Spatial Image Research Articles

Abstract 1430 MALDI Spatial Imaging to Characterize the Competition for Metabolic Resources During Parasitoid Wasp Infection of Fly Larvae

Abstract 1430 MALDI Spatial Imaging to Characterize the Competition for Metabolic Resources During Parasitoid Wasp Infection of Fly Larvae

Combining visual interpretation and image segmentation to derive canopy cover index from high resolution satellite imagery in functionally diverse coniferous forests

Forest canopy cover is one of the most significant structural parameters of the forest stand that can be estimated using of aerial and satellite remote sensing. Even though sub-pixel analysis can be used to estimate the index on low-resolution imagery, high-resolution imagery provides more accurate details on forest canopy variability for ecological and forestry applications. However, the high variability of the images demands a more advanced approach to canopy cover measurement than the visual interpretation of single images or stereo pairs. These traditional methods are inefficient and limited in providing a comprehensive and accurate canopy cover assessment. An improvement of the method could involve classifying high spatial resolution images, separating and extracting the areas corresponding to canopy gaps, and generating canopy cover maps. This study offers valuable insights and reveals key differences between three methods for estimating canopy cover: ground measurements, visual photo interpretation and automatic extraction from classified images using pixel and object-based methods. The texture analysis approach was used for separating the “shadow” objects corresponding to gaps in the canopy from the shades cast on the lower trees. The sample plot-based visual interpretation of the images revealed comparable results between ground and satellite image canopy cover values (correlation coefficient of 0.74 for all plots), with lower correlations (r = 0.39) for uneven-aged stands. The results encourage the use of the texture analysis method, with satisfying accuracy (forest canopy cover differences of maximum 0.06 between ground, photo interpreted and extracted datasets). The method could be further integrated with complementary data like LIDAR or hyperspectral images.

Spatial Transformation Accelerator with Parallel Data Access Scheme for Sample Reconstruction

Spatial image transformation is a commonly used component in many image processing pipelines. It enables the correction of optical distortions, image registration onto a common reference plane, electronic image stabilisation, digital zoom, video mosaicking, etc. With the growing tendency to embed image processing in low-power devices, attaining an efficient transformation solution becomes increasingly decisive. Furthermore, interpolation is the key operation in achieving the high quality of the transformed data from the original data. Fortunately, different implementations have already seen several efficiency improvements in recent years. However, interpolation relies on sampling a set of neighbouring points from memory, which has yet to be addressed efficiently for smaller computational platforms with limited memory resources. In this work, we derive a generic mathematical model and circuit design principles for the spatial transformation accelerator design for N-dimensional data. Furthermore, we present an efficient simultaneous access scheme for high-quality signal reconstruction. Finally, the introduced ideas are verified in field programmable gate arrays using one-dimensional and two-dimensional data transformation use cases. The presented solution is able to transform images with sizes ranging from 256 × 256 to 8192 × 8192 and achieves a transfer rate of 275 frames per second with 512 × 512 images.

Comparison of single and double pulse laser-induced breakdown spectroscopy for the detection of biomolecules tagged with photon-upconversion nanoparticles

BackgroundLaser-induced breakdown spectroscopy (LIBS) is a well-recognized analytical technique used for elemental analysis. This method is gaining considerable attention also in biological applications thanks to its ability for spatial mapping and elemental imaging. The implementation of LIBS in the biomedical field is based on the detection of metals or other elements that either naturally occur in the samples or are present artificially. The artificial implementation of nanoparticle labels (Tag-LIBS) enables the use of LIBS as a readout technique for immunochemical assays. However, one of the biggest challenges for LIBS to meet immunoassay readout standards is its sensitivity. ResultsThis paper focuses on the improvement of LIBS sensitivity for the readout of nanoparticle-based immunoassays. First, the LIBS setup was optimized on photon-upconversion nanoparticle (UCNP) droplets deposited on the microtiter plate wells. Two collection optics systems were compared, with single pulse (SP) and collinear double pulse (DP) LIBS arrangements. By deploying the second laser pulse, the sensitivity was improved up to 30 times. The optimized SP and DP setups were then employed for the indirect detection of human serum albumin based on immunoassay with UCNP-based labels. Compared to our previous LIBS study, the detection limit was enhanced by two orders of magnitude, from 10 ng mL−1 to 0.29 ng mL−1. In addition, two other immunochemical methods were used for reference, based on the readout of upconversion luminescence of UCNPs and absorbance measurement with enzyme labels. Finally, the selectivity of the assay was tested and the practical potential of Tag-LIBS was demonstrated by the successful analysis of urine samples. Significance and noveltyIn this work, we improved the sensitivity of the Tag-LIBS method by combining new labels based on UCNPs with the improved collection optics and collinear DP configuration. In the instrumental setup optimization, the DP LIBS showed better sensitivity and signal-to-noise ratio than SP. The optimizations allowed the LIBS readout to surpass the sensitivity of enzyme immunoassay, approaching the qualities of upconversion luminescence readout, which is nowadays a state-of-the-art readout technique.

