall metrics. This research highlights the disparities between machine learning algorithms and traditional clinical approaches in fundus image analysis. The findings suggest that hyperspectral imaging, coupled with advanced computational techniques such as the ViT algorithm, could significantly enhance glaucoma diagnosis. This understanding offers insights into the potential transformation of glaucoma diagnostics through the integration of hyperspectral imaging and innovative computational methodologies.

Effects of different ground segmentation methods on the accuracy of UAV-based canopy volume measurements.

The nonuniform distribution of fruit tree canopies in space poses a challenge for precision management. In recent years, with the development of Structure from Motion (SFM) technology, unmanned aerial vehicle (UAV) remote sensing has been widely used to measure canopy features in orchards to balance efficiency and accuracy. A pipeline of canopy volume measurement based on UAV remote sensing was developed, in which RGB and digital surface model (DSM) orthophotos were constructed from captured RGB images, and then the canopy was segmented using U-Net, OTSU, and RANSAC methods, and the volume was calculated. The accuracy of the segmentation and the canopy volume measurement were compared. The results show that the U-Net trained with RGB and DSM achieves the best accuracy in the segmentation task, with mean intersection of concatenation (MIoU) of 84.75% and mean pixel accuracy (MPA) of 92.58%. However, in the canopy volume estimation task, the U-Net trained with DSM only achieved the best accuracy with Root mean square error (RMSE) of 0.410 m3, relative root mean square error (rRMSE) of 6.40%, and mean absolute percentage error (MAPE) of 4.74%. The deep learning-based segmentation method achieved higher accuracy in both the segmentation task and the canopy volume measurement task. For canopy volumes up to 7.50 m3, OTSU and RANSAC achieve an RMSE of 0.521 m3 and 0.580 m3, respectively. Therefore, in the case of manually labeled datasets, the use of U-Net to segment the canopy region can achieve higher accuracy of canopy volume measurement. If it is difficult to cover the cost of data labeling, ground segmentation using partitioned OTSU can yield more accurate canopy volumes than RANSAC.

Water pollution classification and detection by hyperspectral imaging.

This study utilizes spectral analysis to quantify water pollutants by analyzing the images of biological oxygen demand (BOD). In this study, a total of 2545 images depicting water quality pollution were generated due to the absence of a standardized water pollution detection method. A novel snap-shot hyperspectral imaging (HSI) conversion algorithm has been developed to conduct spectral analysis on traditional RGB images. In order to demonstrate the effectiveness of the developed HSI algorithm, two distinct three-dimensional convolution neural networks (3D-CNN) are employed to train two separate datasets. One dataset is based on the HSI conversion algorithm (HSI-3DCNN), while the other dataset is the traditional RGB dataset (RGB-3DCNN). The images depicting water quality pollution were categorized into three distinct groups: Good, Normal, and Severe, based on the extent of pollution severity. A comparison was conducted between the HSI and RGB models, focusing on precision, recall, F1-score, and accuracy. The water pollution model's accuracy improved from 76% to 80% when the RGB-3DCNN was substituted with the HSI-3DCNN. The results suggest that the HSI has the capacity to enhance the effectiveness of water pollution detection compared to the RGB model.

Chemical nature evolution of solid supports used in electromembrane extraction procedures: A comparative analysis based on metric tools

BackgroundIn recent decades, green chemistry has been focusing on the adaptation of different chemical methods towards environmental friendliness. Sample preparation procedures, which constitute a fundamental step in analytical methodology, have also been modified and implemented in this direction. In particular, electromembrane extraction (EME) procedures, which have traditionally used plastic supports, have been optimized towards greener approaches through the emergence of alternative materials. In this regard, biopolymer-based membranes (such as agarose or chitosan) have become versatile and very promising substitutes to perform these processes. ResultsDifferent green metric tools (Analytical Eco-Scale, ComplexGAPI and AGREEprep have been applied to study the evolution of solid supports used in EME from nanostructured tissues and polymer inclusion membranes to agar films and chitosan flat membranes. The main goal is to evaluate the usage of these new biomaterials in the analytical procedure to quantify their environmental impact in the frame of Green Analytical Chemistry (GAC). In addition, both RGB model and BAGI metrics have been employed to study the sustainability of the whole procedure, including not only greenness, but also analytical performance and feasibility aspects. Results obtained after the performance of the mentioned metrics have demonstrated that the most efficient and environmentally friendly analytical methods are based on the use of chitosan supports. This improvement is mainly due to the chemical nature of this biopolymer as well as to the removal of organic solvents. SignificanceThis work highlights the advantages of biodegradable materials employment in EME procedures to achieve green analytical methodologies. These materials also contribute to raise the figure of merits regarding to the quantification parameters in a wide range of applications compared to classical supports employed in EME, thus enhancing sustainability of procedures.