Crop mapping in smallholder farms using unmanned aerial vehicle imagery and geospatial cloud computing infrastructure

Smallholder farms are major contributors to agricultural production, food security, and socio-economic growth in many developing countries. However, they generally lack the resources to fully maximize their potential. Subsequently they require innovative, evidence-based and lower-cost solutions to optimize their productivity. Recently, precision agricultural practices facilitated by unmanned aerial vehicles (UAVs) have gained traction in the agricultural sector and have great potential for smallholder farm applications. Furthermore, advances in geospatial cloud computing have opened new and exciting possibilities in the remote sensing arena. In light of these recent developments, the focus of this study was to explore and demonstrate the utility of using the advanced image processing capabilities of the Google Earth Engine (GEE) geospatial cloud computing platform to process and analyse a very high spatial resolution multispectral UAV image for mapping land use land cover (LULC) within smallholder farms. The results showed that LULC could be mapped at a 0.50 m spatial resolution with an overall accuracy of 91%. Overall, we found GEE to be an extremely useful platform for conducting advanced image analysis on UAV imagery and rapid communication of results. Notwithstanding the limitations of the study, the findings presented herein are quite promising and clearly demonstrate how modern agricultural practices can be implemented to facilitate improved agricultural management in smallholder farmers.

Extraction of Coastal Levees Using U-Net Model with Visible and Topographic Images Observed by High-Resolution Satellite Sensors.

Coastal levees play a role in protecting coastal areas from storm surges and high waves, and they provide important input information for inundation damage simulations. However, coastal levee data with uniformity and sufficient accuracy for inundation simulations are not always well developed. Against this background, this study proposed a method to extract coastal levees by inputting high spatial resolution optical satellite image products (RGB images, digital surface models (DSMs), and slope images that can be generated from DSM images), which have high data availability at the locations and times required for simulation, into a deep learning model. The model is based on U-Net, and post-processing for noise removal was introduced to further improve its accuracy. We also proposed a method to calculate levee height using a local maximum filter by giving DSM values to the extracted levee pixels. The validation was conducted in the coastal area of Ibaraki Prefecture in Japan as a test area. The levee mask images for training were manually created by combining these data with satellite images and Google Street View, because the levee GIS data created by the Ibaraki Prefectural Government were incomplete in some parts. First, the deep learning models were compared and evaluated, and it was shown that U-Net was more accurate than Pix2Pix and BBS-Net in identifying levees. Next, three cases of input images were evaluated: (Case 1) RGB image only, (Case 2) RGB and DSM images, and (Case 3) RGB, DSM, and slope images. Case 3 was found to be the most accurate, with an average Matthews correlation coefficient of 0.674. The effectiveness of noise removal post-processing was also demonstrated. In addition, an example of the calculation of levee heights was presented and evaluated for validity. In conclusion, this method was shown to be effective in extracting coastal levees. The evaluation of generalizability and use in actual inundation simulations are future tasks.

A general and simple automated impervious surface mapping approach based on three-dimensional texture features (3DTF) using fine spatial resolution remotely sensed imagery

The mapping of impervious surfaces using remote sensing techniques offer essential technical support for sustainable development objectives and safeguard the environment. In this study, we developed an automated method without training samples for mapping impervious surfaces using texture features. The different aggregated impervious surface patterns and distributions in study areas of Site A-C in China (Beijing, Huainan, Jinhua) were considered. The Site D-E in Dubai and Tehran, surrounded with deserts in arid areas. They were selected to develop and evaluate the performance of the proposed automated method. The texture features of the Contrast, Gabor wavelets, and secondary texture extraction (Con_Gabor) derived from Sentinel-2 images at each site were used to construct the three-dimensional texture features (3DTF) of impervious surfaces. The 3DTF-combined K-means classifier was used to automatically map the impervious surfaces. The results showed that the overall accuracies of mapping impervious surface were 91.15 %, 89.75 %, and 91.90 % in Site A-C. The overall accuracies of mapping impervious surface were 90.95 %, 91.45 % and 88.23 % in rural areas. The distributions of impervious surface on automated method, GHS-BUILT-S and ESA WorldCover were similar in study areas. The automated method for mapping impervious surfaces performed as well as the artificial neural network (ANN) and Random Forest (RF), and the advantage of not requiring training samples. The automated method was tested in the in Dubai and Tehran. The overall accuracies of the automatic method for mapping impervious surfaces >89 % at Site D-E, and >88 % at rural area. In addition, the 3DTF was proved as the simplest and most effective feature combination to map impervious surfaces. The impervious surface mapped using the automated method was robust across bands, seasons and sensors. However, further evaluation is necessary to assess the effectiveness of automated methods using high spatial resolution images for mapping impervious surface in complex areas.

Development of pH‐Sensitive Film for Detection of Implant Infection via Ultrasound Luminescent Chemical Imaging

AbstractA new hybrid ultrasound luminescent chemical imaging technique is described along with a pH sensor to image chemical concentrations at the surface of implanted medical devices. The purpose is to detect and study local biochemistry during infection. The sensor comprises a mechanoluminescent film (SrAl2O4:Eu, Dy microphosphors embedded in a biocompatible polymer film) and a pH indicator dye. A focused ultrasound beam generates green luminescence at the ultrasound focal point. By pulsing the ultrasound ON and OFF, the modulated luminescence can be distinguished from persistent luminescence, for high spatial resolution imaging. A red fluorescent dye and the pH indicator dye bromothymol blue are added to the coating to modulate the red‐light transmittance via pH dependent absorbance. Acidosis is observed as an increase in red luminescence intensity in spectroscopy and imaging. The films are sensitive to biologically relevant changes in pH (6.0–8.0) and can be imaged through optically scattering media to mimic tissue. The images have a knife edge spatial resolution of ≈3 mm through optically scattering phantoms, limited by the focused ultrasound spot size. This novel technique may permit the elucidation of implant infection at the implant surface and can be further developed for the measurement of other relevant chemical species in the future.

Temporal and Spatial Urban Heat Islands in a Coastal Brazilian Area of Tropical Climate

This study explores the urban heat island (UHI) phenomenon in the Metropolitan Region of Vitória (MRV), Brazil. Two methods were employed: the traditional approach estimating the Atmospheric Urban Heat Island (UHI) using air temperature data, and the quantiles method assessing the Surface Urban Heat Island (SUHI) through medium spatial resolution images. The findings demonstrate the existence of an average heat island effect, as estimated by both UHI (+0.9 °C) and SUHI (+2.03 °C). The intensity of the atmospheric UHI is linked with higher air temperature values around midday. Additionally, the study reveals that UHI values in the MRV are more pronounced during spring and summer SUHI estimates peak during the summer months. The utilization of orbital remote sensing enables the identification of hotspots throughout the metropolitan area, with the highest temperatures observed in industrial areas, suburban neighborhoods, and densely populated regions. The advantage of the UHI method lies in its long time series, providing high temporal resolution and statistical significance over an extended period. Conversely, the SUHI method offers the advantage of considering data across the entire domain with high spatial resolution. Thus, these two methods complement each other in identifying the heat island phenomenon.