Software and hardware complex based on the STM32F407VG microcontroller for the study of vibra-tions with the LIS3DSH accelerometer

Commonly used color models are reviewed and analyzed, namely the RGB model and subtractive models, including CMY and CMYK. The characteristics of colorimeters based on the use of RGB, CMY and CMYK models have been studied. A comparative analysis of the advantages and disadvantages of RGB and CMYK models was carried out. The developed digital colorimeter allows for express control of color characteristics with high metrological accuracy and does not require a high level of qualification of service personnel. This device reproduces the perception of color by human vision, and its use is intended for both people with normal and impaired vision. The great advantage of this device is its mobility, the speed of measurements and the absence of the need for a specialized laboratory and highly qualified specialists. Full-scale measurements of the output voltage on the photodiodes of the colorimeter corresponding to red, blue and green colors were carried out.

RGBfast – A user-friendly version of the Red-Green-Blue model for assessing greenness and whiteness of analytical methods

A new version of the popular RGB model has been developed to meet the expectations of users of greenness and whiteness assessment metrics. It has been called “RGBfast” due to its simplicity and significant assessment automation. The criteria were limited to six key parameters that are easy to objectively express numerically, which combine various features of analytical method determining its functionality and sustainability. A customized Excel sheet applies the entire procedure itself after entering the appropriate input data. The assessment outcomes are presented in concise and easy-to-interpret tables, which can be used as pictograms. RGBfast enables reliable comparison of alternative procedures dedicated to the same purpose, thus can be a good support during the validation of new methods and literature reviews. As a case study, the model was applied to compare five methods for the determination of phosphate using different types of (bio)sensors. Subtle differences between them have been captured and analyzed. The mathematical structure of the model can be modified in the future, for example to take potential opportunities offered by artificial intelligence technology.

Self-Supervised 3-D Semantic Representation Learning for Vision-and-Language Navigation.

In vision-and-language navigation (VLN) tasks, most current methods primarily utilize RGB images, overlooking the rich 3-D semantic data inherent to environments. To rectify this, we introduce a novel VLN framework that integrates 3-D semantic information into the navigation process. Our approach features a self-supervised training scheme that incorporates voxel-level 3-D semantic reconstruction to create a detailed 3-D semantic representation. A key component of this framework is a pretext task focused on region queries, which determines the presence of objects in specific 3-D areas. Following this, we devise an long short-term memory (LSTM)-based navigation model that is trained using our 3-D semantic representations. To maximize the utility of these 3-D semantic representations, we implement a cross-modal distillation strategy. This strategy encourages the RGB model's outputs to emulate those from the 3-D semantic feature network, enabling the concurrent training of both branches to merge RGB and 3-D semantic data effectively. Comprehensive evaluations on both the R2R and R4R datasets reveal that our method significantly enhances performance in VLN tasks.

Quantitative analysis of oxygen heterocyclic compounds using liquid chromatography coupled to tandem mass spectrometry, method development and environmental assessment

The latest trends in the field of laboratory method development concern the validation of fast procedures, and the safety of the operator and environment. The present research is focused on the development of fast and environmentally-friendly analytical method of liquid chromatography-triple quadrupole mass spectrometry for the determination of oxygen heterocyclic compounds in Citrus-flavored products. To the best of the authors’ knowledge, this is the first approach for the characterization of thirty-six oxygen heterocyclic compounds in less than four minutes using water/ethanol as mobile phase. Molecules were characterized in Citrus essential oils and beverages. Figures of merit determined were linearity, precision, limits of detection and quantification. The method was compared in terms of greenness with our two methodologies previously published for analysis of oxygen heterocyclic compounds. The analytical greenness metric approach and the “RGB model” were used. According to these approaches, the method proposed here is faster, safer and more eco-friendly.