Detection Method and Application of Nuclear-Shaped Anomaly Areas in Spatial Electric Field Power Spectrum Images

It is found that there are some anomalous high-energy nuclear-shaped regions in the VLF frequency band of the space electric field. To detect and statistically analyze these nuclear-shaped anomaly areas, this paper proposes a nuclear-shaped anomaly area detection method based on the electric field power spectrum image data of the China Seismo Electromagnetic Satellite (CSES-01). First, the logarithm of VLF frequency band data was calculated and rotated counterclockwise to create power spectrum images and label them to form a sample image dataset; then, images were enhanced (which involved resizing, scaling, rotation, gaussian denoising, etc.) to solve the problems of the model overfitting and sample imbalance. Finally, the U-net network model based on the ResNet50 encoder was trained to obtain the optimal kernel anomaly detection model ResNet50_Unet. Comparative experiments with various semantic segmentation algorithms show that the ResNet50_Unet model has the best performance. Applying this model to detect the electric field power spectrum images from November 2021 to February 2022, a total of 101 nuclear-shaped anomaly areas were found, distributed between 45° and 70° of the north–south latitude. This model can quickly detect nuclear-shaped anomaly regions from massive data, providing reference significance for the detection of other types of ionospheric spatial disturbances. At the same time, it has important scientific significance and practical value for understanding the ionosphere and space communication.

Deep Learning Approach to Improve Spatial Resolution of GOES-17 Wildfire Boundaries Using VIIRS Satellite Data

The rising severity and frequency of wildfires in recent years in the United States have raised numerous concerns regarding the improvement in wildfire emergency response management and decision-making systems, which require operational high temporal and spatial resolution monitoring capabilities. Satellites are one of the tools that can be used for wildfire monitoring. However, none of the currently available satellite systems provide both high temporal and spatial resolution. For example, GOES-17 geostationary satellite fire products have high temporal (1–5 min) but low spatial resolution (≥2 km), and VIIRS polar orbiter satellite fire products have low temporal (~12 h) but high spatial resolution (375 m). This work aims to leverage currently available satellite data sources, such as GOES and VIIRS, along with deep learning (DL) advances to achieve an operational high-resolution, both spatially and temporarily, wildfire monitoring tool. Specifically, this study considers the problem of increasing the spatial resolution of high temporal but low spatial resolution GOES-17 data products using low temporal but high spatial resolution VIIRS data products. The main idea is using an Autoencoder DL model to learn how to map GOES-17 geostationary low spatial resolution satellite images to VIIRS polar orbiter high spatial resolution satellite images. In this context, several loss functions and DL architectures are implemented and tested to predict both the fire area and the corresponding brightness temperature. These models are trained and tested on wildfire sites from 2019 to 2021 in the western U.S. The results indicate that DL models can improve the spatial resolution of GOES-17 images, leading to images that mimic the spatial resolution of VIIRS images. Combined with GOES-17 higher temporal resolution, the DL model can provide high-resolution near-real-time wildfire monitoring capability as well as semi-continuous wildfire progression maps.

DaliWS: A High-Resolution Dataset with Precise Annotations for Water Segmentation in Synthetic Aperture Radar Images

The frequent occurrence of global flood disasters leads to millions of people falling into poverty each year, which poses immense pressure on governments and hinders social development. Therefore, providing more data support for flood disaster detection is of paramount importance. To facilitate the development of water body detection algorithms, we create the DaliWS dataset for water segmentation, which contains abundant pixel-level annotations, and consists of high spatial resolution SAR images collected from the GaoFen-3 (GF-3) satellite. For comprehensive analysis, extensive experiments are conducted on the DaliWS dataset to explore the performance of the state-of-the-art segmentation models, including FCN, SegNeXt, U-Net, and DeeplabV3+, and investigate the impact of different polarization modes on water segmentation. Additionally, to probe the generalization of our dataset, we further evaluate the models trained with the DaliWS dataset, on publicly available water segmentation datasets. Through detailed analysis of the experimental results, we establish a valuable benchmark and provide usage guidelines for future researchers working with the DaliWS dataset. The experimental results demonstrate the F1 scores of FCN, SegNeXt, U-Net, and DeeplabV3+ on the dual-polarization data of DaliWS dataset reach to 90.361%, 90.192%, 92.110%, and 91.199%, respectively, and these four models trained using the DaliWS dataset exhibit excellent generalization performance on the public dataset, which further confirms the research value of our dataset.

Advanced spectroscopic investigation of colour centres in LiF crystals irradiated with monochromatic hard x-rays

Nominally-pure lithium fluoride (LiF) crystals were irradiated with monochromatic hard x-rays of energy 5, 7, 9 and 12 keV at the METROLOGIE beamline of the SOLEIL synchrotron facility, in order to understand the role of the selected x-ray energy on their visible photoluminescence (PL) response, which is used for high spatial resolution 2D x-ray imaging detectors characterized by a wide dynamic range. At the energies of 7 and 12 keV the irradiations were performed at five different doses corresponding to five uniformly irradiated areas, while at 5 and 9 keV only two irradiations at two different doses were carried out. The doses were planned in a range between 4 and 1.4 × 103 Gy (10.5 mJ cm−3 to 3.7 J cm−3), depending on the x-ray energy. After irradiation at the energies of 7 and 12 keV, the spectrally-integrated visible PL intensity of the F2 and F3 + colour centres (CCs) generated in the LiF crystals, carefully measured by fluorescence microscopy under blue excitation, exhibits a linear dependence on the irradiation dose in the investigated dose range. This linear behaviour was confirmed by the optical absorption spectra of the irradiated spots, which shows a similar linear behaviour for both the F2 and F3 + CCs, as derived from their overlapping absorption band at around 450 nm. At the highest x-ray energy, the average concentrations of the radiation-induced F, F2 and F3 + CCs were also estimated. The volume distributions of F2 defects in the crystals irradiated with 5 and 9 keV x-rays were reconstructed in 3D by measuring their PL signal using a confocal laser scanning microscope operating in fluorescence mode. On-going investigations are focusing on the results obtained through this z-scanning technique to explore the potential impact of absorption effects at the excitation laser wavelength.

Thin slice photon-counting CT coronary angiography compared to conventional CT: Objective image quality and clinical radiation dose assessment.

Photon-counting CT (PCCT) is the next-generation CT scanner that enables improved spatial resolution and spectral imaging. For full spectral processing, higher tube voltages compared to conventional CT are necessary to achieve the required spectral separation. This generated interest in the potential influence of thin slice high tube voltage PCCT on overall image quality and consequently on radiation dose. This study first evaluated tube voltages and radiation doses applied in patients who underwent coronary CT angiography with PCCT and energy-integrating detector CT (EID-CT). Next, image quality of PCCT and EID-CT was objectively evaluated in a phantom study simulating different patient sizes at these tube voltages and radiation doses. We conducted a retrospective analysis of clinical doses of patients scanned on a conventional and PCCT system. Average patient water equivalent diameters for different tube voltages were extracted from the dose reports for both EID-CT and PCCT. A conical phantom made of polyethylene with multiple diameters (26/31/36cm) representing different patient sizes and containing an iodine insert was scanned with a EID-CT scanner using tube voltages and phantom diameters that match the patient scans and characteristics. Next, phantom scans were made with PCCT at a fixed tube voltage of 120kV and with CTDIVOL values and phantom diameters identical to the EID-CT scans. Clinical image reconstructions at 0.6mm slice thickness for conventional CT were compared to PCCT images with 0.4mm slice thickness. Image quality was quantified using the detectability index (d'), which estimated the visibility of a 3mm diameter contrast-enhanced coronary artery by considering noise, contrast, resolution, and human visual perception. Alongside d', noise, contrast and resolution were also individually assessed. In addition, the influence of various kernels (Bv40/Bv44/Bv48/Bv56), quantum iterative reconstruction strengths (QIR, 3/4) and monoenergetic levels (40/45/50/55keV) for PCCT on d' was investigated. In this study, 143 patients were included: 47 were scanned on PCCT (120kV) and the remaining on EID-CT (74 small-sized at 70kV, 18 medium-sized at 80kV and four large-sized at 90kV). EID-CT showed 7%-17% higher d' than PCCT with Bv40 kernel and strength four for small/medium patients. Lower monoenergetic images (40keV) helped mitigate the difference to 1%-6%. For large patients, PCCT's detectability was up to 31% higher than EID-CT. PCCT has thinner slices but similar noise levels for similar reconstruction parameters. The noise increased with lower keV levels in PCCT (≈30% increase), but higher QIR strengths reduced noise. PCCT's iodine contrast was stable across patient sizes, while EID-CT had 33% less contrast in large patients than in small-sized patients. At 120kV, thin slice PCCT enables CCTA in phantom scans representing large patients without raising radiation dose or affecting vessel detectability. However, higher doses are needed for small and medium-sized patients to obtain a similar image quality as in EID-CT. The alternative of using lower mono-energetic levels requires further evaluation in clinical practice.