Integrated approach of white analytical chemistry and design of experiments to microwave-assisted sensitive and eco-friendly spectrofluorimetric estimation of mirabegron using 4-chloro-7-nitrobezofuran as biosensing fluorescent probe

The USFDA recently approved mirabegron, a novel once-daily β-3 adrenoceptor agonist for oral administration, as a transformative treatment for overactive bladder. Despite the existence of numerous analytical methods for the assay and bioanalysis of mirabegron, it’s perplexing that none have explored the domain of microwave-assisted sensitive spectrofluorimetric method for mirabegron estimation, even after extensive literature review. Adding to the enigma is the insistence of current analytical methods on using expensive and harmful organic solvents, posing a threat to marine life and the broader environment. Recently, the white analytical chemistry approach has been introduced to develop analytical methods that are cost-effective, environmentally friendly, and user-friendly. Consequently, a white analytical chemistry-based, sensitive, and eco-friendly spectrofluorimetric estimation of mirabegron has been initiated, using 4-Chloro-7-nitrobenzofurazan as a fluorescent biosensing probe. The development of this robust method involved a series of experiments designed to minimize solvent and time wastage. Through a combination of fractional factorial and Box-Behnken designs, researchers identified the critical variables and optimized the method to perfection. This method was validated according to the stringent ICH Q2 (R2) and USFDA guidelines, ensuring its reliability and accuracy. Once approved, this sensitive spectrofluorimetric method was tested, accurately estimating mirabegron levels in commercial formulations and rat plasma samples. To further enrich the study, a comprehensive evaluation of existing analytical methods was conducted alongside the proposed spectrofluorimetric method, using advanced tools like the AGREE calculator, GAPI software, and RGB model to assess their eco-friendliness and effectiveness in mirabegron estimation.

Self-calibrating colorimetric sensor assisted deep learning strategy for efficient and precise Fe(II) detection

Colorimetry is to determine the component content usually comparing or measuring the “RGB” values, which has the advantages of quick response and visual comparison. However, the widely reported “RGB” models are limited by requiring professional operation such as accurate sampling, insufficient information, cumbersome calculation, etc., which has significantly restricted the promotional application. Inspired by the PAA-PVA gel with excellent absorbing-swelling ability and superior response ability, indefinite volume addition has a significant difference in the swelling phenomenon of PAA-PVA hydrogel, the extent of this enlargement is directly related to the amount of testing sample absorbed, which can be applied to reflect testing sample volume added. In addition, benefited from a comprehensive and precise parameter network, the deep learning (DL) algorithm solves the defects such as ambient light effect and manual extract RGB values of the colorimetric sensing analysis strategy. Considering Fe(II) is an essential element for life science and environmental monitoring, we have developed a self-calibrating colorimetric sensor assisted deep learning strategy for the Fe(II) detection with indefinite volume. It can directly complete the analysis of Fe(II) concentration via captured images, without going through a complicated process similar to the RGB model. Specifically, we integrated the advantages of colorimetric sensing strategy, absorbing-swelling hydrogel, and DL algorithm to develop a phenanthroline (PHEN) indicator doped PAA-PVA colorimetric sensor (PHEN-PAA-PVA). Then, 2,500 colorimetric response images were collected as the training dataset and assessed 2 DL algorithms and 7 machine learning (ML) algorithms. The convolutional neural network (CNN) model exhibited the most efficiency to quantify 1 × 10−5–9 × 10−4 mol/L Fe(II), with a limit of detection (LOD) of 3.3 × 10−6 mol/L. It provides a promising and blooming in-situ detection approach for Fe(II) analysis, contributes to health management and environment monitoring.

A guided-based approach for deepfake detection: RGB-depth integration via features fusion

Deep fake technology paves the way for a new generation of super realistic artificial content. While this opens the door to extraordinary new applications, the malicious use of deepfakes allows for far more realistic disinformation attacks than ever before. In this paper, we start from the intuition that generating fake content introduces possible inconsistencies in the depth of the generated images. This extra information provides valuable spatial and semantic cues that can reveal inconsistencies facial generative methods introduce. To test this idea, we evaluate different strategies for integrating depth information into an RGB detector and we propose an attention mechanism that makes it possible to integrate information from depth effectively. In addition to being more accurate than an RGB model, our Masked Depthfake Network method is +3.2% more robust against common adversarial attacks on average than a typical RGB detector. Furthermore, we show how this technique allows the model to learn more discriminative features than RGB alone.