Multi-Temporal Sentinel-1 and Sentinel-2 Data for Orchards Discrimination in Khairpur District, Pakistan Using Spectral Separability Analysis and Machine Learning Classification

Generating orchards spatial distribution maps within a heterogeneous landscape is challenging and requires fine spatial and temporal resolution images. This study examines the effectiveness of Sentinel-1 (S1) and Sentinel-2 (S2) satellite data of relatively high spatial and temporal resolutions for discriminating major orchards in the Khairpur district of the Sindh province, Pakistan using machine learning methods such as random forest (RF) and a support vector machine. A Multicollinearity test (MCT) was performed among the multi-temporal S1 and S2 variables to remove those with high correlations. Six different feature combination schemes were tested, with the fusion of multi-temporal S1 and S2 (scheme-6) outperforming all other combination schemes. The spectral separability between orchards pairs was assessed using Jeffries-Matusita (JM) distance, revealing that orchard pairs were completely separable in the multi-temporal fusion of both sensors, especially the indistinguishable pair of dates-mango. The performance difference between RF and SVM was not significant, SVM showed a slightly higher accuracy, except for scheme-4 where RF performed better. This study concludes that multi-temporal fusion of S1 and S2 data, coupled with robust ML methods, offers a reliable approach for orchard classification. Prospectively, these findings will be helpful for orchard monitoring, improvement of yield estimation and precision based agricultural practices.

Multi-branch residual image semantic segmentation combined with inverse weight gated-control

The loss of pixel-level information in the multi-class segmentation task based on the U-net model results in unclear boundaries and low semantic segmentation accuracy. Aiming at this, a deep multi-branch residual Unet (IWG-MRUN) with fused inverse weight gated-control is proposed to improve the quality of image semantic segmentation. Specifically, we first introduce a deep multi-branch residual module, which used parallel convolution mode to capture the contextual feature to extract the detailed features of the input image at a deeper level. Then, we adopt an inverse weight gated-control module to enhance the diversity of up-sampling information by counterclockwise transmitting attention horizontally to improve the restoration accuracy of up-sampled image pixels. Finally, to obtain finer granularity features from low spatial resolution images, we adopt the different receptive field pyramid attention mechanisms at the highest level of the U-shaped encoder to capture high-level context information at different scales, thereby improving the accuracy of semantic segmentation. The experimental results show that the segmentation accuracy of the proposed algorithm reaches 91.80% and the CCE loss is reduced to 0.21. When compared to the Unet, BiSeNet, DeeplabV3+ and U-net + BLR model, the pixel accuracy of semantic segmentation is improved by 15.0%, 1.98%, 0.9% and 6.5%, respectively. The semantic segmentation model proposed in this paper provides an end-to-end semantic segmentation capability with the enriched finer granularity features of the target boundary and realizes the accurate segmentation of the objects in different categories.

Semantic segmentation of very-high spatial resolution satellite images: A comparative analysis of 3D-CNN and traditional machine learning algorithms for automatic vineyard detection

The Erzincan (Cimin) grape, which is an endemic product, plays a significant role in the economy of both the region it is cultivated in and the overall country. Therefore, it is crucial to closely monitor and promote this product. The objective of this study was to analyze the spatial distribution of vineyards by utilizing advanced machine learning and deep learning algorithms to classify high-resolution satellite images. A deep learning model based on a 3D Convolutional Neural Network (CNN) was developed for vineyard classification. The proposed model was compared with traditional machine learning algorithms, specifically Support Vector Machine (SVM), Random Forest (RF), and Rotation Forest (ROTF). The accuracy of the classifications was assessed through error matrices, kappa analysis, and McNemar tests. The best overall classification accuracies and kappa values were achieved by the 3D CNN and RF methods, with scores of 86.47% (0.8308) and 70.53% (0.6279) respectively. Notably, when Gabor texture features were incorporated, the accuracy of the RF method increased to 75.94% (0.6364). Nevertheless, the 3D CNN classifier outperformed all others, yielding the highest classification accuracy with an 11% advantage (86.47%). The statistical analysis using McNemar's test confirmed that the χ2 values for all classification outcomes exceeded 3.84 at the 95% confidence interval, indicating a significant enhancement in classification accuracy provided by the 3D CNN classifier. Additionally, the 3D CNN method demonstrated successful classification performance, as evidenced by the minimum-maximum F1-score (0.79-0.97), specificity (0.95-0.99), and accuracy (0.91-0.99) values.

Detecting Multiple Mixed Bacteria Using Dual-Mode Hyperspectral Imaging and Deep Neural Networks

Identifying and analyzing mixed pathogenic bacteria is important for clinical diagnosis and antibiotic therapy of multiple bacterial infection. In this paper, a dual-mode hyperspectral microscopic detection technology with hybrid deep neural networks (DNNs) was proposed for simultaneous quantitative analysis of four kinds of pathogenic bacteria in mixed samples. To acquire both transmission and fluorescence spectra regarding the mixed pathogens, we developed a dual-mode hyperspectral detection system with fine spectral resolution and wide wavelength range, which can also generate spatial images that can be used to calculate the total amount of mixed bacteria. The dual-mode spectra were regarded as mixed proportion characteristics and the input of the neural network for predicting the proportion of each bacterium present in the mixture. To better analyze the dual-mode spectral data, we customized a mixed bacteria measurement network (MB-Net) with hybrid DNNs architectures based on spectral feature fusion. Using the fusion strategy, two DNNs frameworks applied for transmission/fluorescence spectral feature processing were stacked to form the MB-Net that processes these features simultaneously, and the achieved average coefficient of determination (R2) and RMSE of validation set are 0.96 and 0.03, respectively. To the best of our knowledge, it is the first time of simultaneously detecting four types of mixed pathogenic bacteria using spectral detection technology, showing excellent potential in clinical practice.

Advancing the path to in-vivo imaging in freely moving mice via multimode-multicore fiber based holographic endoscopy.

Hair-thin multimode optical fiber-based holographic endoscopes have gained considerable interest in modern neuroscience for their ability to achieve cellular and even subcellular resolution during in-vivo deep brain imaging. However, the application of multimode fibers in freely moving animals presents a persistent challenge as it is difficult to maintain optimal imaging performance while the fiber undergoes deformations. We propose a fiber solution for challenging in-vivo applications with the capability of deep brain high spatial resolution imaging and neuronal activity monitoring in anesthetized as well as awake behaving mice. We used our previously developed multimode-multicore fiber to record fluorescently labeled neurons in anesthetized mice. Our exhibits a cascaded refractive index structure, enabling two distinct regimes of light transport that imitate either a multimode or a multicore fiber. The has been specifically designed for use in the initial phase of an in-vivo experiment, allowing for the navigation of the endoscope's distal end toward the targeted brain structure. The multicore regime enables the transfer of light to and from each individual neuron within the field of view. For chronic experiments in awake behaving mice, it is crucial to allow for disconnecting the fiber and the animal between experiments. Therefore, we provide here an effective solution and establish a protocol for reconnection of two segments of with hexagonally arranged corelets. We successfully utilized the to image neurons in anaesthetized transgenic mice expressing enhanced green fluorescent protein. Additionally, we compared imaging results obtained with the with larger numerical aperture (NA) fibers in fixed whole-brain tissue. This study focuses on addressing challenges and providing insights into the use of multimode-multicore fibers as imaging solutions for in-vivo applications. We suggest that the upcoming version of the increases the overall NA between the two cladding layers to allow for access to high resolution spatial imaging. As the NA increases in the multimode regime, the fiber diameter and ring structure must be reduced to minimize the computational burden and invasiveness.

Extended Kalman filtering for enhancing a cyclic pattern-updating scheme of single-pixel dynamic fluorescence spatial frequency domain imaging: publisher's note.

This publisher's note contains a correction to Opt. Lett.48, 5443 (2023)10.1364/OL.495099.